JP2024088570A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

To provide an image processing apparatus that appropriately performs color mapping on a print color gamut on the basis of a type of content of image data.SOLUTION: An image processing apparatus comprises: input means that inputs image data; generation means that, by using conversion means that converts a color gamut of the image data input by the input means into a color gamut of a device outputting the image data, generates image data in which the color gamut is converted from the image data input by the input means; correction means that corrects the conversion means on the basis of a result of the conversion of the color gamut; and control means that controls execution of correction, by the correction means, of a content included in the image data input by the input means. The control means controls the execution of the correction of the content by the correction means on the basis of a kind of the content.SELECTED DRAWING: Figure 2

Description

The present invention relates to an image processing device, an image processing method, and a program capable of performing color mapping.

There is a known image processing device that receives a digital document described in a specific color space, maps each color in that color space to a color gamut that can be reproduced by a printer, and outputs the result. Patent Document 1 describes "perceptual" mapping and "absolute colorimetric" mapping. Patent Document 2 describes the determination of whether or not to perform color space compression and the direction of compression for an input color image signal.

JP 2020-27948 A Japanese Patent Application Laid-Open No. 07-203234

If color mapping is performed to the print gamut regardless of the type of image data content, the output colors may not be appropriate for the content.

The present invention aims to provide an image processing device, an image processing method, and a program that appropriately performs color mapping to the print gamut based on the type of image data content.

In order to solve the above problem, the image processing device according to the present invention comprises an input means for inputting image data, a generation means for generating image data whose gamut has been converted from the image data input by the input means using a conversion means for converting the gamut of the image data input by the input means into the gamut of a device that outputs the image data, a correction means for correcting the conversion means based on the result of the gamut conversion, and a control means for controlling the execution of correction by the correction means on the content included in the image data input by the input means, the control means controls the execution of correction by the correction means on the content based on the type of the content, and when the conversion means has been corrected by the correction means, the generation means uses the corrected conversion means to generate image data whose gamut has been converted from the image data input by the input means, and the image data whose gamut has been converted by the conversion means after the correction has expanded the color difference in the image data whose gamut has been converted by the conversion means before the correction.

According to the present invention, it is possible to appropriately map colors to the print gamut based on the type of content of the image data.

FIG. 1 is a block diagram showing a configuration of an image processing device. 13 is a flowchart showing image processing. 13 is a flowchart showing an area setting process. FIG. 13 is a diagram for explaining pixel analysis. FIG. 11 is a diagram for explaining determination of a type of content. 13 is a flowchart showing a process of creating a table after color degeneration correction. FIG. 13 is a diagram for explaining color degeneracy. FIG. 2 is a diagram for explaining a manuscript page. FIG. 2 is a diagram showing a UI screen of an image processing application. FIG. 1 is a diagram showing a configuration including an application. 13 is a flowchart showing image processing. 11 is a flowchart showing a process for determining whether or not color degeneration correction is required for each region. FIG. 11 is a diagram for explaining the color degeneration determination process in step S302. FIG. 11 is a diagram for explaining the color degeneration correction process in step S305. FIG. 13 is a diagram showing a brightness correction table. FIG. 2 is a diagram showing the configuration around a recording head. FIG. 13 is a diagram showing a UI screen. FIG. 2 is a diagram for explaining the relationship between contents and areas. FIG. 13 is a diagram showing the relationship between the number of region pixels and a similar pixel number threshold. 13 is a flowchart showing a process for determining the type of content. FIG. 11 is a diagram for explaining differences due to positional relationships of contents. 13 is a flowchart showing a process for determining the type of content. 13 is a flowchart showing a process for determining the type of content. 13 is a flowchart showing a process for determining the type of content. 13 is a flowchart showing a process for determining the type of content.

The following embodiments are described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe multiple features, not all of these multiple features are necessarily essential to the invention, and multiple features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicate explanations are omitted.

[First embodiment]
The terms used in this specification are defined as follows.

(Color reproduction range)
The "color reproduction range" is also called the color reproduction range, color gamut, or gamut. Generally, the "color reproduction range" refers to the range of colors that can be reproduced in an arbitrary color space. In addition, the color gamut volume is an index that indicates the width of the color reproduction range. The color gamut volume is a three-dimensional volume in an arbitrary color space. The chromaticity points that constitute the color reproduction range may be discrete. For example, a specific color reproduction range may be represented by 729 points on the CIE-L*a*b*, and the points between them may be obtained using known interpolation calculations such as tetrahedral interpolation and cubic interpolation. In such a case, the corresponding color gamut volume can be obtained by calculating and accumulating the volumes of tetrahedrons and cubes that constitute the color reproduction range on the CIE-L*a*b* in accordance with the interpolation calculation method. The color reproduction range and color gamut in this embodiment are not limited to a specific color space, but in this embodiment, the color reproduction range in the CIE-L*a*b* space is described as an example. Furthermore, the numerical values of the color reproduction range in this embodiment indicate volumes obtained by cumulative calculation in the CIE-L*a*b* space on the premise of tetrahedral interpolation.

(Gamut mapping)
Gamut mapping is a process of converting between different color gamuts, for example, mapping an input color gamut to an output color gamut of a device such as a printer. Common examples include Perceptual, Saturation, and Colorimetric ICC profiles. The mapping process may be realized by conversion using a three-dimensional lookup table (3DLUT). Also, the mapping process may be performed after color space conversion to a standard color space. For example, if the input color space is sRGB, it is converted to the CIE-L*a*b* color space, and then mapping process is performed to the output color gamut on the CIE-L*a*b* color space. The mapping process may be conversion using a 3DLUT or a conversion formula. Also, conversion of the color space at the time of input and the color space at the time of output may be performed simultaneously. For example, the color space at the time of input may be sRGB, and the color space may be converted to the printer-specific RGB value or CMYK value at the time of output.

(Manuscript data)
The manuscript data refers to the entire input digital data to be processed. The manuscript data includes one to multiple pages. Each single page may be held as image data or expressed as a drawing command. When expressed as a drawing command, rendering may be performed and the data may be converted into image data before processing. The image data is composed of multiple pixels arranged two-dimensionally. The pixels hold information representing colors in a color space. Examples of information representing colors include RGB values, CMYK values, K values, CIE-L*a*b* values, HSV values, and HLS values. Note that this embodiment is applicable to both one page and multiple pages. In this embodiment, as an example, manuscript data of one page will be described as image data.

(Color degeneracy)
In this embodiment, when performing gamut mapping for any two colors, the distance between the colors after mapping in a specified color space becomes smaller than the distance between the colors before mapping, which is defined as color degeneracy. Specifically, assume that there are colors A and B in a digital document, and color A is converted to color C and color B is converted to color D by mapping to the printer's color gamut. In this case, color degeneracy is defined as the distance between colors C and D becoming smaller than the distance between colors A and B. When color degeneracy occurs, colors that were recognized as different in the digital document are recognized as the same color when printed. For example, a graph makes different items recognized as different items by making different items different colors. When color degeneracy occurs, different colors are recognized as the same color, and different items in the graph may be mistakenly recognized as the same item. The specified color space for calculating the distance between colors here may be any color space. For example, these include the sRGB color space, Adobe RGB color space, CIE-L*a*b* color space, CIE-LUV color space, XYZ color space, xyY color space, HSV color space, and HLS color space.

FIG. 1 is a block diagram showing an example of the configuration of an image processing device in this embodiment. For example, a PC, a tablet, a server, or a recording device is used as the image processing device 101. FIG. 1 shows an example in which the image processing device 101 is configured separately from the recording device 108. The CPU 102 executes various image processes by reading a program stored in a storage medium 104 such as an HDD or ROM into a RAM 103 as a work area and executing the program. For example, the CPU 102 acquires a command from a user via a Human Interface Device (HID) I/F (not shown). Then, various image processes are executed according to the acquired command and the program stored in the storage medium 104. The CPU 102 also performs a predetermined process on the manuscript data acquired via the data transfer I/F 106 according to the program stored in the storage medium 104. Then, the CPU 102 displays the result and various information on a display (not shown) and transmits them via the data transfer I/F 106.

The image processing accelerator 105 is hardware capable of executing image processing faster than the CPU 102. The image processing accelerator 105 is started when the CPU 102 writes parameters and data required for image processing to a specified address in the RAM 103. After reading the parameters and data, the image processing accelerator 105 executes image processing on the data. However, the image processing accelerator 105 is not an essential element, and equivalent processing may be executed by the CPU 102. Specifically, the image processing accelerator is a GPU or a specially designed electric circuit. The parameters may be stored in the storage medium 104, or may be obtained from the outside via the data transfer I/F 106.

In the recording device 108, the CPU 111 reads out a program stored in the storage medium 113 into the RAM 112 as a work area and executes it, thereby controlling the recording device 108 as a whole. The image processing accelerator 109 is hardware capable of executing image processing faster than the CPU 111. The image processing accelerator 109 is started by the CPU 111 writing parameters and data required for image processing to a specified address in the RAM 112. After reading the above parameters and data, the image processing accelerator 109 executes image processing on the data. However, the image processing accelerator 109 is not an essential element, and equivalent processing may be executed by the CPU 111. The above parameters may be stored in the storage medium 113, or may be stored in a storage (not shown) such as a flash memory or HDD.

Here, we will explain the image processing performed by the CPU 111 or the image processing accelerator 109. Image processing is, for example, processing that generates data indicating the ink dot formation positions for each scan by the print head 115 based on the acquired print data. The CPU 111 or the image processing accelerator 109 performs color conversion processing and quantization processing on the acquired print data.

The color conversion process is a process of separating colors into ink densities that can be handled by the recording device 108. For example, the acquired recording data includes image data that represents an image. If the image data represents an image in color space coordinates such as sRGB, which is the representation color of the monitor, the data that represents the image in the sRGB color coordinates (R, G, B) is converted into ink data (CMYK) that can be handled by the recording device 108. The color conversion method is realized by matrix calculation processing or processing using a 3DLUT or 4DLUT.

In this embodiment, the recording device 108 uses black (K), cyan (C), magenta (M), and yellow (Y) inks for recording, as an example. Therefore, the image data of the RGB signal is converted into image data consisting of 8-bit color signals for each of K, C, M, and Y. The color signal for each color corresponds to the amount of each ink applied. In addition, although four colors, K, C, M, and Y, are given as an example of the number of ink colors, other ink colors may be used to improve image quality, such as fluorescent ink (F), light cyan (Lc), light magenta (Lm), and gray (Gy) inks with low density. In that case, color signals corresponding to those inks are generated.

After the color conversion process, a quantization process is performed on the ink data. The quantization process is a process that reduces the number of gradation levels of the ink data. In this embodiment, quantization is performed using a dither matrix that arranges threshold values for comparison with the ink data value for each pixel. After the quantization process, binary data is finally generated that indicates whether or not a dot will be formed at each dot formation position.

After image processing, the printhead controller 114 transfers the binary data to the printhead 115. At the same time, the CPU 111 performs printing control via the printhead controller 114 to operate a carriage motor (not shown) that operates the printhead 115, and further to operate a transport motor that transports the printing medium. The printhead 115 scans over the printing medium, and at the same time, the printhead 115 ejects ink droplets onto the printing medium, forming an image.

The image processing device 101 and the recording device 108 are connected via a communication line 107. In this embodiment, a local area network (LAN) is described as an example of the communication line 107, but it may also be a USB hub, a wireless communication network using a wireless access point, a connection using a Wi-Fi direct communication function, etc.

The following description will assume that the recording head 115 has nozzle rows for four color inks: cyan (C), magenta (M), yellow (Y), and black (K).

Figure 16 is a diagram for explaining the recording head 115 in this embodiment. In this embodiment, an image is recorded in a unit area of one nozzle row by multiple scans N times. The recording head 115 has a carriage 116, nozzle rows 115k, 115c, 115m, and 115y, and an optical sensor 118. The carriage 116, which is equipped with five nozzle rows 115k, 115c, 115m, and 115y and the optical sensor 118, can be moved back and forth along the X direction (main scanning direction) in the figure by the driving force of a carriage motor transmitted via a belt 117. As the carriage 116 moves in the X direction relative to the recording medium, ink droplets are ejected from each nozzle of the nozzle row in the direction of gravity (-z direction in the figure) based on the recording data. As a result, an image of 1/N main scans is recorded on the recording medium placed on the platen 119. When one main scan is completed, the recording medium is transported along a transport direction that intersects with the main scan direction (-y direction in the figure) a distance corresponding to the width of 1/Nth of a main scan. Through these operations, an image the width of one nozzle row is recorded over N multiple scans. By alternating between such main scans and transport operations, an image is gradually formed on the recording medium. In this way, control is performed to complete image recording on a specified area.

Image data input by the user contains a variety of content information. Examples include various categories such as text, photographs, graphics such as figures, and gradations as design expressions. In addition, categories that are linked to the processing described below, such as "emphasis on emphasizing color differences" and "emphasis on color continuity", can also be treated as content information. For example, user settings for content such as "emphasis on emphasizing color differences" and "emphasis on color continuity" can be treated as content information. In this case, "emphasis on emphasizing color differences" corresponds to "text or graphics" below, and "emphasis on color continuity" corresponds to "photo or gradation" below.

When mapping image data into a color space that can be represented by a printer, if the colors present in the image data are reduced in color, the user may perceive different colors as the same in the printed result. If the image data contains "text and a background that is a similar color to the text," it is undesirable for the user to be unable to read the text due to color reduction. The same applies to graphics. For such content, it is desirable to carry out "color reduction correction" to reduce the degree of color reduction after mapping.

On the other hand, color degeneration correction places emphasis on maintaining the distance between colors within the limited color space that can be represented by the printer, so the color continuity between close colors is reduced compared to before color degeneration correction. The "photos" in image data contain many areas where the color changes continuously and minutely, such as human skin and the sky, but color degeneration correction tries to maintain the distance between colors even at the expense of shifting the saturation and brightness, changing the characteristic of color continuity. This results in output results that are undesirable for the user. The same applies to gradations. It is not advisable to perform "color degeneration correction" after mapping on such content.

In addition, image data input by a user may contain a mixture of various contents. Uniformly applying color reduction correction to image data, or not applying color reduction correction at all, may result in an output result that is undesirable for some contents. Therefore, in this embodiment, the content in the image data is analyzed in advance to determine whether the content places importance on maintaining color difference. With such a configuration, color reduction correction can be applied to content for which it is desirable to apply color reduction correction, and it is possible to prevent an output result that is undesirable for some contents.

Figure 2 is a flowchart showing image processing by the image processing device 101 in this embodiment. In this embodiment, the process in Figure 2 increases the distance between colors in a specified color space for color combinations that cause color degeneracy. As a result, the degree of color degeneracy can be reduced. The process in Figure 2 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. The process in Figure 2 may also be executed by the image processing accelerator 105.

In S101, the CPU 102 inputs image data. For example, the CPU 102 inputs image data stored in the storage medium 104. The CPU 102 may also input image data via the data transfer I/F 106. The CPU 102 acquires image data including color information from the input original data (acquisition of color information). The image data includes values that represent colors expressed in a predetermined color space. In acquiring color information, values that represent colors are acquired. The values that represent colors are values acquired from, for example, sRGB data, Adobe RGB data, CIE-L*a*b* data, CIE-LUV data, XYZ color system data, xyY color system data, HSV data, and HLS data.

At S102, the CPU 102 obtains a gamut mapping table that is pre-recorded on the storage medium 104.

In S103, the CPU 102 performs gamut mapping on the color information of each pixel of the image data using the gamut mapping table acquired in S102. The image data after gamut mapping is stored in the RAM 103 or the storage medium 104. Specifically, the gamut mapping table is a three-dimensional lookup table. The three-dimensional lookup table can calculate a combination of output pixel values (Rout, Gout, Bout) for a combination of input pixel values (Rin, Gin, Bin). When the input values Rin, Gin, and Bin each have 256 gradations, it is preferable to use table Table1 [256] [256] [256] [3], which has a total of 16,777,216 sets of output values of 256 x 256 x 256. The CPU 102 performs color conversion using the gamut mapping table. Specifically, this is achieved by carrying out the following process for each pixel of an image composed of RGB pixel values of the manuscript data input in S101.

Rout = Table1 [Rin] [Gin] [Bin] [0] ... (1)
Gout = Table1 [Rin] [Gin] [Bin] [1] ... (2)
Bout = Table1 [Rin] [Gin] [Bin] [2] ... (3)
Also, the number of grids in the LUT may be reduced from 256 grids to, for example, 16 grids, and the table size may be reduced by determining the output value by interpolating table values of a plurality of grids, etc. Also, in addition to a three-dimensional LUT, a 3×3 matrix calculation formula may be used.

In S104, the CPU 102 sets an area in the manuscript data input in S101. There are various possible methods for setting the area, but as an example, a method is used in which the white pixel value of the input image is used as the judgment criterion.

Figure 3 is a flowchart showing the area setting process using the white pixel value as the judgment criterion. The information "seg" in Figure 3 indicates each set area, and each area is managed by information such as 0, 1, 2, etc. as a segment number.

In S201, the CPU 102 acquires one pixel from the image data. In this flowchart, pixel analysis is performed with this pixel as the pixel of interest.

The rectangular information in Figure 4(a) represents image data. Black pixels in the image data are considered to be non-white areas, and white pixels are considered to be white areas. The pixel of interest is changed by scanning the image data in the direction of the arrow. For the image data in Figure 4(a), first the top pixel row is scanned from left to right, then the middle pixel row is scanned from left to right, and finally the bottom pixel row is scanned from left to right.

In S202, it is determined whether the pixel of interest is a white pixel. A white pixel is, for example, a pixel where R=G=B=255 in 8-bit information. In this process, the determination of whether a pixel is white or not may be performed after gamut mapping in S103, or it may be performed on the pixel value before gamut mapping. However, when making a determination from a pixel after gamut mapping, a table is used that ensures that a pixel where R=G=B=255 retains its value even after gamut mapping, in other words, that it remains a white pixel. If it is determined to be white, proceed to S203, and if it is determined not to be white, transition to S204.

In S203, the CPU 102 sets a white segment to the pixel of interest. In this embodiment, the segment is expressed as a number, and the white segment is treated as segment number 0.

In S204, the CPU 102 provisionally sets a seg number for the pixel of interest. The pixel is inspected according to the pixel scanning direction shown in FIG. 4A, and the seg number is provisionally set based on the seg number of the pixel to the left of the pixel of interest. In this embodiment, as an example, the provisional setting of the seg number is performed according to the following rules.

-When the pixel to the left has a segment number of 0, assign a new segment number to the pixel of interest.

When the pixel to the left has seg number=N, the pixel of interest is assigned seg number=N.

In S205, the CPU 102 judges whether or not a provisional seg number has been assigned to all pixels of the image data. If it is determined that a provisional seg number has been assigned to all pixels, the process proceeds to S206, and if it is determined that a provisional seg number has not been assigned to all pixels, the process repeats from S201.

Figure 4 (b) shows the result of provisionally setting the segment numbers. The above setting shows that provisional segment numbers 0 to 3 have been assigned within a 3x3 range.

In S206, the CPU 102 corrects the provisional segment number. The pixel of interest is corrected in the same scanning direction as in FIG. 4A. The CPU 102 compares the segment number of the pixel of interest with the smallest segment number excluding segment number 0 among the four pixels above, below, left, and right, and replaces the segment number of the pixel of interest with the smaller segment number.

In S207, the CPU 102 determines whether or not correction of the provisional segment numbers has been performed for all pixels of the image data. If it is determined that correction of the provisional segment numbers has been performed for all pixels, the process of FIG. 3 ends. If it is determined that correction of the provisional segment numbers has not been performed for all pixels, the process from S206 is repeated.

Figure 4 (c) shows the result of the correction of the provisional seg number. The provisional seg number = 3 that existed in Figure 4 (b) is corrected to provisional seg number = 2. The pixels with provisional seg numbers 0, 1, and 2 remain with the same numbers.

The above area setting is one example. Area setting may be based on pixels other than white pixels in the pixel information, or the determination of the segment number may be achieved using a different flow. For example, a reduction process may be performed in advance, and all overlapping pixels may be determined to have the same segment number as a result of the reduction. By using such a reduction process, the processing time for area setting can be shortened. Also, information other than the R, G, and B information of pixels may be used.

In S105, the CPU 102 analyzes the area set in S104. This analysis process is a process for generating information for determining whether to create and apply a color difference correction table in the subsequent stage. In this embodiment, the CPU 102 generates information for each area as the determination information, indicating whether the area is "text or graphic" or "photograph or gradation."

In this embodiment, by analyzing the content contained in the image data in this way, it becomes possible to perform control such that color reduction correction is performed on "text or graphics" but not on "photos or gradations." In other words, a configuration is adopted in which color reduction correction is not performed uniformly on image data, but only on content for which it is appropriate to perform color reduction correction. With this configuration, it is possible to reduce image damage caused by uniformly performing color reduction correction.

An example of the analysis process is shown in Fig. 5(a) to Fig. 5(c). The image data shown in Fig. 5(a) shows a state in which the character "I" is set as one area 400 as a result of the area setting in S104. On the other hand, the image data shown in Fig. 5(b) shows a state in which a circular gradation is set as one area 401 as a result of the area setting in S104. Areas 400 and 401 are set as rectangles that surround the character and circular gradation, but if the method described as an example in S104 is used, the result will be that they are separated by a boundary with white pixels. For the sake of explanation, a rectangular area is set based on the Y coordinates of the pixel located at the top and the pixel located at the bottom of the area, and the X coordinates of the pixel located at the rightmost and the pixel located at the left of the area, based on the result of the area setting in S104.

In S105, the CPU 102 uses edge detection areas 402 and 403 having a predetermined area to determine whether an edge exists within the mesh frame area. There are various methods for determining whether an edge exists, but here, as an example, a method using "same pixel", "similar pixel", and "different pixel" is used. As shown in FIG. 5(c), pixels included in the pixel value range THa to THb are classified as same pixels, pixels included in the range THc to THd are classified as similar pixels, and other pixels are classified as different pixels. Here, the value to be compared with TH may be RGB information or YCbCr information, or information of any one channel (CH), and there is no limit to the number and type of CH. The values of THa to THd may be changed for each color information, and for example, different threshold values may be set for all RGB. By comparing all pixels in the edge detection areas 402 and 403 with the central pixel, it is possible to know how many same pixels, similar pixels, and different pixels are included in the edge detection areas 402 and 403. Edge detection areas 402 and 403 scan the entire regions 400 and 401, and a histogram is created that accumulates the number of identical pixels, similar pixels, and different pixels. Because the central pixel of edge detection area 402 needs to scan the entire region 400, the range where the edge of detection area 402 is outside region 400 can also be scanned. Similarly, the edge of edge detection area 403 can also be scanned in a range outside region 401.

Histogram 404 corresponds to FIG. 5(a), and histogram 405 corresponds to FIG. 5(b). Therefore, it can be seen that "i" in FIG. 5(a) has many identical pixels and different pixels, but few similar pixels. It can also be seen that the circular gradation in FIG. 5(b) has a comparatively large number of similar pixels.

The frequency distribution resulting from the histogram makes it possible to classify an area as either "text or graphics" or "photography or gradation." Since it can be determined that an area is "text or graphics" when the number of similar pixels is low, classification is performed by setting thresholds for same pixels, similar pixels, and different pixels. For example, the CPU 102 determines that an area is "text or graphics" based on whether it meets the conditions in equation (4).

Number of same pixels＞TH_same && Number of similar pixels＜TH_near && Number of different pixels＞TH_other ... (4)
This analysis process is an example, and the method is not limited to the above. For example, the number of pixels resulting from the judgment result may be converted into the area ratio of the region and then compared using formula (4). In addition, in this embodiment, edge detection using the edge detection areas 402 and 403 is used for judgment, but judgment may also be based on color distribution within the region. Here, a configuration may be adopted in which a threshold value is appropriately set. Such a configuration will be described in the sixth embodiment and onwards.

In S106, the CPU 102 performs a process of determining whether or not color degeneration correction is necessary for each area. For this determination, the information on "text or graphics" or "photograph or gradation" determined in S105 is used. In this embodiment, if the area is a "text or graphics" area, it is determined that color degeneration correction is necessary, and the process transitions to S107. On the other hand, if the area is a "photograph or gradation" area, it is determined that color degeneration correction is not necessary, and the process transitions to S109.

In S107, the CPU 102 creates a table after color degeneration correction using the image data input in S101, the gamut mapping table acquired in S102, and the image data after gamut mapping performed in S103. The table after color degeneration correction is created for each area divided in S104. The table after color degeneration correction has the same format as the gamut mapping table.

In S108, the CPU 102 generates area data after color degeneration correction by performing calculations on the area data of the corresponding area of the image data input in S101 using the table after color degeneration correction created in S107. The generated area data after color degeneration correction is stored in the RAM 103 or the storage medium 104. In S109, the CPU 102 stores the results of the gamut mapping in S103 for the area data in the RAM 103 or the storage medium 104.

For area data determined to be "photo or gradation", the gamut mapping table may be switched to one for "photo or gradation". By applying a table that emphasizes color differences to "text or graphics" and a table that emphasizes color continuity to "photo or gradation", color conversion can be performed that makes the most of the characteristics of each content. Also, to maintain color continuity, the gamut mapping table may not be applied to "photo or gradation" areas. By not applying the gamut mapping table to some content, the effect of reducing memory capacity can be achieved. Similarly, the table length of the gamut mapping table may be changed depending on the content, or a simple configuration such as a unique conversion coefficient instead of a table may be replaced, with the same effect of reducing memory capacity.

In S110, the CPU 102 determines whether or not the entire area of the image data has been processed. If it is determined that the entire area has been processed, the process proceeds to S111, and if it is determined that the entire area has not been processed, the process repeats from S105.

In S111, the CPU 102 outputs the area data after color degeneration correction stored in S108, or the result after gamut mapping stored in S109, from the image processing device 101 via the data transfer I/F 106. The gamut mapping may be mapping from the sRGB color space to the color reproduction gamut of the recording device 108. In this case, it is possible to suppress the reduction in saturation and color difference due to gamut mapping into the color reproduction gamut of the recording device 108.

The creation of the table after color degeneration correction in S107 will be described in detail with reference to FIG. 6. The process in FIG. 6 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. The process in FIG. 6 may also be executed by the image processing accelerator 105.

In S301, the CPU 102 detects unique colors present in each region. In this embodiment, the term "unique colors" refers to colors used in the image data. For example, for text data with black characters on a white background, the unique colors are white and black. Also, for an image such as a photograph, the unique colors are the colors used in the photograph. The CPU 102 stores the detection results as a unique color list in the RAM 103 or the storage medium 104. The unique color list is initialized at the start of S201. The CPU 102 repeats the detection process for each pixel of the image data, and determines whether the color of each pixel is different from the unique colors detected so far for all pixels included in the target object. If it is determined that the color is unique, the color is stored as a unique color in the unique color list.

As a method of judgment, it may be possible to judge whether the color of the target pixel is included in the unique color list created up to that point, and if it is judged that the color is not included, to add new color information to the unique color list. In this way, it is possible to detect the unique color list included in the area. For example, if the input image data is sRGB data, each has 256 gradations, so a total of 16,777,216 unique colors of 256 x 256 x 256 are detected. In this case, the number of colors becomes enormous, and the processing speed decreases. Therefore, it is possible to detect unique colors discretely. For example, it is possible to reduce the 256 gradations to 16 gradations and then detect unique colors. When reducing the number of colors, it is possible to reduce the number of colors to the color of the nearest grid. In this way, it is possible to detect a total of 4,096 unique colors of 16 x 16 x 16, and the processing speed can be improved.

ＣＰＵ１０２は、色差ΔＥ６０８が色差ΔＥ６０７に比べ、小さい場合、色縮退していると判定する。さらに、ＣＰＵ１０２は、色差ΔＥ６０８が色差を識別できるだけの大きさが無かった場合、色縮退していると判定する。これは、色６０５と色６０６が、人の視覚特性に基づき、異なる色だと識別できる程度の色差があれば、色差を補正する必要が無いからである。視覚特性上、異なる色だと識別できる色差ΔＥとしては、２．０が用いられても良い。つまり、色差ΔＥ６０８が色差ΔＥ６０７に比べ小さい場合で且つ色差ΔＥ６０８が２．０より小さい場合に色縮退していると判定するようにしても良い。 In S302, the CPU 102 detects the number of color combinations that are color degenerate among the combinations of unique colors present in the region based on the unique color list detected in S301. FIG. 7 is a diagram for explaining color degeneration. A color gamut 601 is the color gamut of the input image data. A color gamut 602 is the color gamut after gamut mapping in S103. In other words, the color gamut 602 corresponds to the color gamut of the device. A color 603 and a color 604 are included in the input image data. A color 605 is the color when gamut mapping is performed on the color 603. A color 606 is the color when gamut mapping is performed on the color 604. If a color difference 608 between the color 605 and the color 606 is smaller than a color difference 607 between the color 603 and the color 604, it is determined that color degeneration occurs. The CPU 102 repeats this determination process for the number of color combinations in the unique color list. A Euclidean distance in a color space is used as a method for calculating the color difference. In this embodiment, the Euclidean distance (hereinafter, referred to as color difference ΔE) in the CIE-L*a*b* color space is used as a suitable example for explanation. Since the CIE-L*a*b* color space is a visually uniform color space, the Euclidean distance can be approximated as the amount of color change. Therefore, when the Euclidean distance in the CIE-L*a*b* color space is small, humans perceive the colors as being closer to each other, and when the Euclidean distance is large, humans perceive the colors as being farther apart. If the input data is not CIE-L*a*b* data, color space conversion is performed. For the color space conversion at that time, a known process may be used. The color information in the CIE-L*a*b* color space is represented by a color space with three axes, L*, a*, and b*. For example, color 603 is represented by L603, a603, and b603. Color 604 is represented by L604, a604, and b604. Color 605 is represented by L605, a605, and b605. The color 606 is represented by L 606, a 606, and b 606. If the input image data is represented in another color space, it is converted to the CIE-L*a*b* color space using a known method. The color difference ΔE 607 and the color difference ΔE 608 are calculated by the following equations (5) and (6).

The CPU 102 determines that color degeneration has occurred when the color difference ΔE 608 is smaller than the color difference ΔE 607. Furthermore, the CPU 102 determines that color degeneration has occurred when the color difference ΔE 608 is not large enough to distinguish the color difference. This is because if the color 605 and the color 606 have a color difference that allows them to be distinguished as different colors based on human visual characteristics, there is no need to correct the color difference. A color difference ΔE of 2.0 may be used as a color difference that allows the colors to be distinguished as different colors based on visual characteristics. In other words, it may be determined that color degeneration has occurred when the color difference ΔE 608 is smaller than the color difference ΔE 607 and is smaller than 2.0.

In S303, the CPU 102 determines whether the number of color combinations determined to be color degenerate in S302 is zero. If the number of color combinations determined to be color degenerate is zero, i.e., it is determined that there is a color difference, the process proceeds to S304, where the CPU 102 determines that this is an area in which color degeneracy correction is not required, terminates the processing of FIG. 5, and proceeds to S108. On the other hand, if the number of color combinations determined to be color degenerate in S303 is not zero, i.e., it is determined that there is no color difference, the process proceeds to S305, where color degeneracy correction (color difference correction) is performed.

Because color degeneracy correction results in color changes, it also changes the colors of color combinations that are not color degenerate, resulting in unnecessary color changes. For this reason, a ratio based on the total number of unique color combinations and the number of color combinations that are color degenerate may be used as a condition for determining the need for color degeneracy correction. Specifically, it may be determined that color degeneracy correction is necessary when the number of color combinations that are color degenerate is more than half of the total number of unique color combinations. In this way, color changes caused by excessive color degeneracy correction can be suppressed.

In S305, the CPU 102 performs color reduction correction for the color combinations that are subject to color reduction based on the input image data, the image data after gamut mapping, and the gamut mapping table.

The color degeneracy correction will be explained in detail with reference to FIG. 7. Color 603 and color 604 are input colors included in the input image data. Color 605 is the color after color 603 is color converted by gamut mapping. Color 606 is the color after color 604 is color converted by gamut mapping. In FIG. 7, the combination of colors 603 and 604 indicates that the colors are degenerated. Therefore, color degeneracy can be corrected by separating colors 605 and 606 from each other by the color distance in a predetermined color space. Specifically, a correction process is performed to increase the color distance between colors 605 and 606 beyond the color distance at which colors 605 and 606 can be identified as different colors based on human visual characteristics. The color difference ΔE is 2.0 or more for the color distance at which colors can be identified as different colors based on visual characteristics. More preferably, the color difference between colors 605 and 606 is about the same as color difference ΔE607. The CPU 102 repeats the color degeneration correction process for the number of color combinations that will be color degenerated. The results of color degeneration correction for the number of color combinations are stored in a table, with color information before and after correction. In FIG. 7, the color information is color information in the CIE-L*a*b* color space. Therefore, the input image data and the color space of the image data at the time of output may be converted. In that case, the color information before correction in the color space of the input image data and the color information after correction in the color space of the output image data are stored in a table.

Next, the color degeneration correction process will be described in detail. A color difference correction amount 609 that widens the color difference ΔE is obtained from the color difference ΔE 608. The difference between the color difference ΔE of 2.0, which is a color difference that can be recognized as a different color in terms of visual characteristics, and the color difference ΔE 608 is the color difference correction amount 609. More preferably, the difference between the color difference ΔE 607 and the color difference ΔE 608 is the color difference correction amount 609. The result of correcting the color 605 by the color difference correction amount 609 on the extension line in the CIE-L*a*b* color space from the color 606 to the color 605 is the color 610. The color 610 is separated from the color 606 by the color difference ΔE 608 and the color difference correction amount 609. In the above, it was on the extension line from the color 606 to the color 605, but this is not limited to this in this embodiment. As long as the color difference ΔE between the color 606 and the color 610 is a color difference separation that is a combination of the color difference ΔE 608 and the color difference correction amount 609, the color difference ΔE may be in any direction of the lightness direction, the saturation direction, or the hue angle direction in the CIE-L*a*b* color space. Also, the color difference ΔE may be in one direction or a combination of the lightness direction, the saturation direction, and the hue angle direction. Furthermore, in the above, the color degeneration is corrected by changing the color 605, but the color 606 may also be changed. Also, both the color 605 and the color 606 may be changed. When changing the color 606, since it cannot be changed to the outside of the color gamut 602, the color 606 is changed by moving it on the boundary surface of the color gamut 602. In that case, the color degeneration correction may be performed by changing the color 605 for the insufficient color difference ΔE.

In S306, the CPU 102 changes the gamut mapping table using the result of the color degeneration correction in S305. The gamut mapping table before the change is a table that converts the input color 603 to the output color 605. As a result of S305, the table is changed to one that converts the output color of the input color 603 to color 610. In this way, a table after color degeneration correction can be created. The CPU 102 repeats the change of the gamut mapping table for the number of color combinations that are color degenerated.

As described above, by applying the gamut mapping table after color degeneration correction to the region to which the input image data is applied, a correction can be performed to increase the distance between colors for color combinations that are color degenerate among the unique color combinations that the region has. As a result, color degeneration can be efficiently reduced for color combinations that are color degenerate. For example, when the input image data is sRGB data, the gamut mapping table is created on the premise that the input image data has 16,777,216 colors. The gamut mapping table created under this premise is created taking into account color degeneration and saturation even for colors that are not actually included in the input image data. In this embodiment, the gamut mapping table can be adaptively corrected for the region data by detecting the colors that the region data has. Then, a gamut mapping table can be created that targets the colors that the region data has. As a result, suitable adaptive gamut mapping can be performed for the region data, so that color degeneration can be efficiently reduced.

In this embodiment, regions were set for the input image data, and each of the set regions was analyzed. Then, using the analysis results, it was determined whether or not to perform color degeneracy correction, and if it was determined that color degeneracy correction should be performed, color degeneracy correction was performed. The purpose of performing color degeneracy correction is to reduce color degeneracy due to gamut mapping. Information such as characters and graphics is important not in color but in the content (the meaning of the characters and the shape of the graphics). Therefore, if the distance between colors becomes small and the user cannot recognize the shape of the characters or graphics after printing, this leads to a decrease in the visibility and readability of the content. Therefore, color degeneracy correction is performed in the case of characters and graphics. On the other hand, information such as photographs and gradations often has importance in the color itself. For example, in a portrait, just because the color distance between the person's skin color and the color of the background wall in the color space is close, it is not desirable for the user to correct the skin to a different color. The gradation has a design meaning in itself because it has continuous gradations, but correcting the color distance in the color space of the continuous gradation to create a step in the gradation impairs the design and is not desirable for the user. The purpose of the region analysis and color degeneration correction judgment process in this embodiment is to determine the user's intention (the meaning of the data) from the characteristics of the image data entered by the user. According to this embodiment, it is possible to apply appropriate gamut mapping for each region depending on the judgment result of the color degeneration correction.

In this embodiment, the processing has been described for the case where the input image data is one page. The input image data may be multiple pages. When the input image data is multiple pages, the processing flow in FIG. 2 may be performed across all pages, or the processing in FIG. 2 may be performed for each page. In this way, even when the input image data is multiple pages, the degree of color degeneracy due to gamut mapping can be reduced.

In addition, in this embodiment, multiple regions are set from the input image data, but the decision on whether to perform color degeneration correction may be made for the entire image. In this case, instead of setting multiple regions, the decision on whether to perform "text or graphics" or "photograph or gradation" is made for the entire image.

In this embodiment, the two types of classification are "text or graphics" and "photograph or gradation", but the classification is not limited to this. For example, gray lines (pixel values on the gray axis from white to black), black points, and white points in the input image must not deviate from the gray line through color degeneracy correction, so color degeneracy correction may not be performed on them. For example, a printer may eject only K ink on the input gray line portion, and deviation from the gray line may change the control of ink usage. Therefore, the results of analyzing the gray line information may be used to determine whether or not to perform color degeneracy correction.

In this embodiment, the gamut mapping table after the color degeneration correction is applied to the input image, but a correction table for performing color degeneration correction on the image data after gamut mapping may be created. In that case, a correction table for converting the color information before correction to the color information after correction may be generated based on the color degeneration correction result in S305. The generated correction table is a table for converting the color 605 to the color 610 in FIG. 7. In S108, the CPU 102 applies the generated correction table to the image data after gamut mapping. In this way, by correcting the image data after gamut mapping, the degree of color degeneration due to gamut mapping can be reduced. In addition, after applying the gamut mapping table reflecting the color degeneration correction to each region of the input image, a unique gamut mapping table may be applied to the entire image to set the output gamut. For example, overall color adjustment, etc. may be performed after the processing of this embodiment. In that case, the evaluation of the color distance in the color degeneration correction and the amount and direction of color degeneration correction take into account that the gamut mapping table is applied after the color degeneration correction.

In this embodiment, the user may be able to input an instruction as to whether or not to perform color degeneracy correction processing. In that case, the UI screen shown in FIG. 17 may be displayed on a display unit (not shown) mounted on the image processing device 101 or the recording device 108 to accept user instructions. In the UI screen shown in FIG. 17, the user can select the type of color correction using a toggle button. Furthermore, it is possible to select ON/OFF using a toggle button whether or not to perform "adaptive gamut mapping", which indicates the processing described in this embodiment. With this configuration, it is possible to switch whether or not to perform adaptive gamut mapping according to a user instruction. As a result, when the user wants to reduce the degree of color degeneracy, the gamut mapping described in this embodiment can be performed.

[Second embodiment]
The second embodiment will be described below with respect to the differences from the first embodiment. In the first embodiment, the color degeneration correction is performed on a single color in S305. Therefore, depending on the combination of colors in the input image data, although the degree of color degeneration is reduced, a change in color tone may occur. Specifically, when color degeneration correction is performed on two colors with different hue angles, if the color is changed by changing the hue angle, the color tone will differ from that of the color in the input image data. For example, if color degeneration correction is performed on blue and purple by changing the hue angle, the purple will change to red. If the color tone changes, the user may be reminded of a device problem such as poor ink ejection.

In the first embodiment, the color degeneracy correction is repeated for the number of unique color combinations in the input image data. This ensures that the distance between colors can be increased. However, when the number of unique colors in the input image data increases, changing a color to increase the distance between colors may result in the distance between the colors being smaller than that of another unique color after the change. Therefore, the CPU 102 must repeatedly perform the color degeneracy correction of S305 so that the expected distance between colors is achieved for all unique color combinations in the input image data. This increases the processing time because the amount of processing required to increase the distance between colors is enormous.

Therefore, in this embodiment, for each specified hue angle, multiple unique colors are treated as one color group and color degeneracy correction is performed in the same direction. In order to correct multiple unique colors as one color group, in this embodiment, a reference unique color (described later) is selected from the color group. Also, by limiting the correction direction to the lightness direction, it is possible to suppress changes in color tone. Also, by correcting multiple unique colors as one color group in the lightness direction, it is no longer necessary to process all color combinations contained in the input image data, and processing time can be reduced.

Figure 13 is a diagram for explaining the color degeneracy determination process of S302 in this embodiment. Figure 13 is a diagram showing two axes, the a* axis and the b* axis, in a plane in the CIE-L*a*b* color space. The hue range 1201 represents a range in which multiple unique colors within a specified hue angle are grouped into one color group. In Figure 13, the hue angle of 360 degrees is divided into six equal parts, so the hue range 1201 represents a range from 0 degrees to 60 degrees. The hue range is preferably a hue range that can be recognized as the same color. For example, the hue angle in the CIE-L*A*B* color space is determined in 60-degree increments from 30 degrees. When the hue angle is determined in 60-degree increments, six colors, red, green, blue, cyan, magenta, and yellow, can be separated. When the hue angle is determined in 30-degree increments, the colors can be further separated by colors between the colors separated in 60-degree increments. The hue range may be determined fixedly as in Figure 13. It may also be dynamically determined based on the unique colors contained in the input image data.

As in the first embodiment, the CPU 102 detects the number of color combinations that are color degenerate within the hue range 1201 for the unique color combinations in the input image data. In FIG. 13, colors 1204, 1205, 1206, and 1207 represent input colors. In this case, the CPU 102 determines whether or not the four color combinations of colors 1204, 1205, 1206, and 1207 are color degenerate. The CPU 102 then repeats this process for all hue ranges. In this way, the number of color combinations that are color degenerate for each hue range is detected.

In FIG. 13, for example, six color combinations are detected. In this embodiment, the hue range is determined for every 60-degree hue angle, but this is not limited to the above. For example, the hue range may be determined for every 30-degree hue angle, or the hue range may be determined without dividing equally. Preferably, the range of hue angles is determined as the hue range so as to be visually uniform. With such a configuration, colors within the same color group are visually perceived as the same color, so that color degeneracy correction can be performed on the same color. Furthermore, for each hue range, the number of color combinations that are color degenerate in a hue range including two adjacent hue ranges may be detected.

Figure 14 is a diagram for explaining the color degeneration correction process of S305 in this embodiment. Figure 14 is a diagram showing two axes, the L* axis and the C* axis in the CIE-L*a*b* color space, on a plane. L* represents lightness, and C* represents saturation. In Figure 14, color 1301, color 1302, color 1303, and color 1304 are input colors. Color 1301, color 1302, color 1303, and color 1304 represent colors included in the range of the hue range 1201 in Figure 13. Color 1305 is the color after color 1301 is converted by gamut mapping. Color 1306 is the color after color 1302 is converted by gamut mapping. Color 1307 is the color after color 1303 is converted by gamut mapping. Color 1304 represents the color after color conversion by gamut mapping is the same color.

First, the CPU 102 determines a unique color (reference color) that serves as the basis for the color degeneracy correction process for each hue range. In this embodiment, the maximum lightness color, the minimum lightness color, and the maximum saturation color are determined as the reference colors. In FIG. 14, color 1301 is the maximum lightness color, color 1302 is the minimum lightness color, and color 1303 is the maximum saturation color.

Next, the CPU 102 calculates a correction factor R for each hue range from the number of unique color combinations and the number of degenerate color combinations in the target hue range. A suitable calculation formula is shown below.

Correction rate R=number of degenerate color combinations/number of unique color combinations (7)
The correction factor R is smaller as the number of color combinations that are color degenerate is smaller, and is larger as the number of color combinations that are color degenerate is larger. In this way, the more the number of color combinations that are color degenerate is, the stronger the color degeneration correction can be applied. FIG. 14 shows that there are four colors within the range of the hue range 1201 in FIG. 13. Therefore, the number of unique color combinations is six. For example, the number of color combinations that are color degenerate among the six combinations is four. In this case, the correction factor is 0.667. FIG. 14 shows an example in which all combinations are color degenerated by gamut mapping. However, even after color conversion by gamut mapping, if the color difference is greater than the minimum distinguishable color difference, the color combination is not considered to be color degenerate. Therefore, the combinations of color 1304 and color 1303 and color 1304 and color 1302 are not considered to be color degenerate. Here, the minimum distinguishable color difference ΔE is, for example, 2.0.

このように、ガマットマッピング後に保持すべき色差ΔＥを算出する。ガマットマッピング後に保持すべき色差ΔＥは、ガマットマッピング前の色差ΔＥである。図１４では、補正量Ｍｈは色差１３０８に補正率Ｒを乗算した値であり、補正量Ｍｌは色差ΔＥ１３０９に補正率Ｒを乗算した値である。また、ガマットマッピング前の色差ΔＥが識別可能な最小色差より大きい場合、保持すべき色差ΔＥは、識別可能な最小色差ΔＥより大きければよい。このように処理を行うことで、ガマットマッピングによって低下した色差ΔＥを、識別可能な色差ΔＥまで回復することができる。さらに、保持すべき色差ΔＥは、ガマットマッピング前の色差ΔＥであっても良い。この場合は、ガマットマッピング前の識別しやすさに近づけることができる。さらに、保持すべき色差ΔＥは、ガマットマッピング前の色差より大きくても良い。この場合は、ガマットマッピング前より識別しやすくすることができる。 Next, the CPU 102 calculates a correction amount for each hue range from the correction rate R and the color information of the maximum lightness color, the minimum lightness color, and the maximum saturation color. The CPU 102 calculates a correction amount Mh on the side brighter than the maximum saturation color and a correction amount Ml on the side darker than the maximum saturation color as the correction amount. As in the first embodiment, color information in the CIE-L*a*b* color space is represented by a color space with three axes of L*, a*, and b*. The color 1301, which is the maximum lightness color, is represented as L1301, a1301, and b1301. The color 1302, which is the minimum lightness color, is represented as L1302, a1302, and b1302. The color 1303, which is the maximum saturation color, is represented as L1303, a1303, and b1303. The suitable correction amount Mh is a value obtained by multiplying the color difference ΔE between the maximum lightness color and the maximum chroma color by a correction rate R. The suitable correction amount Ml is a value obtained by multiplying the color difference ΔE between the maximum chroma color and the minimum lightness color by a correction rate R. The correction amounts Mh and Ml are calculated by the following formulas (8) and (9).

In this way, the color difference ΔE to be retained after gamut mapping is calculated. The color difference ΔE to be retained after gamut mapping is the color difference ΔE before gamut mapping. In FIG. 14, the correction amount Mh is a value obtained by multiplying the color difference 1308 by the correction rate R, and the correction amount Ml is a value obtained by multiplying the color difference ΔE 1309 by the correction rate R. Furthermore, if the color difference ΔE before gamut mapping is larger than the minimum distinguishable color difference, the color difference ΔE to be retained may be larger than the minimum distinguishable color difference ΔE. By performing the process in this way, the color difference ΔE reduced by gamut mapping can be restored to the distinguishable color difference ΔE. Furthermore, the color difference ΔE to be retained may be the color difference ΔE before gamut mapping. In this case, it is possible to approach the ease of distinction before gamut mapping. Furthermore, the color difference ΔE to be retained may be larger than the color difference before gamut mapping. In this case, it is possible to make it easier to distinguish than before gamut mapping.

Next, the CPU 102 generates a brightness correction table for each hue range. The brightness correction table is a table for expanding the brightness between colors in the brightness direction based on the brightness of the maximum saturation color, the correction amount Mh, and the correction amount Ml. In FIG. 14, the brightness of the maximum saturation color is the brightness L1303 of color 1303. The correction amount Mh is a value based on the color difference ΔE1308 and the correction rate R. The correction amount Ml is a value based on the color difference ΔE1309 and the correction rate R. A method for creating a brightness correction table for expanding the brightness in the brightness direction is described below.

The brightness correction table is a 1DLUT. In a 1DLUT, the input brightness is the brightness before correction, and the output brightness is the brightness after correction. The brightness after correction is determined by characteristics based on three points: the minimum brightness after correction, the brightness of the maximum saturation color after gamut mapping, and the maximum brightness after correction. The maximum brightness after correction is the brightness obtained by adding the correction amount Mh to the brightness of the maximum saturation color after gamut mapping. The minimum brightness after correction is the brightness obtained by subtracting the correction amount Ml from the brightness of the maximum saturation color after gamut mapping. In the brightness correction table, the characteristics are specified as linearly changing between the minimum brightness after correction and the brightness of the maximum saturation color after gamut mapping. Also, the characteristics are specified as linearly changing between the brightness of the maximum saturation color after gamut mapping and the maximum brightness after correction. In FIG. 14, the maximum brightness before correction is the brightness L1305 of the color 1305, which is the maximum brightness color. The minimum brightness before correction is the brightness L1306 of color 1306, which is the minimum brightness color. The brightness of the maximum saturation color after gamut mapping is the brightness L1307 of color 1307. The maximum brightness after correction is brightness L1310, which is the brightness L1307 plus color difference ΔE1308, which is the correction amount Mh. In other words, it can be said that the color difference between the maximum brightness color and the maximum saturation color is converted into a brightness difference. The minimum brightness after correction is brightness L1311, which is the brightness L1307 minus color difference 1309, which is the correction amount Ml. In other words, it can be said that the color difference between the minimum brightness color and the maximum saturation color is converted into a brightness difference.

Figure 15 is a diagram showing an example of a brightness correction table for expanding brightness in the brightness direction in Figure 14. In this embodiment, color degeneration correction is performed by converting the color difference ΔE to a brightness difference. In terms of visual characteristics, brightness difference is highly sensitive. Therefore, by converting the saturation difference to a brightness difference, even a small brightness difference in terms of visual characteristics can be made to feel as if a color difference ΔE has been added. In addition, due to the relationship between the sRGB color gamut and the color gamut of the recording device 108, the brightness difference is smaller than the saturation difference. Therefore, by converting to a brightness difference, it is possible to effectively utilize a narrow color gamut. In addition, in this embodiment, the brightness of the maximum saturation color is not changed. In this embodiment, since the brightness of the most saturated color is not changed, the color difference ΔE can be corrected while maintaining the brightness of the maximum saturation color. The correction of values greater than the maximum brightness and values less than the minimum brightness may be indefinite because they are not included in the input image data. In addition, the brightness correction table may be supplemented, and in that case, the values may be supplemented so as to change linearly as shown in Figure 15. In this way, it is possible to reduce the number of grids in the brightness correction table, thereby reducing capacity and reducing the processing time required to transfer the brightness correction table.

In addition, if the maximum brightness after correction exceeds the maximum brightness of the color gamut after gamut mapping, the CPU 102 performs maximum value clipping processing. The maximum value clipping processing is a process of subtracting the difference between the maximum brightness after correction and the maximum brightness of the color gamut after gamut mapping in the entire brightness correction table. In other words, the entire brightness correction table is shifted in the direction of lower brightness until the maximum brightness of the color gamut after gamut mapping becomes the maximum brightness after correction. In this case, the brightness of the maximum saturation color after gamut mapping is also moved to the lower brightness side. In this way, if the unique color of the input image data is biased toward the high brightness side, the color difference ΔE can be improved and color degeneration can be reduced by using the brightness gradation range on the low brightness side. On the other hand, if the minimum brightness after correction is lower than the minimum brightness of the color gamut after gamut mapping, the CPU 102 performs minimum value clipping processing. The minimum value clipping processing adds the difference between the minimum brightness after correction and the minimum brightness of the color gamut after gamut mapping in the entire brightness correction table. In other words, the entire brightness correction table is shifted toward higher brightness until the minimum brightness of the color gamut after gamut mapping becomes the minimum brightness after correction. In this way, if the unique colors in the input image data are biased toward the low brightness side, the color difference ΔE can be improved and color degeneration can be reduced by using the brightness gradation range on the high brightness side.

Next, the CPU 102 applies the brightness correction table created for each hue range to the gamut mapping table. First, the CPU 102 determines which hue angle brightness correction table to apply from the color information of the output value of the gamut mapping. For example, if the hue angle of the output value of the gamut mapping is 25 degrees, it determines that the brightness correction table of the hue range 1201 in FIG. 13 is to be applied. Then, the CPU 102 applies the determined brightness correction table to the output value of the gamut mapping table to correct it. The CPU 102 sets the corrected color information as the output value after the new gamut mapping. For example, in FIG. 14, the CPU 102 applies the determined brightness correction table to the color 1305, which is the output value of the gamut mapping table, and corrects the brightness of the color 1305. Then, the CPU 102 sets the brightness of the corrected color 1312 as the output value after the new gamut mapping.

As described above, in this embodiment, the brightness correction table created using the reference color is also applied to colors other than the reference color within the hue range 1201. Then, for the color after brightness correction, for example color 1312, mapping to the color gamut 1316 is performed without changing the hue, as described below. In other words, in the hue range 1201, the direction of color degeneracy correction is limited to the brightness direction. With such a configuration, it is possible to suppress changes in color tone. Furthermore, there is no need to perform color degeneracy correction processing for all unique color combinations in the input image data, and processing time can be reduced.

Furthermore, the brightness correction tables of adjacent hue ranges may be synthesized according to the hue angle of the output value of the gamut mapping. For example, when the hue angle of the output value of the gamut mapping is Hn degrees, the brightness correction table of the hue range 1201 and the brightness correction table of the hue range 1202 are synthesized. Specifically, the brightness value of the output value after gamut mapping is corrected with the brightness correction table of the hue range 1201 to obtain the brightness value Lc1201. Also, the brightness value of the output value after gamut mapping is corrected with the brightness correction table of the hue range 1202 to obtain the brightness value Lc1202. At that time, the angle of the intermediate hue angle of the hue range 1201 is the hue angle H1201, and the angle of the intermediate hue angle of the hue range 1202 is the hue angle H1202. In this case, the corrected brightness value Lc is calculated by complementing the corrected brightness value Lc1201 and the corrected brightness value Lc1202. The brightness value Lc after interpolation is calculated using the following formula (10):

このように、色相角によって、適用する明度補正テーブルを合成することにより、色相角の変化での補正強度の急峻な変化を抑制することができる。

In this way, by synthesizing the brightness correction table to be applied depending on the hue angle, it is possible to suppress abrupt changes in correction strength due to changes in the hue angle.

If the color space of the corrected color information is different from the color space of the output value after gamut mapping, the color space is converted to the output value after gamut mapping. For example, if the color space of the corrected color information is the CIE-L*a*b* color space, the following search is performed to obtain the output value after gamut mapping.

ΔＨ=ΔＥ-(ΔＬ＋ΔＣ) ・・・（１４）
ΔＥｗ=Ｗｌ×ΔＬ＋Ｗｃ×ΔＣ＋Ｗｈ×ΔＨ ・・・（１５）
明度方向に色差ΔＥを変換して拡張したため、明度を彩度より重視してマッピングする。つまり、明度の重みＷｌは、彩度の重みＷｃより大きい。さらに、色相は色味に対する影響が大きいため、色相は、明度及び彩度より重視してマッピングする方が補正前後で色味の変化を最小限に抑えることができる。つまり、色相の重みＷｈは、明度の重みＷｌ以上であり、且つ、彩度の重みＷｃより大きい。このように、本実施形態では、色味を維持しつつ色差ΔＥを補正することができる。 If the value after the lightness correction exceeds the color gamut after gamut mapping, mapping to the color gamut after gamut mapping is performed. For example, color 1312 in FIG. 14 exceeds the color gamut 1316 after gamut mapping. In that case, mapping from color 1312 to color 1314 is performed. The mapping method used here is color difference minimum mapping that emphasizes lightness and hue. In color difference minimum mapping that emphasizes lightness and hue, the color difference ΔE is calculated by the following formula. In the CIE-L*a*b* color space, the color information of the color that exceeds the color gamut after gamut mapping is set to Ls, as, and bs. The color information of the color within the color gamut after gamut mapping is set to Lt, at, and bt. The lightness difference is set to ΔL, the saturation difference is set to ΔC, and the hue difference is set to ΔH. The weight of lightness is set to Wl, the weight of saturation is set to Wc, the weight of hue angle is set to Wh, and the weighted color difference is set to ΔEw.

ΔH=ΔE−(ΔL+ΔC) (14)
ΔEw=Wl×ΔL+Wc×ΔC+Wh×ΔH (15)
Since the color difference ΔE is converted and expanded in the lightness direction, lightness is given more importance than saturation when mapping. That is, the lightness weight Wl is greater than the saturation weight Wc. Furthermore, since the hue has a large effect on color, the change in color before and after correction can be minimized by mapping the hue with more importance than lightness and saturation. That is, the hue weight Wh is equal to or greater than the lightness weight Wl and greater than the saturation weight Wc. In this way, in this embodiment, the color difference ΔE can be corrected while maintaining the color.

Furthermore, the color space may be converted when performing color difference minimum mapping. It is known that in the CIE-L*a*b* color space, color changes in the saturation direction do not result in equal hues. Therefore, if the change in hue angle is suppressed by increasing the weight of the hue, the colors will not be mapped to equal hues. For this reason, it is possible to convert to a color space in which the hue angle is bent so that color changes in the saturation direction result in equal hues. In this way, by performing weighted color difference minimum mapping, changes in color can be suppressed.

In FIG. 14, color 1305, which is the result of gamut mapping of color 1301, is corrected to color 1312 by the brightness correction table. Since color 1312 exceeds color gamut 1316 after gamut mapping, color 1312 is mapped to color gamut 1316. In other words, color 1312 is mapped to color 1314. As a result, in this embodiment, when the input of the corrected gamut mapping table is color 1301, the output is color 1314.

In this embodiment, the brightness correction table is created for each hue range. However, a brightness correction table may be created for adjacent hue ranges. Specifically, the number of combinations of degenerated colors is detected in the hue range 1201 and 1202 in FIG. 13. Next, the number of combinations of degenerated colors is detected in the hue range 1202 and 1203. In other words, by performing detection with one hue range overlapping, it is possible to suppress a sharp change in the number of combinations of degenerated colors when crossing hue ranges. In this case, a suitable hue range is preferably a hue angle range in which the two hue ranges can be combined and recognized as the same color. For example, the hue angle in the CIE-L*a*b* color space is 30 degrees. In other words, one hue angle range is 15 degrees. In this way, it is possible to suppress a sharp change in the correction strength of color degeneration across hue ranges.

In this embodiment, the color difference ΔE is corrected in the lightness direction by treating multiple unique colors as one group. It is known that the sensitivity to lightness difference varies depending on saturation as a visual characteristic, and lightness difference of low saturation is more sensitive than lightness difference of high saturation. Therefore, the correction amount in the lightness direction may be controlled by the saturation value. In other words, at low saturation, the correction amount in the lightness direction is controlled to be small, and at high saturation, correction is performed with the correction amount in the lightness direction described above. Specifically, when correcting lightness using the lightness correction table, the lightness value Ln before correction and the lightness value Lc after correction are divided internally by the saturation correction rate S. The saturation correction rate S is calculated by the following formula based on the saturation value Sn of the output value after gamut mapping and the maximum saturation value Sm of the color gamut after gamut mapping at the hue angle of the output value after gamut mapping.

S = Sn / Sm (16)
Lc'=S×Lc+(1−S)×Ln (17)
That is, the closer to the maximum saturation value Sm of the color gamut after gamut mapping, the closer the saturation correction rate S is to 1, and Lc' is closer to the corrected lightness value Lc obtained by the lightness correction table. On the other hand, the lower the saturation value Sn of the output value after gamut mapping, the closer the saturation correction rate S is to 0, and Lc' is closer to the lightness value Ln before correction. In other words, the lower the saturation value Sn of the output value after gamut mapping, the smaller the lightness correction amount. Also, the correction amount may be zero in a color gamut with low saturation. With this configuration, it is possible to suppress color changes near the gray axis. Also, since color degeneration correction can be performed according to visual sensitivity, overcorrection can be suppressed.

[Third embodiment]
The third embodiment will be described below with a focus on differences from the first and second embodiments. In this embodiment, an example will be described in which content analysis is performed using information other than pixel values of input image data.

In this embodiment, the determination is made using a drawing command, which will be described later, as information different from the pixel values of the image data. The drawing command includes descriptive information such as "photo" or "text" that is added in the application software when the user creates the image data. In this way, by using the drawing command, it is possible to apply color degeneration correction only to content for which it is appropriate to perform the correction.

The area setting process in S104 will be explained using information other than pixel values.

Figure 8 is a diagram for explaining an example of a page of image data (hereinafter referred to as manuscript data) input in S101 of Figure 2 in this embodiment. Here, it is assumed that the document data is described in PDL. PDL is an abbreviation for Page Description Language, and is composed of a set of drawing commands on a page-by-page basis. The types of drawing commands are defined for each PDL specification, but in this embodiment, the following three types are used as an example.

Command 1) TEXT drawing command (X1, Y1, color, font information, character string information)
Command 2) BOX drawing command (X1, Y1, X2, Y2, color, fill shape)
Command 3) IMAGE drawing command (X1, Y1, X2, Y2, image file information)
Other drawing commands may be used as appropriate depending on the application, such as a DOT drawing command for drawing a point, a LINE drawing command for drawing a line, a CIRCLE drawing command for drawing an arc, etc. For example, a general PDL such as Portable Document Format (PDF) proposed by Adobe, XPS proposed by Microsoft, or HP-GL/2 proposed by HP may be used.

The manuscript page 700 in FIG. 8 represents one page of manuscript data, and as an example, the number of pixels is 600 pixels wide and 800 pixels high. Below, an example of PDL corresponding to the document data of the manuscript page 700 in FIG. 8 is shown.

<PAGE=001>
<TEXT>50,50,550,100,BLACK,STD-18,“ABCDEFGHIJKLMNOPQR”</TEXT>
<TEXT>50,100,550,150,BLACK,STD-18,“abcdefghijklmnopqrstuv”</TEXT>
<TEXT>50,150,550,200,BLACK,STD-18,“1234567890123456789”</TEXT>
<BOX>50, 350, 200, 550, GRAY, STRIPE</BOX>
<IMAGE>250,300,580,700,“PORTRAIT.jpg”</IMAGE>
</PAGE>
<PAGE=001> on the first line is a tag that indicates the page number in this embodiment. Normally, PDL is designed to be able to describe multiple pages, so tags indicating page breaks are described in the PDL. In this example, </PAGE> indicates that this is the first page. In this embodiment, this corresponds to manuscript page 700 in FIG. 8. If there is a second page, <PAGE=002> will be described following the above PDL.

The text from <TEXT> on the second line to </TEXT> on the third line is drawing command 1, which corresponds to the first line of area 701 in Figure 8. The first two coordinates indicate the coordinates (X1, Y1) of the upper left corner of the drawing area, and the next two coordinates indicate the coordinates (X2, Y2) of the lower right corner of the drawing area. Next, it is written that the color is BLACK (R=0, G=0, B=0), the font is "STD" (standard), the character size is 18 points, and the string to be written is "ABCDEFGHIJKLMNOPQR".

The text from <TEXT> on the fourth line to </TEXT> on the fifth line is drawing command 2, which corresponds to the second line of area 701 in FIG. 8. The first four coordinates and two character strings indicate the drawing area, character color, and character font, respectively, just like command 1, and state that the character string to be written is "abcdefghijklmnopqrstuv".

Drawing command 3 begins with <TEXT> on the sixth line and ends with </TEXT> on the seventh line, and corresponds to the third line of area 701 in FIG. 8. The first four coordinates and two character strings indicate the drawing area, character color, and character font, just like drawing command 1 and drawing command 2, and state that the character string to be written is "1234567890123456789".

The section from <BOX> to </BOX> on the 8th line is drawing command 4, which corresponds to area 702 in Figure 8. The first two coordinates indicate the upper left coordinates (X1, Y1) which are the drawing start point, and the next two coordinates indicate the lower right coordinates (X2, Y2) which are the drawing end point. Next, the color is specified as GRAY (gray: R = 128, G = 128, B = 128), and the fill shape is specified as STRIPE, which is a striped pattern. In this embodiment, the direction of the stripes is a line to the lower right, but the angle and period of the lines may also be specified in the BOX command.

Next, the IMAGE command on lines 9 and 10 corresponds to area 703 in Figure 8. Here, it is written that the file name of the image in that area is "PORTRAIT.jpg", which indicates that it is a JPEG file, a commonly used image compression format. Then, </PAGE> on line 11 indicates that drawing of that page has finished.

In actual PDL files, in addition to the above group of drawing commands, there are cases where the "STD" font data and the "PORTRAIT.jpg" image file are also included as one file. This is because if the font data and image files are managed separately, the text and image parts cannot be formed with drawing commands alone, and there is insufficient information to form the image in Figure 8. Also, area 704 in Figure 8 is an area where there are no drawing commands, and is blank.

In the case of a manuscript page described in PDL, such as manuscript page 700 in FIG. 8, the area setting process in S104 in FIG. 2 can be realized by analyzing the above PDL. Specifically, the start and end points of the drawing Y coordinates of each drawing command are as follows, and are continuous in area.

Drawing command Y start point Y end point First TEXT command 50 100
Second TEXT command 100 150
Third TEXT command 150 200
BOX command 350 550
IMAGE command 300 700
It can also be seen that the BOX command and the IMAGE command are both spaced 100 pixels from the TEXT command in the Y direction.

Next, for the BOX command and the IMAGE command, the start and end points of the drawing X coordinates are as follows, and we can see that they are 50 pixels apart in the X direction.

Drawing command X start point X end point BOX command 50 200
IMAGE command 250 580
From the above, three regions can be set as follows:

Area X start point Y start point X end point Y end point First area 50 50 550 200
Second area 50 350 200 550
Third area 250 300 580 700
In S105, the CPU 102 determines that the content of the first region is "text or graphics." In S105, the CPU 102 determines that the content of the second region is "text or graphics." In S105, the CPU 102 determines that the content of the third region is "photograph or gradation."

As described above, according to this embodiment, content is analyzed using drawing commands as information different from the pixel values of the input image data. By using drawing commands, different areas can be set between overlapping content that is difficult to judge from the input image data alone. In-store POP images in retail stores may display price information partially overlapping a product photo, but even in such cases, it is possible to perform color degeneration correction on the price information text and not on the product photo. In addition, information in which characters are lined up side by side with spaces between them, such as the character string in area 701, can also be set as the same area. Areas can also be set even if the background is not white but a solid color.

[Fourth embodiment]
The fourth embodiment will be described below with respect to the differences from the first to third embodiments. In this embodiment, a configuration will be described in which it is determined whether or not to perform color degeneration correction based on information other than pixel information and PDL description information.

In this embodiment, an example of an application that allows a user to assign information on whether or not to perform color degeneration correction to pixels of an input image is described. In the first embodiment, it was described that emphasis should be placed on visibility and readability of graphics. However, even within graphics, there is information that gives meaning to colors. For example, the colors used in a company logo are called corporate colors, and represent colors that impress the image of the company. Therefore, when a company logo is superimposed on a background graphic, it may not be desirable for the colors contained in the logo to change for the sake of visibility and readability. In this embodiment, in cases where there is an area within the graphic where color degeneration correction should not be performed, whether or not to perform color degeneration correction is determined from information different from pixel information and PDL description information.

In this embodiment, the application can receive a setting operation from the user as to whether or not to perform color reduction correction on a specified pixel. This configuration can reflect the user's intention regarding the implementation of color reduction correction.

Figure 9(a) shows the UI screen of an image processing application that can be operated by the user. The display screen 801 of the entire application includes an image display section 802 for the image being created by the application, a color specification palette section 803, and a fill palette 804. The corporate color is set by defining a color in the color specification palette section 803. Four colors are defined in the color specification palette section 803, but the user can add colors to be specified as corporate colors.

When the "+" area is pressed, a color specification window 808 as shown in FIG. 9(b) is displayed. The color specification window 808 has a gradation specification section 809 that allows a color to be specified within a gradation, and an absolute value specification section 810 that allows a color to be specified using RGB values or Lab values. FIG. 9(b) shows, as an example, a state in which a color of R=G=B=128 has been specified as a color. When the OK button 811 is pressed for the determined color, the color specified by the user is added to the color specification palette section 803. When the cancel button 812 is pressed, the color specification palette is not updated. The color specified in the color specification palette section 803 is set as a color for which color degeneration correction is not performed during image processing.

When the fill palette 804 is pressed, the color specification window 808 in FIG. 9(b) is also displayed. When the OK button 811 is pressed, the color specified by the user is added to the fill palette 804. The user can specify two colors on the UI using the fill palette 804.

The user can paint the image display section 802 by selecting a specified color with the pointer and sliding it over the image display section 802 to overlap it. The user can use a number of methods using the area selection section 805 to determine which part of the image displayed in the image display section 802 to apply the color to. For example, the user can paint the entire overlapping object area by sliding an arrow. In addition, the user can specify a specified area in the image display section 802 so that it is displayed with a solid or dotted frame, and then paint within that area. Figure 9(a) shows an example in which the text section "ABC" and the heart in the graphic section are filled with a color slid from the color specification palette section 803.

The applications in Figs. 9(a) and 9(b) are just examples, and the UI that the user uses is not limited to these. For example, a UI that specifies an RIP (Raster Image Processor) may be used. Also, a UI that specifies which area of image data drawn in RGB colors should have a specified color may be used.

Figure 10 is a diagram showing an example of the relationship between the application in this embodiment and the configuration of the image processing device 101 and the recording device 108. The user terminal 901 includes an editing application 902, which is an application in this embodiment, and the image processing device 101. Note that the image processing device 101 may be included in the user terminal 901, or the user terminal 901 may be the image processing device 101. In other words, the user terminal 901 may be the image processing device 101, and the editing application 902 may be installed in the image processing device 101. Also, in Figure 10, the image processing device 101 may be configured inside the recording device 108.

In addition to the RGB information of the image data, the editing application 902 outputs α plane information for the area specified in the color specification palette section 803. The α plane information may be handled as binary information of 0 and 1, or as multi-value information. In other words, the α plane information is area information for which the user has specified that color degeneracy correction is not to be performed. The RGB information and α plane information are input to the image processing device 101, and gamut mapping is performed using this information. The RGB information resulting from gamut mapping is input to the recording device 108 and converted into quantized data for printing. The α plane information may be used in the ink decomposition process, gamma conversion process, and quantization process performed inside the recording device 108, as shown by the dotted lines.

Figure 11 is a flowchart showing the gamut mapping process in this embodiment. The process in Figure 11 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. The process in Figure 2 may also be executed by the image processing accelerator 105.

In S401, the CPU 102 inputs the input image data and the corresponding α plane data. The α plane data has On/Off information for determining whether or not color degeneration correction is to be performed for each pixel. In this embodiment, α=1 (On) indicates that color degeneration correction is not to be performed, and α=0 (Off) indicates that color degeneration correction is to be performed.

S102 and S103 in Figure 11 are the same as those in Figure 2, so their explanation will be omitted.

In S402, area setting is performed in the same manner as in the first embodiment, but in addition, each area is set so that the alpha plane data is composed of only On or only Off.

S105 in Figure 11 is the same as the explanation for S105 in Figure 2, so the explanation will be omitted.

In S403, the CPU 102 determines whether color degeneration correction is required for each area.

Figure 12 is a flowchart showing the process of S403. In this embodiment, the process of Figure 12 determines whether or not color degeneration correction is required for each region.

In S501, the CPU 102 determines whether the alpha plane information assigned to the segment number N is alpha = 1 (On). If the determination result is alpha = 1 (On), the process proceeds to S503, and if the determination result is alpha = 0 (Off), the process proceeds to S502.

In S502, the CPU 102 determines whether the area with segment number=N is "text or graphics." If it is determined that it is "text or graphics," the process proceeds to S504, and if it is determined that it is not "text or graphics" (i.e., it is "photograph or gradation"), the process proceeds to S503.

In S503, the CPU 102 determines that color degeneracy correction is not to be performed for segment number=N, and stores the result of the determination in a storage area such as the RAM 103. In S504, the CPU 102 determines that color degeneracy correction is not to be performed for segment number=N, and stores the result of the determination in a storage area such as the RAM 103.

When α = 1 (On), in other words, the user has specified that color reduction correction should not be performed in the area, and in that case, it is determined that color reduction correction should not be performed without judging the content. On the other hand, when α = 0 (Off), the need to perform color reduction correction is determined based on the content judgment result.

As described above, in this embodiment, a user is asked whether or not to perform color degeneracy correction on a pixel specified on the application. According to this embodiment, by using the information specified by the user on the application, appropriate gamut mapping can be performed for each region in accordance with the user's intention for creating image data.

In this embodiment, the color specified on the application is not limited to the corporate color. For example, in train route maps displayed on ticket vending machines, each route is assigned an image color, so such an image color may be used. In addition, stamp information containing words such as "confidential" or "no copying" is often set to the warning color red, and since it is desirable for this color to remain constant from document to document, the color may be used in such stamp information. In addition, the color may be used in the legend information of a graph, which is desired to be consistent between pages. According to this embodiment, it is possible to control so that color degeneracy correction is not performed for such colors for which it is undesirable to change the color from the standpoint of visibility and readability.

[Fifth embodiment]
The fifth embodiment will be described below with respect to the differences from the first to fourth embodiments. In this embodiment, when the content information of each region is determined by the likelihood, the correction strength of the color degeneration correction is controlled according to the likelihood.

In this embodiment, we consider a case where the probability that the content analysis result is "photo or gradation" and the probability that it is "text or graphics" cannot be uniquely determined. In such a case, if the configuration is such that the higher probability is determined, the result of the color degeneration correction may be undesirable for the content, such as losing color continuity. Therefore, in this embodiment, the correction strength of the color degeneration correction is controlled according to the likelihood of the content information for each area. With such a configuration, it is possible to reduce the possibility that the result of the color degeneration correction may be undesirable for the content, such as losing color continuity.

Here, as an example, assume that the likelihood of a certain region set in S104 in FIG. 2 being "text or graphics" is 80% and the likelihood of it being "photograph or gradation" is 20%. In this embodiment, the likelihood is set, for example, based on the number of similar pixels in histograms 404 and 405. In the first embodiment, the number of similar pixels in histograms 404 and 405 was compared with a threshold value. In this embodiment, the ratio of the number of similar pixels to the total number of pixels in region 400 or 401 is treated as the likelihood. For example, if the number of similar pixels is 20% of the total number of pixels in region 400, the likelihood of it being "text or graphics" is 80%, and the likelihood of it being "photograph or gradation" is 20%. In this embodiment, the likelihood is used to control the correction strength of the color degeneracy correction. As in the above example, if there is an 80% probability that it is "text or graphics", the result of color degeneracy correction (pixel value Kout) and the state before color degeneracy correction (pixel value Kin) are used to calculate the pixel value of the gamut mapping as shown in the following formula (18).

K_out = 0.8 × K_in + 0.2 × K_out ... (18)
That is, in the above example, since the likelihood of being a "character or graphic" is 80%, the correction strength of the color degeneration correction is weakened from 100% to 80%. As a result, it is possible to reduce the possibility that the color degeneration correction will be excessive.

In the determination of S106 in FIG. 2, if the likelihood of being a "character or graphic" is greater than 0%, it is determined that color degeneracy correction is necessary. In that case, in S305 in FIG. 6, the CPU 102 controls the correction strength of the color degeneracy correction as described above.

As described above, according to this embodiment, the correction strength can be controlled based on the likelihood of the content resulting from the analysis of the area. With this configuration, gamut mapping that reflects the likelihood can be performed even for areas where the analysis result is difficult to determine as either "text or graphics" or "photograph or gradation."

The correction strength of color degeneracy correction, in other words, is the degree to which color continuity is maintained. In this embodiment, for content that is deemed prone to erroneous judgment, it is possible to achieve control that maintains a certain degree of color continuity or ensures a certain degree of distance between colors.

Sixth Embodiment
The sixth embodiment will be described below with respect to the differences from the first to fifth embodiments. In this embodiment, it is assumed that the result of the region setting in S104 is not one content per region. FIG. 18A shows an example in which there is no white pixel between graphic content 1800 and photo content 1801, and they overlap. In FIG. 18A, since the two contents overlap, the result of the region setting in S104 is region 1802, and two contents are included in one region. FIG. 18B shows an example in which graphic content 1803 and photo content 1804 are adjacent to each other. This region setting result is region 1805, and similarly to FIG. 18A, two contents are included in one region.

First, for such an area, it is assumed that the judgment criterion in S105 is to use a value THratio determined by the ratio between the number of area pixels and the number of similar pixels within the area as a threshold value THnear. In the judgment in S105, the target pixel is inspected based on the edge detection area. When the result of comparing the target pixel with its surroundings shows that a pixel with a large proportion of similar pixels whose pixel difference is within a specified range and a small proportion of same pixels/different pixels is determined to be a similar pixel, the target pixel is judged to be a "similar pixel." An edge detection filter is scanned within the area, centered on the target pixel, to obtain the number of pixels judged to be similar within the area (= number of similar pixels).

Figure 19 (a) shows the relationship between the number of region pixels, THratio, and the judgment threshold THnear. For example, if the similar pixel rate in a region exceeds 50%, and it is judged to be a "photograph or gradation," THratio increases monotonically according to the number of region pixels. For example, if the number of region pixels is 100, the threshold THratio is 50, and if the number of similar pixels is greater than 50, it is judged to be a "photograph or gradation."

Here, the region including multiple types of content as described above has a lower similar pixel ratio than a region in which the entire region is photo content. Therefore, such a region may be determined as "text or graphics". For example, the number of region pixels of the graphic content 1803 in FIG. 18B is 200, the number of similar pixels is 0, and the number of region pixels of the photo content 1804 is 100, and the number of similar pixels is 80. Also, if the similar pixel ratio in the region exceeds 50%, it is determined as "photo or gradation". At this time, if the region is only the photo content 1804, the TH ratio is 50, so the determination result is "photo or gradation". On the other hand, the number of region pixels of the region 1805 is 300, so the TH ratio is 150. The number of similar pixels of the region 1805 is 80, so it is determined as "text or graphics". Here, it is possible to make the ratio smaller and the determination result "photo or gradation". However, depending on the size of the adjacent graphic content, the desired judgment results may not be obtained.

For example, consider the case where THratio is calculated assuming that the similar pixel rate in the region is 20%. In this case, the THratio, which is the judgment threshold for region 1805, is 60, and it is judged as "photo or gradation". On the other hand, FIG. 18C shows an example in which the same photo content as in FIG. 18B is adjacent to graphic content 1806, which has a larger number of pixels. When the number of region pixels of graphic content 1806 in FIG. 18C is 800 and the number of similar pixels is 0, the judgment threshold THratio for region 1808 is 180. However, since the number of similar pixels of region 1808 is 80, it is judged as "text or graphics". In other words, although the purpose is to make the judgment result "photo or gradation" by setting THratio to a smaller value, the judgment result of the type of content differs depending on the size of the adjacent graphic content. As a result, color degeneration correction is performed based on the judgment results of different types for a region that includes content to be subjected to color degeneration correction and content not to be subjected to color degeneration correction in one region, which is a processing unit.

For example, in the case of FIG. 18(a), when color degeneration correction is performed on the photo content 1801, the color distance increases, but this may cause the color continuity that existed in the photo content 1801 to be lost. For example, ID photos and the like are printed in a size that allows the person's information to be visually recognized, so it is thought that color changes in skin and the like are easily noticeable by visual inspection. In this way, the change in the characteristics of the color continuity of the photo content results in undesirable results for the user. In a situation where multiple contents exist within one area as shown in FIG. 18(a) to FIG. 18(c), the same thing can be said when color correction is made to differ greatly depending on the type of content. For example, color correction that emphasizes saturation is performed on the graphic content, and contrast enhancement processing is performed on the photo content. In this case, if the content that should be determined as "photo or gradation" is determined as "text or graphics," saturation enhancement processing is performed on the photo area, and for example, the redness of the skin in a face photo will give an impression different from the redness remembered by the user.

Therefore, in this embodiment, a value THabs determined independently of the number of pixels in the region is used as the threshold value THnear. Regions that contain both photo content and graphic content have a lower similar pixel rate compared to regions that contain only photo content, but the number of similar pixels is the same. In this embodiment, by determining the type of content based only on the number of similar pixels, independent of the number of pixels in the region, it is possible to reduce the effect of a decrease in the similar pixel rate due to graphic content, for example, in regions that contain both photo content and graphic content.

In this embodiment, in S105, the number of similar pixels present in the region is obtained. Here, a unique value THabs that is independent of the total number of pixels in the region is set as the threshold for determining that the region is a "photograph or gradation."

Fig. 19(b) is a diagram showing the relationship between the number of pixels in a region, THabs, and THnear, which is the judgment threshold. Also, Fig. 20 is a flowchart showing the content type judgment process in this embodiment. For example, the process in Fig. 20 is realized by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. Also, the process in Fig. 20 is executed in S105 in Fig. 2. First, in S2000, the CPU 102 obtains the number of similar pixels in the region. Next, in S2001, the CPU 102 sets THabs to THnear, which is the judgment threshold.

In S2002, the CPU 102 compares the number of similar pixels obtained in S2000 with THnear. If it is determined that the condition of number of similar pixels > THnear is met, then in S2003 the CPU 102 determines the type of content to be "photo or gradation." On the other hand, if it is determined that the above condition is not met, then in S2004 the CPU 102 determines the type of content to be "text or graphics."

The value of THabs may be determined according to the number of pixels of the content to be determined as "photograph or gradation." In FIG. 19(b), T19 is shown as a unique value that is independent of the total number of pixels in the region. In this case, a region with a number of similar pixels greater than T19 is determined as "photograph or gradation" regardless of the number of pixels in the region.

In this way, in this embodiment, a fixed threshold value is set that is independent of the number of pixels in the area. This makes it possible to determine that an area is a "photo or gradation" when there is photo content of a certain size or more in an area that contains both photo content and graphic content within a single area, which is a processing unit. This maintains the characteristic of color continuity without performing color degeneration correction on an area that contains both photo content and graphic content, for example. Furthermore, in addition to color degeneration correction, it is also possible to prevent the user from feeling uncomfortable when performing different color corrections depending on the type of content, as described above.

The method of determining the type of content is not limited to the above, and other methods may be used. For example, if the image data is in a PDL such as PDF and has an attribute value added for each area, the attribute value may be used to determine the type of content. For example, text attributes or vector attributes may be determined to be a graphic type, and bitmap attributes may be determined to be a photo type. By using attribute values in this way, it is no longer necessary to scan the image data for determination processing, and processing can be performed faster.

[Seventh embodiment]
The seventh embodiment will be described below with respect to the differences from the first to sixth embodiments. In the sixth embodiment, the threshold value THnear is set to a value THabs that is independent of the number of pixels in the region. In this case, all regions with a small number of similar pixels are determined to be "text or graphics."

Here, we assume that depending on the content arrangement, the user's focus may differ even for content of the same size. Figure 21(a) is an example in which graphic content 2101 and photo content 2102 are adjacent to each other. On the other hand, Figure 21(b) is an example in which only photo content 2104, which is exactly the same as photo content 2102, is arranged. When a user looks at the image in Figure 21(a), the graphic content attracts more attention than the photo content. On the other hand, when there is only photo content, such as photo content 2104 in Figure 21(b), the user is likely to focus on the contents of the photo.

Here, for example, photo content smaller than the A4 size for passport photos is determined to be "text or graphics." When photo content 2102 and 2104 are small and have a small number of similar pixels and are therefore determined to be "text or graphics," in FIG. 21(a), the effect of ensuring the color distance by performing color degeneracy correction on graphic content 2101 is large. However, in FIG. 21(b), since only photo content 2104 exists, performing color degeneracy correction changes the characteristics of the color continuity, causing a sense of discomfort to the user.

In this embodiment, in addition to the value THabs, which is independent of the number of pixels in the region, a value THratio, which is dependent on the number of pixels in the region, is used. Regions that contain only photo content have a higher similar pixel rate than regions that contain both photo content and graphic content. Therefore, by using the similar pixel rate as the judgment threshold, it becomes possible to judge a region that contains only photo content and does not have a similar pixel count equal to or greater than the judgment threshold as "photo or gradation."

In this embodiment, in S105 of FIG. 2, in addition to the number of similar pixels present in the region, the number of region pixels is obtained. Then, as the threshold for determining "photograph or gradation", both a value that depends on the total number of pixels in the region (region pixel count) and a value that does not depend on it are set.

Fig. 19(c) is a diagram showing the relationship between the number of area pixels, THratio, THabs, and the judgment threshold THnear. Also, Fig. 22 is a flowchart showing the content type judgment process in this embodiment. The process in Fig. 22 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. Also, the process in Fig. 22 is executed in S105 in Fig. 2.

In this embodiment, for example, 50% of the number of region pixels is set as THratio, and THabs is set as 100. If the number of similar pixels in the region exceeds either of these, it is determined to be a "photograph or gradation." In this example, when the number of region pixels is 200, THratio and THabs match. If the number of region pixels is greater than 201, THabs is used as the criterion for determining that the region is a "photograph or gradation." The number of region pixels where the values of THratio and THabs match is the inflection point IF of the threshold. In FIG. 19(c), the inflection point IF is shown as (P19, T19), which is (200, 100).

In the following explanation, it is assumed that the photo contents 2102 and 2104 have a similar pixel count of 50 and a region pixel count of 80. It is also assumed that the graphic content 2101 has a similar pixel count of 0 and a region pixel count of 500. It is also assumed that 50% of the region pixel count is set as THratio and THabs is set to 100.

First, in S2200, the CPU 102 obtains the number of region pixels. Region 2103 has a region pixel count of 580, so THratio is 290. In S2201, the CPU 102 obtains the number of similar pixels in the region. Then, in S2202, the CPU 102 determines whether or not the condition THabs>THratio is satisfied. If it is determined that the condition is satisfied, then in S2203, the CPU 102 sets THratio to THnear. On the other hand, if it is determined that the condition is not satisfied, then in S2204, the CPU 102 sets THabs to THnear.

In this example, THabs is 100 and THratio is 290, so it is determined that the condition is not met, and in S2204, THabs = 100 is set to THnear.

In S2205, the CPU 102 compares the number of similar pixels obtained in S2200 with THnear. If it is determined that the condition of number of similar pixels > THnear is met, then in S2206 the CPU 102 determines that the type of content is "photo or gradation." On the other hand, if it is determined that the above condition is not met, then in S2207 the CPU 102 determines that the type of content is "text or graphics."

In this example, the number of similar pixels is 50 and THnear is 100, so it is determined that the condition is not met and in S2207 it is determined to be "text or graphic". On the other hand, for example, in the case of area 2105, the number of area pixels is 80, the number of similar pixels is 50, and THratio is 40. Therefore, in S2202, the condition THabs>THratio is met, so in S2203 THratio=40 is set to THnear. Then, in S2205, the condition number of similar pixels>THnear is met, so in S2206 it is determined to be "photo or gradation".

As described above, in the case of content with a small number of area pixels, it is possible to determine that an area containing only standalone photographic content is a "photo or gradation" by making a judgment using a threshold based on the similar pixel rate. This makes it possible to maintain the characteristic of color continuity by not performing color degeneration correction on areas containing only standalone photographic content.

Here, as an example of graphic content with a high similar pixel rate but a small number of similar pixels, consider graphic content with a small number of area pixels, blurred edges, and solid color. Since such graphic content has a small number of area pixels, it is considered that the benefits of correcting the solid color to improve color discrimination and enhancing saturation to improve appearance are greater than the disadvantages of correcting the blurred parts, such as a decrease in gradation. In the above process, such graphic content is determined to be "photograph or gradation." Here, it is noted that graphic content tends to have many areas where the same color continues. By using a threshold value TH_ratio_same, which is obtained by setting a ratio to the number of area pixels, for the number of pixels, it is possible to improve the accuracy of the determination in this embodiment. The number of pixels in this embodiment is described below.

In this embodiment, in the judgment at S105, the target pixel is inspected based on the edge detection area. If, as a result of comparing the target pixel with its surroundings, a pixel with a high proportion of identical pixels and a low proportion of similar pixels/different pixels is determined to be the same pixel, the target pixel is judged to be the "same pixel." An edge detection filter is scanned within the area, centered on the target pixel, to obtain the number of pixels determined to be the same pixel within the area (= the number of identical pixels).

Figure 23 is a flowchart showing the processing when this pixel rate is used. The processing in Figure 23 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. The processing in Figure 23 is also executed in S105 in Figure 2. S2300 to S2301 are the same as those explained in S2200 to S2201 in Figure 22, so their explanation will be omitted.

In S2302, the CPU 102 obtains the number of identical pixels in the region as described above. In S2303, the CPU 102 compares the number of identical pixels obtained in S2302 with THratio_same. If it is determined that the condition of number of identical pixels < THratio_same is met, the process proceeds to S2304. S2304 to S2309 are the same as those described in S2202 to S2207 of FIG. 22, and therefore their description will be omitted.

On the other hand, if it is determined that the condition of the number of similar pixels<THratio_same is not satisfied, the process proceeds to S2310. In S2310, the CPU 102 compares the number of similar pixels acquired in S2301 with TH_abs. If it is determined that the condition of the number of similar pixels>THabs is satisfied, in S2311, the CPU 102 determines the type of the content to be "photo or gradation". On the other hand, if it is determined that the condition of the number of similar pixels>THabs is not satisfied, in S2312, the CPU 102 determines the type of the content to be "text or graphic". This allows the above-mentioned graphic content with a high similar pixel ratio but a low number of similar pixels to be determined to be "text or graphic". Note that the flowchart in FIG. 23 is an example, and the determination order and the method of processing within each block are not limited to the above. For example, the determination may be made using not only the threshold value of the same pixel ratio obtained by setting the ratio to the number of area pixels, but also the threshold value of the number of similar pixels, or both may be used.

[Eighth embodiment]
The following describes the differences from the first to seventh embodiments. In the seventh embodiment, when the number of similar pixels is greater than either THratio or THabs, it is determined to be a "photograph or gradation." That is, as shown in FIG. 19C, when the number of region pixels exceeds the inflection point IF determined by THratio and THabs, the determination threshold THnear is always THabs. In other words, for any region where the number of region pixels exceeds the inflection point IF, any region that contains a number of similar pixels equal to or greater than THabs is determined to be a "photograph or gradation."

Here, the "text or graphic" content does not necessarily contain no similar pixels. Examples include anti-aliasing and drop shadows. For images with a sufficiently large area, users may view the printed matter from a distance. For example, in-store posters are required to have emphasized saturation, and to be eye-catching and have text that is easy to see. In such cases, it is considered to prioritize the effect of the entire content over the reduction in image quality caused by color degeneration correction of the anti-aliasing part that occurs in the details of the content. If such graphic content contains similar pixels equal to or greater than the judgment threshold THabs, it will be uniformly determined as "photograph or gradation" when the number of region pixels exceeds the inflection point IF. Here, it is possible to determine the above graphic content as "text or graphic" by increasing THabs. However, depending on the size of the graphic content, the desired judgment result cannot be obtained.

For example, suppose there is a graphic content with a similar pixel count of 10 and a region pixel count of 100, and an enlarged version of the graphic content with a similar pixel count of 100 and a region pixel count of 1000. Here, as an example, suppose 50% of the region pixel count is set as THratio and THabs is set to 100. In this case, the region pixel count at the inflection point IF is 200, which is between 100 and 1000. Therefore, for the former graphic content, THratio = 50, and for the latter graphic content, THnear is THratio = 100, and for the latter graphic content, THnear is THabs = 100. As a result, the former graphic content is determined to be "text or graphics", while the latter graphic content is determined to be "photo or gradation".

In order to determine that such graphic content, which has a low similar pixel rate but contains similar pixels, is "text or graphics" if the size is large enough, it is possible to set a determination threshold that depends on the similar pixel rate, even if the number of pixels in the area is large. However, as described in the sixth embodiment, if the determination threshold based on the number of similar pixels is eliminated, an area that contains both graphic content and photo content may be determined to be "text or graphics".

Therefore, in this embodiment, THabs is not fixed, but is controlled using a weighting factor α that depends on the number of pixels in the region. This makes it possible to determine that a region containing both graphic content and photo content is a "photo or gradation." On the other hand, it makes it possible to determine that graphic content that has a low similar pixel rate but contains similar pixels is a "text or graphic." Furthermore, for photo content of a certain size that is the target of THabs, as the number of pixels in the region increases, the proportion of the region that is occupied by photo content decreases. Therefore, the greater the number of pixels in the region, the greater the effect of applying color degeneration correction to the graphic content.

Fig. 19(d) is a diagram showing the relationship between the number of area pixels, THratio, α+THabs, and the judgment threshold THnear. Also, Fig. 24 is a flowchart showing the content type judgment process in this embodiment. The process in Fig. 24 is realized, for example, by the CPU 102 reading a program stored in the storage medium 104 into the RAM 103 and executing it. Also, the process in Fig. 24 is executed in S105 in Fig. 2. S2400, S2401, and S2403 to S2405 are the same as those explained in S2200, S2201, and S2205 to S2207 in Fig. 22, so their explanation will be omitted.

As an example, 50% of the number of region pixels is set as THratio, THabs is set to 100, and weighting factor α is set to (number of region pixels - number of region pixels at inflection point IF) x 0.1. In this case, when the number of region pixels is 200, α + THabs and THratio are the same. That is, in Figure 19(d), P19 is 200 and T19 is 100. If the number of region pixels is greater than 201, the threshold based on the number of similar pixels is used as the judgment criterion.

The processing of S2402 will be explained below with reference to FIG. 25. As an example, a region with 100 similar pixels and 1000 region pixels is processed. First, in S2500, CPU 102 calculates coefficient α. In this example, α = (1000-200) x 0.1 = 80. In S2501, CPU 102 compares α + THabs with THratio and determines whether the condition α + THabs < THratio is satisfied. In this example, α + THabs = 180 and THratio = 500, so it is determined that the condition is satisfied. If it is determined that the condition is satisfied, in S2502, CPU 102 sets α + THabs to THnear. On the other hand, if it is determined in S2501 that the condition is not satisfied, in S2503 the CPU 102 sets THratio to THnear. After S2502 and S2503, the processing in FIG. 25 ends, and then proceeds to S2403 in FIG. 24. Thereafter, the subsequent determination processing is performed based on THnear set in S2402. In this example, THnear is set to 180, so it is determined in S2403 that the condition of number of similar pixels > THnear is not satisfied, and the result is determined to be "text or graphics" in S2405.

As described above, in this embodiment, when the number of area pixels is large, the number of area pixels is weighted in the determination by THabs. This makes it possible to maintain the effect of THabs while also dealing with an increase in the number of similar pixels caused by an increase in the number of area pixels. For example, because graphic content is a certain size or larger, content with a low similar pixel rate but a large number of similar pixels can be appropriately determined to be "text or graphics."

As described above, according to each embodiment, color degeneration correction can be appropriately performed for each content in image data. Such a configuration can reduce degradation in image quality caused by uniformly performing color degeneration correction. Furthermore, the configurations are not limited to those described in each embodiment, and other configurations such as the area setting method and color conversion method may be used as long as the effects described above are obtained. For example, in each embodiment, the image processing device 101 sets areas and performs color conversion using the CPU 102, but the CPU 102 may be replaced with an ASIC, GPU, FPGA, etc.

The present invention can also be realized by supplying a program that realizes one or more of the functions of the above-mentioned embodiments to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program. It can also be realized by a circuit (e.g., an ASIC) that realizes one or more functions.

The disclosure of the present embodiment includes the following image processing device, image processing method, and program.
(Item 1)
An input means for inputting image data;
a generating means for generating image data having a converted color gamut from the image data input by the input means, using a converting means for converting the color gamut of the image data input by the input means into the color gamut of a device that outputs the image data;
a correction unit that corrects the conversion unit based on a result of the conversion of the color gamut;
a control unit that controls the execution of correction by the correction unit on content included in the image data input by the input unit;
Equipped with
the control means controls execution of correction by the correction means on the content based on a type of the content;
When the conversion unit has been corrected by the correction unit, the generation unit generates image data having a converted color gamut from the image data input by the input unit, using the corrected conversion unit;
In the image data whose color gamut has been converted by the conversion means after the correction, the color difference in the image data whose color gamut has been converted by the conversion means before the correction is expanded.
13. An image processing device comprising:
(Item 2)
The image processing device described in item 1, characterized in that the control means controls the correction means to perform correction on the content when the content is of a first type, and controls the correction means not to perform correction on the content when the content is of a second type different from the first type.
(Item 3)
The apparatus further includes a receiving unit for receiving an instruction not to execute the correction by the correction unit,
the control means controls the correction means not to correct the content when the instruction is received by the reception means, even if the content is of the first type.
3. The image processing device according to item 2,
(Item 4)
a first determination means for determining whether the content is of the first type or the second type,
the control means controls execution of correction by the correction means on the content based on a result of the determination by the first determination means.
3. The image processing device according to item 2,
(Item 5)
The image data input by the input means is further analyzed by an analysis means.
5. The image processing device according to claim 4, wherein the first determination means determines whether the content is of the first type or the second type based on a result of the analysis by the analysis means.
(Item 6)
6. The image processing apparatus according to claim 5, wherein the analysis means analyzes a distribution of pixel values in the image data input by the input means.
(Item 7)
7. The image processing device according to item 6, wherein the distribution of pixel values is a distribution of the number of pixels having a first pixel value, the number of pixels having a second pixel value that is in a range similar to the first pixel value, and the number of pixels having a third pixel value that is not in a range similar to the first pixel value.
(Item 8)
The image processing device described in item 7, characterized in that the first determination means determines that the content is of the first type when the number of pixels of the first pixel value is greater than a first threshold, the number of pixels of the second pixel value is smaller than a second threshold, and the number of pixels of the third pixel value is greater than a third threshold.
(Item 9)
9. The image processing device according to item 7 or 8, wherein the first determination means determines that the content is of the second type when the number of pixels of the second pixel value is greater than a fourth threshold value.
(Item 10)
5. The image processing device according to item 4, wherein the first determination means determines whether the content is of the first type or the second type based on a drawing command for image data input by the input means.
(Item 11)
a second determination means for determining a likelihood that the content is of the first type or the second type,
the control means controls execution of correction by the correction means on the content based on a result of the determination by the second determination means.
3. The image processing device according to item 2,
(Item 12)
12. The image processing device according to item 11, wherein the control means controls the correction strength applied to the content by the correction means based on the result of the determination by the second determination means.
(Item 13)
The image processing device described in item 11 or 12, characterized in that the second determination means analyzes the distribution of the number of pixels of a first pixel value in the image data input by the input means, the number of pixels of a second pixel value that is within a similar range to the first pixel value, and the number of pixels of a third pixel value that is not within a similar range to the first pixel value, and determines the likelihood of the image being of the first type or the second type based on the number of pixels of the second pixel value.
(Item 14)
Item 14. The image processing device according to item 13, wherein the control means increases the strength of correction applied by the correction means to the content as the likelihood of the first type increases.
(Item 15)
15. The image processing device according to any one of items 1 to 14, wherein the direction of the expansion of the color difference is the direction of lightness.
(Item 16)
16. The image processing device according to any one of items 1 to 15, wherein the direction of the expansion of the color difference is the direction of saturation.
(Item 17)
17. The image processing device according to any one of items 1 to 16, wherein the direction of the expansion of the color difference is the direction of the hue angle.
(Item 18)
16. The image processing device according to item 15, wherein the color difference between colors included in a predetermined hue angle is expanded in the direction of lightness.
(Item 19)
The image processing device described in item 1, characterized in that the control means controls the correction means to perform correction on the area containing the content when the area containing the content is a first type, and controls the correction means not to perform correction on the area containing the content when the area containing the content is a second type different from the first type.
(Item 20)
The control means
When the area including the content is the first type, control is performed so that the correction means performs correction on the area, and performs color correction corresponding to the first type of content;
when the area including the content is of the second type, control is performed so that the correction means does not perform correction on the area, and color correction corresponding to the second type of content is performed;
3. The image processing device according to item 2,
(Item 21)
a third determination means for determining whether an area including the content is of the first type or the second type,
The control means controls the execution of correction by the correction means for the region based on a result of the determination by the third determination means.
21. The image processing device according to item 20,
(Item 22)
The image data input by the input means is further analyzed by an analysis means.
22. The image processing device according to item 21, wherein the third determination means determines whether an area including the content is of the first type or the second type based on a result of the analysis by the analysis means.
(Item 23)
23. The image processing apparatus according to item 22, wherein the analysis means analyzes a distribution of pixel values of the image data input by the input means.
(Item 24)
The analysis means, when focusing on a specific pixel within the region including the content, acquires the number of similar pixels that are surrounded by pixels whose pixel values are not the same and whose pixel value difference is within a predetermined range;
the third determination means determines the region to be of the second type when the number of similar pixels exceeds a preset threshold based on a result of the analysis by the analysis means;
24. The image processing device according to item 23,
(Item 25)
25. The image processing device according to item 24, wherein the area including the content is an area including the first type of content and the second type of content.
(Item 26)
26. The image processing device according to item 25, wherein the third determination means determines that an area including the content is the second type of content even if the area includes the first type of content.
(Item 27)
27. The image processing device according to item 25 or 26, wherein the first type of content and the second type of content are adjacent to each other within the area.
(Item 28)
27. The image processing device according to item 25 or 26, wherein the first type of content and the second type of content are superimposed on each other.
(Item 29)
29. The image processing device according to any one of items 24 to 28, wherein the threshold value includes a first threshold value that is constant regardless of the number of pixels in the area including the content.
(Item 30)
29. The image processing device according to any one of items 24 to 28, wherein the threshold value includes a second threshold value that increases as the number of pixels in the area including the content increases.
(Item 31)
The image processing device described in any one of items 24 to 28, characterized in that the threshold value includes a first threshold value that is constant regardless of the number of pixels in the area including the content, and a second threshold value that increases as the number of pixels in the area including the content increases.
(Item 32)
An image processing method executed in an image processing device, comprising:
an input step of inputting image data;
a generating step of generating image data that has been gamut-converted from the image data input in the input step, using a conversion means for converting the color gamut of the image data input in the input step into the color gamut of a device that outputs the image data;
a correction step of correcting the conversion means based on a result of the conversion of the color gamut;
a control step of controlling execution of correction in the correction step on content included in the image data input in the input step;
having
In the control step, execution of the correction on the content in the correction step is controlled based on a type of the content;
When the conversion means is corrected in the correction step, the generation step generates image data whose color gamut has been converted from the image data input in the input step using the corrected conversion means,
In the image data whose color gamut has been converted by the conversion means after the correction, the color difference in the image data whose color gamut has been converted by the conversion means before the correction is expanded.
13. An image processing method comprising:
(Item 33)
32. A program for causing a computer to function as each of the means of the image processing device according to any one of items 1 to 31.

The invention is not limited to the above-described embodiment, and various modifications and variations are possible without departing from the spirit and scope of the invention. Therefore, the following claims are appended to disclose the scope of the invention.

101 Image processing device: 108 Recording device: 102, 111 CPU: 103, 112 RAM: 104, 113 Storage medium

Claims (33)

An input means for inputting image data;
a generating means for generating image data having a converted color gamut from the image data input by the input means, using a converting means for converting the color gamut of the image data input by the input means into the color gamut of a device that outputs the image data;
a correction unit that corrects the conversion unit based on a result of the conversion of the color gamut;
a control unit that controls the execution of correction by the correction unit on content included in the image data input by the input unit;
Equipped with
the control means controls execution of correction by the correction means on the content based on a type of the content;
When the conversion unit has been corrected by the correction unit, the generation unit generates image data having a converted color gamut from the image data input by the input unit, using the corrected conversion unit;
In the image data whose color gamut has been converted by the conversion means after the correction, the color difference in the image data whose color gamut has been converted by the conversion means before the correction is expanded.
13. An image processing device comprising: The image processing device according to claim 1, characterized in that the control means controls the correction means to perform correction on the content when the content is of a first type, and controls the correction means not to perform correction on the content when the content is of a second type different from the first type. The apparatus further includes a receiving unit for receiving an instruction not to execute the correction by the correction unit,
the control means controls the correction means not to correct the content when the instruction is received by the receiving means, even if the content is of the first type.
3. The image processing device according to claim 2. a first determination means for determining whether the content is of the first type or the second type,
the control means controls execution of correction by the correction means on the content based on a result of the determination by the first determination means.
3. The image processing device according to claim 2. The image data input by the input means is further analyzed by an analysis means.
5. The image processing apparatus according to claim 4, wherein the first determination means determines whether the content is of the first type or the second type based on a result of the analysis by the analysis means. The image processing device according to claim 5, characterized in that the analysis means analyzes the distribution of pixel values in the image data input by the input means. The image processing device according to claim 6, characterized in that the distribution of pixel values is a distribution of the number of pixels having a first pixel value, the number of pixels having a second pixel value that is in a range similar to the first pixel value, and the number of pixels having a third pixel value that is not in a range similar to the first pixel value. The image processing device according to claim 7, characterized in that the first determination means determines that the content is of the first type when the number of pixels of the first pixel value is greater than a first threshold, the number of pixels of the second pixel value is less than a second threshold, and the number of pixels of the third pixel value is greater than a third threshold. The image processing device according to claim 7, characterized in that the first determination means determines that the content is of the second type when the number of pixels of the second pixel value is greater than a fourth threshold value. The image processing device according to claim 4, characterized in that the first determination means determines whether the content is the first type or the second type based on a drawing command for image data input by the input means. a second determination means for determining a likelihood that the content is of the first type or the second type,
the control means controls execution of correction by the correction means on the content based on a result of the determination by the second determination means.
3. The image processing device according to claim 2. The image processing device according to claim 11, characterized in that the control means controls the correction strength applied by the correction means to the content based on the result of the judgment by the second judgment means. The image processing device according to claim 11, characterized in that the second determination means analyzes the distribution of the number of pixels of a first pixel value in the image data input by the input means, the number of pixels of a second pixel value that is in a similar range to the first pixel value, and the number of pixels of a third pixel value that is not in a similar range to the first pixel value, and determines the likelihood of the first type or the second type based on the number of pixels of the second pixel value. The image processing device according to claim 13, characterized in that the control means increases the correction strength applied by the correction means to the content as the likelihood of the first type increases. The image processing device according to claim 1, characterized in that the direction of the color difference expansion is the direction of brightness. The image processing device according to claim 1, characterized in that the direction of the expansion of the color difference is the direction of saturation. The image processing device according to claim 1, characterized in that the direction of the color difference expansion is the direction of the hue angle. The image processing device according to claim 15, characterized in that the color difference between colors included in a predetermined hue angle is expanded in the direction of lightness. The image processing device according to claim 1, characterized in that the control means controls the correction means to perform correction on the area containing the content if the area is a first type, and controls the correction means not to perform correction on the area containing the content if the area is a second type different from the first type. The control means
When the area including the content is the first type, control is performed so that the correction means performs correction on the area, and performs color correction corresponding to the first type of content;
if the area including the content is of the second type, control is performed so that the correction means does not perform correction on the area, and color correction corresponding to the second type of content is performed.
3. The image processing device according to claim 2. a third determination means for determining whether an area including the content is of the first type or the second type,
The control means controls the execution of correction by the correction means for the region based on a result of the determination by the third determination means.
21. The image processing device according to claim 20. The image data input by the input means is further analyzed by an analysis means.
22. The image processing apparatus according to claim 21, wherein the third determination means determines whether an area including the content is of the first type or the second type based on a result of the analysis by the analysis means. The image processing device according to claim 22, characterized in that the analysis means analyzes the distribution of pixel values of the image data input by the input means. The analysis means, when focusing on a specific pixel within the region including the content, acquires the number of similar pixels that are surrounded by pixels whose pixel values are not the same and whose pixel value difference is within a predetermined range;
the third determination means determines the region to be of the second type when the number of similar pixels exceeds a preset threshold based on a result of the analysis by the analysis means;
24. The image processing device according to claim 23 . The image processing device according to claim 24, characterized in that the area containing the content is an area containing the first type of content and the second type of content. The image processing device according to claim 25, characterized in that the third determination means determines that the area containing the content is the second type of content even if the area contains the first type of content. The image processing device according to claim 25, characterized in that the first type of content and the second type of content are adjacent to each other within the area. The image processing device according to claim 25, characterized in that the first type of content and the second type of content are superimposed on each other. The image processing device according to claim 24, characterized in that the threshold value includes a first threshold value that is constant regardless of the number of pixels in the area including the content. The image processing device according to claim 24, characterized in that the threshold includes a second threshold that increases as the number of pixels in the area including the content increases. The image processing device according to claim 24, characterized in that the thresholds include a first threshold that is constant regardless of the number of pixels in the region including the content, and a second threshold that increases as the number of pixels in the region including the content increases. An image processing method executed in an image processing device, comprising:
an input step of inputting image data;
a generating step of generating image data that has been subjected to color gamut conversion from the image data input in the input step, using a conversion means for converting the color gamut of the image data input in the input step into the color gamut of a device that outputs the image data;
a correction step of correcting the conversion means based on a result of the conversion of the color gamut;
a control step of controlling execution of correction in the correction step on content included in the image data input in the input step;
having
In the control step, execution of the correction on the content in the correction step is controlled based on a type of the content;
When the conversion means is corrected in the correction step, the generation step generates image data in which a color gamut has been converted from the image data input in the input step using the corrected conversion means,
In the image data whose color gamut has been converted by the conversion means after the correction, the color difference in the image data whose color gamut has been converted by the conversion means before the correction is expanded.
13. An image processing method comprising: A program for causing a computer to function as each of the means of the image processing device according to any one of claims 1 to 31.

Images