JP5757904B2 - Image compression device - Google Patents

Description

  The present invention relates to an image compression apparatus.

  An image compression apparatus temporarily stores input image data in a buffer, and sequentially reads the stored image data for each block having a predetermined number of pixels. This image compression device extracts color information of characters / line drawings from the read image data of each block, and identifies whether the area in the image is a character / line drawing area or a non-character / line drawing area To extract attribute information. The image compression apparatus replaces the color information of the character / line drawing area in the block with the average value of the color information of the area other than the character / line drawing area in the block. The image compression apparatus compresses the color information and attribute information of the character / line drawing by the lossless compression method, and the color information of the area other than the character / line drawing area (original color information not replaced) and the character / line drawing. The image data including the color information (replaced color information) of the area is compressed by the irreversible compression method (see, for example, Patent Document 1).

  In addition, a certain image compression apparatus converts the color information of the character / line drawing area in the block into the DC component value of the frequency conversion when it is assumed that all the areas in the block are areas other than the character / line drawing area. Replace with approximate values. The image compression apparatus compresses the color information and attribute information of the character / line drawing by the lossless compression method, and the color information of the area other than the character / line drawing area (original color information not replaced) and the character / line drawing. The image data composed of the color information (replaced color information) of the area is compressed by the lossy compression method (see, for example, Patent Document 2).

JP-A-4-326669 JP 2008-104120 A

  However, since the image compression apparatuses described in Patent Documents 1 and 2 compress character / line image color information by a reversible compression method, the compression rate is low. For this reason, in order to increase the compression rate, it is conceivable to compress the color information of characters / line drawings using an irreversible compression method. However, if this is done, an image such as mosquito noise will be added to the edges of the characters or line drawings. Disturbance may occur and the contrast of characters and line drawings may be impaired.

  The present invention has been made in view of the above problems, and an object of the present invention is to obtain an image compression apparatus that compresses image data at a high compression rate without impairing the contrast around an edge in an image.

  In order to solve the above problems, the present invention is configured as follows.

  An image compression device according to the present invention includes a pixel block extraction unit that extracts image data and attribute information corresponding to the image data by dividing the image data into pixel blocks having a predetermined number of pixels, and each of the pixel block extraction units extracted by the pixel block extraction unit. In the pixel block, the attribute pixel in the pixel block is determined for each attribute according to the attribute information, and the pixel value of the pixel other than the attribute pixel is replaced with a predetermined value obtained from the attribute pixel. A color information generation unit that generates color information including a predetermined value of a pixel other than the pixel and a pixel value of an attribute pixel, an irreversible compression unit that compresses color information for each attribute using an irreversible compression method, and attribute information And a reversible compression unit that compresses the image using a reversible compression method.

  As a result, the image data is compressed at a high compression rate without impairing the contrast around the edges in the image.

The image compression apparatus according to the present invention includes a lossy decompression unit for decompressing the color information for each attribute that is compressed by the irreversible compression unit respectively, and the lossless decompression unit for decompressing the attribute information compressed by lossless compression unit And a color information selection unit that determines an attribute of each pixel based on the attribute information expanded by the reversible expansion unit and selects a pixel value from the color information for each pixel according to the determined attribute.

  As a result, the decompression process is executed in correspondence with the above-described compression process, so that the contrast around the edge in the image obtained by decompressing the compressed image data is not impaired.

In the image compression device according to the present invention , the attribute information indicates whether the pixel attribute is a character attribute or a non-character attribute, and the color information generation unit includes the pixel value of the pixel of the character attribute in the pixel block. Is replaced with a predetermined value derived from the pixel value of the non-character attribute pixel in the pixel block, and non-character attribute color information including the pixel value of the non-character attribute pixel and the predetermined value of the character attribute pixel is obtained. The non-character attribute color information generation unit to be generated and the pixel value of the non-character attribute pixel in the pixel block are replaced with a predetermined value derived from the pixel value of the character attribute pixel in the pixel block. A character attribute color information generation unit configured to generate character attribute color information including pixel values of pixels and predetermined values of non-character attribute pixels;

  Thereby, the image data is compressed at a high compression rate without losing the contrast around the edge of the character.

In the image compression apparatus according to the present invention, the predetermined value described above are representative of attributes of the pixels in the pixel block. Then, the maximum value of each of R, G, and B of the non-character attribute pixel in the pixel block is used as the representative value for the non-character attribute pixel, and the representative value for the character attribute pixel is the pixel value. The minimum values of R, G, and B for character attribute pixels in the block are used.

  This reduces the size of the color information after compression.

The image compression apparatus according to the present invention, in addition to the image compression equipment described above may be as follows. In this case, when the color information generation unit determines that all the pixels in the pixel block are pixels other than the pixels having the above-described attribute, the pixel value of all the pixels in the pixel block is set to a predetermined value. Replace.

  This reduces the size of the color information after compression.

  The image compression apparatus according to the present invention may be as follows in addition to any of the image compression apparatuses described above. In this case, the irreversible compression unit compresses the color information for each attribute using JPEG, and the irreversible decompression unit decompresses the color information for each attribute compressed by the irreversible compression unit using JPEG.

  As a result, the color information is compressed at a high compression rate by JPEG.

  According to the present invention, in an image compression apparatus, image data is compressed at a high compression rate without impairing the contrast around the edges in the image.

FIG. 1 is a block diagram showing a configuration of an image compression apparatus according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of attribute information detected by the character attribute detection unit in FIG. 1 and a pixel block extracted by the block extraction unit. FIG. 3 is a flowchart for explaining the operation at the time of image compression in the image compression apparatus shown in FIG. FIG. 4 is a flowchart for explaining the operation at the time of image expansion in the image compression apparatus shown in FIG.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an image compression apparatus according to an embodiment of the present invention. In this embodiment, in describing the configuration and operation of the image compression apparatus shown in FIG. 1, an image in which a character or line drawing image and an image having gradation such as a photo or a pattern other than the character and line drawing are mixed. Assume that image data is compression-encoded. In this embodiment, a character or line drawing image may be simply referred to as a “character image”. In addition, an image having a gradation other than characters and line drawings may be referred to as a “gradation image”.

  The image compression apparatus shown in FIG. 1 includes an image compression processing unit 1 that executes compression processing and decompression processing of image data. As shown in FIG. 1, the image compression processing unit 1 includes a compression processing unit 10, an expansion processing unit 20, and a storage unit 30.

  The compression processing unit 10 is a processing unit that executes compression processing of image data. As shown in FIG. 1, the compression processing unit 10 includes a character attribute detection unit 11, a block extraction unit 12, a first replacement unit 13, a second replacement unit 14, a first JPEG compression unit 15, a second JPEG compression unit 16, and An MMR compression unit 17 is provided.

  The character attribute detection unit 11 reads out image data of an input image having a predetermined size (for example, the number of vertical × horizontal pixels is 640 × 480 pixels, 1024 × 768 pixels, etc.) stored in the storage unit 30, and the read image This is a processing unit that detects, for each pixel, attribute information (see FIG. 2) for identifying whether the attribute of each pixel of data is a character attribute or a non-character attribute. Here, “character attribute” refers to an attribute indicating that an area in the image is a character or line drawing area, and “non-character attribute” refers to an area in the image other than a character or line drawing (ie , A tone region). In the example illustrated in FIG. 2, the attribute information indicating “character attribute” is “1”, and the attribute information indicating “non-character attribute” is “0”.

  The block extraction unit 12 converts image data and attribute information of a predetermined size (for example, vertical × horizontal number of pixels of 640 × 480 pixels, 1024 × 768 pixels, etc.) into blocks of a predetermined number of pixels (in this embodiment, 8 The color information and attribute information of each pixel in the block are extracted by dividing into blocks of × 8 pixels (see FIG. 2). Here, “color information” is information representing the color of each pixel. For example, when each color of R (blue), G (green), and B (blue) has 256 gradations, the “color information” of each pixel is data of 256 gradations of each color (R, G, B, respectively). 8 bits of data). In this embodiment, the value of the color information of each pixel may be referred to as “pixel value”. For example, when each color of R, G, and B has 256 gradations, the “pixel value” is one of 0 to 255 in each color.

  Based on the attribute information, the first replacement unit 13 sets the pixel value of the pixel with the character attribute in the block extracted by the block extraction unit 12 to the representative value or the average value of the pixel value of the pixel with the non-character attribute in the block. (That is, a predetermined one value) is replaced with a pixel value (original pixel value) of a non-character attribute pixel and a pixel value (post-replacement value) of a character attribute pixel in the block. One block of color information for the non-character attribute is generated. The first replacement unit 13 performs such processing for all the blocks extracted by the block extraction unit 12.

  Based on the attribute information, the second replacement unit 14 sets the pixel value of the non-character attribute pixel in the block extracted by the block extraction unit 12 to the representative value or average value of the pixel value of the character attribute pixel in the block. (That is, a predetermined one value) is replaced with a pixel value (original pixel value) of a character attribute pixel and a pixel value (post-replacement value) of a non-character attribute pixel in the block. One block of color information is generated for the character attribute. The second replacement unit 14 performs such processing for all the blocks extracted by the block extraction unit 12. The above-described representative value is, for example, the maximum value or the minimum value for each color of RGB. Further, the above average value is, for example, an average value for each color of RGB.

  The first JPEG compression unit 15 compresses the color information for one block for the non-character attribute generated by the first replacement unit 13 using JPEG (Joint Photographic Experts Group), which is an irreversible compression method, and the non-character attribute. Is stored in the storage unit 30. The second JPEG compression unit 16 compresses the color information for one block for the character attribute generated by the second replacement unit 14 by JPEG, and stores the compressed color information for one block for the character attribute in the storage unit 30. To do.

  Here, in JPEG, image data (color information) for one block is subjected to YUV conversion, DCT (Discrete Cosine Transform) conversion is performed on the data after YUV conversion, the DCT converted data is quantized, The image data is compressed by encoding the converted data with a Huffman code.

  The MMR compression unit 17 compresses the attribute information using MMR (Modified Modified READ), which is a lossless compression method, and stores the compressed attribute information in the storage unit 30.

  The decompression processing unit 20 is a processing unit that performs image data decompression processing. As illustrated in FIG. 1, the decompression processing unit 20 includes a first JPEG decompression unit 21, a second JPEG decompression unit 22, an MMR decompression unit 23, and a selector unit 24.

  The first JPEG decompression unit 21 decompresses the color information about the non-character attribute compressed by the first JPEG compression unit 15 using the lossy compression method using the lossy compression method. The second JPEG decompression unit 22 decompresses the color information about the character attribute compressed by the second JPEG compression unit 16 using the lossy compression method and the lossy compression method. The MMR decompression unit 23 decompresses the attribute information compressed by the MMR compression unit 17 using the lossless compression method using the lossless compression method.

  The selector unit 24 determines whether the attribute of each pixel is a non-character attribute or a character attribute based on the attribute information expanded by the MMR compression unit 17, and the first JPEG expansion unit 21 or the second JPEG expansion unit 22. The color information of the non-character attribute or the character attribute expanded by the above is selected for each pixel, and the image data is reconstructed.

  The storage unit 30 is a storage device that can store various data including image data. For the storage unit 30, for example, a semiconductor memory such as a RAM (Random Access Memory) is used. In this embodiment, the storage unit 30 stores at least input image data, image data that has been subjected to compression processing by the compression processing unit 10, and image data that has undergone expansion processing by the expansion processing unit 20. It can be memorized.

  As shown in FIG. 1, a display 41, a printer 42, and a scanner 43 are connected to the image compression processing unit 1. The display 41 is a device that displays an image on a display screen based on image data. The printer 42 is a device that prints an image based on image data. The scanner 43 is a device that optically reads an image and generates image data of the image.

  Although not shown in FIG. 1, the image compression processing unit 1 is connected to a communication network, and has a network interface for performing data communication with other devices (for example, personal computers) connected to the communication network. I have. Therefore, the image compression processing unit 1 can receive the image data transmitted from another device via the network interface, and execute the compression process of the received image data.

  Next, the operation of the image compression apparatus will be described.

  FIG. 3 is a flowchart for explaining the operation at the time of image compression in the image compression apparatus shown in FIG.

  By receiving image data captured by the scanner 43 or transmitted from another device via the network interface, the image data is input and stored in the storage unit 30. When there is a request for compression processing on the predetermined image data stored in the storage unit 30, the compression processing unit 10 executes each process shown in FIG.

  The character attribute detection unit 11 of the compression processing unit 10 reads the image data from the storage unit 30, specifies whether the attribute of each pixel of the read image data is a character attribute or a non-character attribute, and sets the attribute for each pixel. Information is output (step S1).

  For example, the character attribute detection unit 11 performs a character or line drawing region and a non-character region other than a character and a line drawing (that is, a photograph or a picture) based on a known region separation algorithm such as edge extraction processing from the image data of the input image. Etc.). Then, for example, as shown in FIG. 2, the character attribute detecting unit 11 is an attribute for identifying a character attribute indicating a character or line drawing area and a non-character attribute indicating a non-character area. Generate information. The character attribute detection unit 11 outputs the attribute information of each pixel to the block extraction unit 12.

  Next, the block extraction unit 12 divides the image data (color information) into blocks of an image size of 8 × 8 pixels, and the attribute information output from the character attribute detection unit 11 as shown in FIG. 2, for example. The block is divided into 8 × 8 pixel image blocks, and color information and attribute information of each pixel in the block are extracted block by block (step S2).

  The block extraction unit 12 outputs the color information and attribute information of each pixel in the extracted block to the first replacement unit 13 and the second replacement unit 14. Further, the block extraction unit 12 outputs the attribute information of each pixel in the extracted block to the MMR compression unit 17.

  Next, when the color information and attribute information of each block is input from the block extraction unit 12, the first replacement unit 13 determines whether all the pixels in the block are pixels with character attributes based on the attribute information. It is determined whether or not the attribute information of all the pixels in the block is “1” (step S3).

  When it is determined that at least one pixel in the block is not a character attribute pixel (that is, when some of the pixels in the block are character attribute pixels), the first replacement unit 13 The pixel value of a pixel with a character attribute (a pixel set with “1” in FIG. 2) is the pixel value of a pixel with a non-character attribute within the block (a pixel with “0” set in FIG. 2) Is replaced with a representative value or an average value (step S4). By this processing, color information for one block is generated for the non-character attribute composed of the pixel value (original pixel value) of the non-character attribute pixel and the pixel value (post-replacement value) of the character attribute pixel. Is done.

  Here, as the “representative value” of the non-character attribute pixel in the block, for example, the maximum value of each of R, G, B of the non-character attribute pixel in the block is used. Specifically, when the gradation of each color is 256 gradations, and the maximum values of R, G, B of non-character attribute pixels in the block are values of (250, 200, 100), The maximum value (250, 200, 100) of each color is set as a representative value for each pixel (character attribute pixel) for which “1” is set in 2.

  In addition, the “average value” of non-character attribute pixels in the block is the sum of the pixel values of the non-character attribute pixels in the block for each color, and the added value for each color is the non-character attribute in the block. It is calculated by dividing by the number of pixels. If the calculated average value is not an integer, the value after the decimal point is rounded down or rounded off to the integer value.

  On the other hand, when it is determined that all the pixels in the block are character attribute pixels, the first replacement unit 13 predetermines pixel values of all the pixels (8 × 8 pixels in FIG. 2) in the block. The obtained value is replaced (step S5). By performing such processing, color information for one block is generated for a non-character attribute, in which all pixel values are predetermined values in a certain block.

  Here, the “predetermined value” may be an arbitrary value. For example, if the gradation of each color is 256 gradations, the “predetermined value” is any one value from “0” to “255”. However, all blocks to be replaced with “predetermined value” are conditional on being replaced with the same value (constant value) as “predetermined value”. In this way, if the same value is replaced between blocks, the value can be encoded into the same data and compressed.

  The first replacement unit 13 performs such processing (steps S3 to S5) for all the blocks extracted by the block extraction unit 12. Further, the first replacement unit 13 outputs the color information of each block regarding the generated non-character attribute to the first JPEG compression unit 15.

  In addition, when the color information and attribute information of each block is input from the block extraction unit 12, the second replacement unit 14 determines whether all the pixels in the block are non-character attribute pixels based on the attribute information. It is determined whether or not the attribute information of all the pixels in the block is “0” (step S6).

  When it is determined that at least one pixel in the block is not a non-character attribute pixel (that is, when some pixels in the block are non-character attribute pixels), the second replacement unit 14 The pixel values of all the non-character attribute pixels (pixels set to “0” in FIG. 2) in the pixel are the pixel values of the character attribute pixels (pixels set to “1” in FIG. 2) in the block. The pixel value is replaced with a representative value or an average value (step S7). By this process, color information for one block is generated for the character attribute composed of the pixel value (original pixel value) of the pixel of the character attribute and the pixel value (post-replacement value) of the pixel of the non-character attribute. The

  Here, as the “representative value” of the character attribute pixel in the block, for example, the minimum values of R, G, and B of the character attribute pixel in the block are used. Specifically, when the gradation of each color is 256 gradations and the minimum values of R, G, and B of the character attribute pixels in the block are (0, 0, 0), “ The minimum value (0, 0, 0) of each color is set as a representative value for each of the pixels for which “0” is set (non-character attribute pixels).

  In addition, the “average value” of the character attribute pixels in the block is obtained by adding the pixel values of the character attribute pixels in the block for each color, and adding the values for each color of the character attribute pixels in the block. Calculated by dividing by number. If the calculated average value is not an integer, the value after the decimal point is rounded down or rounded off to the integer value.

  On the other hand, when it is determined that all the pixels in the block are non-character attribute pixels, the second replacement unit 14 selects all the non-character attribute pixels (8 × 8 pixels in FIG. 2) in the block. The value is replaced with a predetermined value (step S8). By performing such processing, color information for one block is generated for a character attribute in which all pixel values are predetermined values in a certain block. Again, the “predetermined value” may be an arbitrary value.

  The second replacement unit 14 performs such processing (steps S6 to S8) for all the blocks extracted by the block extraction unit 12. Further, the second replacement unit 14 outputs the color information of each block regarding the generated character attribute to the second JPEG compression unit 16. Note that the processing of steps S6 to S8 may be performed in parallel with the processing of steps S3 to S5.

  Next, the first JPEG compression unit 15 compresses the color information of each block for the non-character attribute generated by the first replacement unit 13 with JPEG, which is an irreversible compression method, and the second JPEG compression unit 16 The color information of each block for the character attribute generated by the two replacement unit 14 is compressed with JPEG, which is an irreversible compression method (step S9). The non-character attribute color information and the character attribute color information compressed by the first JPEG compression unit 15 and the second JPEG compression unit 16 are stored in the storage unit 30 in association with each other. Note that the first JPEG compression unit 15 and the second JPEG compression unit 16 may operate in parallel.

  Further, when the attribute information is input from the block extraction unit 12, the MMR compression unit 17 compresses the attribute information, for example, for each block by MMR, which is a lossless compression method (step S10). The attribute information compressed by the MMR compression unit 17 is also stored in the storage unit 30 in association with the compressed color information. In addition, you may make it perform the process of step S10 in parallel with the process of step S3-S5, the process of step S6-S8, the process of step S9, etc.

  FIG. 4 is a flowchart for explaining the operation at the time of image expansion in the image compression apparatus shown in FIG.

  When there is a request for decompression processing on the image data after compression processing (compressed color information and attribute information associated with each other) stored in the storage unit 30, each of the decompression processing units 20 is shown in FIG. Execute the process.

  The first JPEG decompression unit 21 stores, from the storage unit 30, the compressed color information about the non-character attribute (color information about the non-character attribute compressed by the first JPEG compression unit 15) of the image data requested to be decompressed. The color information about the read and read non-character attribute is expanded by JPEG (step S11). Further, the second JPEG decompression unit 22 stores the compressed color information about the character attribute (color information about the character attribute compressed by the second JPEG compression unit 16) of the image data requested to be decompressed from the storage unit 30. The color information on the read and read character attributes is expanded by JPEG (step S11). The color information about the non-character attribute and the color information about the character attribute expanded by the first JPEG expansion unit 21 and the second JPEG expansion unit 22 are output to the selector unit 24.

  Further, the MMR decompression unit 23 reads the attribute information (attribute information compressed by the MMR compression unit 17) of the image data requested for decompression processing from the storage unit 30, and decompresses the read attribute information by MMR (step S12). ). The attribute information expanded by the MMR expansion unit 23 is output to the selector unit 24. Note that the processing in steps S11 and S12 may be performed in parallel.

  The selector unit 24 determines whether the attribute of each pixel is a non-character attribute or a character attribute based on the attribute information expanded by the MMR expansion unit 23, and the first JPEG expansion unit 21 or the second JPEG expansion unit 22. The color information of the non-character attribute or the character attribute expanded by the above is selected for each pixel to reconstruct the image data (step S13).

  For example, an image based on the image data reconstructed by the selector unit 24 is output to the display 41 and displayed on the display screen, or printed by the printer 42.

  As described above, according to the above-described embodiment, the block extracting unit 12 extracts the image data and the attribute information by dividing them into blocks having a predetermined number of pixels. Then, the first replacement unit 13 replaces the pixel value of the pixel of the character attribute in the block with the representative value or the average value of the pixel value of the pixel of the non-character attribute in the block. Generate block color information. In addition, the second replacement unit 14 replaces the pixel value of the non-character attribute pixel in the block with the representative value or the average value of the pixel value of the character attribute pixel in the block, and each block for the character attribute Generate color information for. The first JPEG compression unit 15 compresses the color information of each block for the non-character attribute generated by the first replacement unit 13 using an irreversible compression method. The second JPEG compression unit 16 compresses the color information of each block for the character attribute generated by the second replacement unit 14 using an irreversible compression method. Further, the MMR compression unit 17 compresses the attribute information by a lossless compression method.

  As a result, not only non-character attribute color information (that is, image data of gradation images) but also character attribute color information (character / line drawing image data) is compressed by the lossy compression method. Image data is compressed at a compression rate. In addition, even if the color information about the character attribute is compressed by the lossy compression method, the pixel value of the non-character attribute around the character attribute pixel is not used for the reconstruction of the image data after decompression for the color information. The occurrence of mosquito noise can be suppressed, and the contrast around the edges of characters and the like is not impaired.

  The above-described embodiments are preferred examples of the present invention, but the present invention is not limited to these, and various modifications and changes can be made without departing from the scope of the present invention. is there.

  For example, in the above-described embodiment, the image compression processing unit 1 is realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) in which a plurality of functional circuits are combined into one to perform a specific application. That is, the image compression processing unit 1 is realized by, for example, an IC chip equipped with a plurality of circuits that execute processing of various processing units. However, the image compression processing unit 1 includes a computer having a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM, and the like, and loads the program from the ROM or a storage device (not shown) to the RAM. Various processing units may be realized by being executed by the CPU.

  In the above embodiment, in addition to the display 41, the printer 42, and the scanner 43, a device such as a facsimile device may be connected to the image compression processing unit 1.

  In the above embodiment, instead of the “average value” of non-character attribute and character attribute pixels in the block, “median value” or “mode value” of non-character attribute and character attribute pixels in the block ”May be calculated and used.

  In the above embodiment, JPEG is used as the irreversible compression method, but an irreversible compression method other than JPEG may be used. In the above-described embodiment, MMR is used as the lossless compression method. However, lossless compression methods such as MR (Modified READ) and JBIG (Joint Bi-level Image experts Group) can also be used.

  In the above embodiment, the case where the color space of the image data is RGB has been described. However, CMY (Cyan, Magenta, Yellow) may be used. In the case of the CMY color space, the maximum value and the minimum value described above are opposite to those in the case of the RGB color space.

  In the above embodiment, the number of attributes is two (character attributes and non-character attributes), but three or more attributes may be used. In this case, the same number of JPEG compression units and JPEG expansion units as the number of attributes are provided, and compressed color information of each block is generated for each of three or more attributes. Is done. For example, the attribute information may indicate three attributes: a character, a vector graphic, and others.

  The present invention is applicable to, for example, a multifunction machine.

DESCRIPTION OF SYMBOLS 1 Image compression process part 10 Compression process part 11 Character attribute detection part 12 Block extraction part (an example of a pixel block extraction part)
13 1st replacement part (a part of an example of a color information generation part, an example of a non-character attribute color information generation part)
14 Second replacement unit (a part of an example of a color information generation unit, an example of a character attribute color information generation unit)
15 First JPEG compression unit (part of an example of irreversible compression unit)
16 Second JPEG compression unit (part of an example of irreversible compression unit)
17 MMR compression unit (an example of a lossless compression unit)
20 Decompression processing unit 21 First JPEG decompression unit (part of an example of irreversible decompression unit)
22 Second JPEG decompression unit (part of an example of irreversible decompression unit)
23 MMR expansion unit (an example of a reversible expansion unit)
24 selector section (example of color information selection section)
30 storage unit

Claims (3)

A pixel block extraction unit that extracts image data and attribute information corresponding to the image data by dividing the image data into pixel blocks having a predetermined number of pixels;
In each pixel block extracted by the pixel block extraction unit, for each attribute according to the attribute information, the pixel having the attribute in the pixel block is determined, and the predetermined value obtained from the determined pixel having the attribute is determined. Replacing the pixel value of a pixel other than the attribute pixel, and generating color information including the predetermined value of the pixel other than the attribute pixel and the pixel value of the attribute pixel;
An irreversible compression unit that compresses the color information for each attribute by an irreversible compression method;
A reversible compression unit that compresses the attribute information by a reversible compression method;
An irreversible decompression unit for decompressing the color information for each attribute compressed by the irreversible compression unit;
A lossless decompression unit that decompresses the attribute information compressed by the lossless compression unit;
A color information selection unit that determines an attribute of each pixel based on the attribute information expanded by the reversible expansion unit, and selects a pixel value from the color information for each pixel according to the determined attribute;
Equipped with a,
The attribute information indicates whether the attribute of the pixel is a character attribute or a non-character attribute,
The color information generation unit
The pixel value of the pixel of the character attribute in the pixel block is replaced with the predetermined value derived from the pixel value of the pixel of the non-character attribute in the pixel block, and the pixel value of the pixel of the non-character attribute A non-character attribute color information generation unit that generates non-character attribute color information including the predetermined value of the pixel of the character attribute;
The pixel value of the non-character attribute pixel in the pixel block is replaced with the predetermined value derived from the pixel value of the character attribute pixel in the pixel block, and the pixel value of the character attribute pixel and the pixel value A character attribute color information generation unit that generates character attribute color information including the predetermined value of the non-character attribute pixel;
The predetermined value is a representative value of the pixel of the attribute in the pixel block,
As the representative value for the non-character attribute pixel, the maximum value of each of R, G, and B of the non-character attribute pixel in the pixel block is used.
For the representative value for the character attribute pixel, the minimum value of each of R, G, B of the character attribute pixel in the pixel block is used.
An image compression apparatus characterized by the above. If the color information generation unit determines that all the pixels in the pixel block are pixels other than the pixel having the attribute, the color information generation unit replaces the pixel values of all the pixels in the pixel block with predetermined values. image compression apparatus according to claim 1 Symbol mounting, characterized in that. The irreversible compression unit, the image compression apparatus according to claim 1 or claim 2, wherein the compressing the color information of each of the attributes in JPEG, respectively.

Images