JPH04313744A - Surface removing system for image recorder - Google Patents

Abstract

PURPOSE:To obtain an optimum surface removing level every original and to clearly remove a surface. CONSTITUTION:A sampling part 13 samples the image data of an original which is read at the time of prescanning at specified intervals in a main scanning direction, and generates gate signals in the main scanning direction and in a subscanning direction at the same time. The sampled data and the gate signal are inputted in a histogram generation part 14, and the histogram of every specified density area is generated. A surface removing level decision part 15 searches the frequency of the histogram from a high density side, detects the density area which exceeds a specified value first, and obtains the threshold of surface removing density corresponding to the detected density area. The decided threshold is given to a surface removing part 11 at the time of copying scanning. The surface removing part 11 outputs the density value which is set at zero to a picture element whose density is equal to or under the threshold, outputs the inputted density value as it is to a picture element whose density >=1.5 the threshold, and outputs the inputted density value which is made three times as large as the inputted density value to a picture element whose density exceeds the threshold and which is <=1.5 the threshold.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to background removal in image recording apparatuses such as copying machines, facsimiles, and printers.

0002

[Prior Art] Image recording devices such as copying machines, facsimile machines, and printers handle not only originals using normal white paper, but also paper with various background densities, such as newspapers, straw paper, recycled paper, blueprints, etc. Manuscripts using this are subject to reading. Originals using paper other than normal white paper have a high background density, so if digital copying machines output the image data read by a document reading device such as a CCD sensor as is, the background of the original will also be affected. It ends up being reproduced and becomes unclean. Therefore, in an image recording apparatus, a background removal circuit is generally provided in order to remove the background from a document having a constant density. An example is shown in FIG. FIG. 6 is a diagram showing a schematic configuration of a real-time background removal method previously proposed by the present applicant, in which density data of a predetermined number of bits, for example 8 bits, is input to the background level detection section 1. Such digital density data is obtained by performing predetermined signal processing such as shading correction on three-color analog signals of green (G), blue (B), and red (R) output from a full-color sensor, and then digitizing them. It is clear to those skilled in the art that it can be obtained by performing the calculation. In addition, in an image recording device that uses a full-color sensor as described above, it is possible to record colors other than black, and therefore, it is possible to record colors other than black, and therefore it is possible to record information that indicates what color the pixel is in addition to the density data of the pixel. Although a color flag is also required, it is omitted in FIG. 6 because it is not an essential matter.

Background level detection in the background level detection section 1 is performed as follows. The maximum background detection level MSL and the minimum background detection level LSL are supplied to the background level detection unit 1 from the control unit 3, and as shown in FIG. 7, the range of these two levels becomes the background density range. and now,
When density data as shown by 4 in FIG. 7 is input, the background level detection section 1 outputs the density data as is to the background removal section 2, and samples each predetermined pixel from the input density data. In FIG. 7, density data is imported at a rate of 1 in 4 pixels.
Pixels marked with "*" are captured. Then, it is determined whether the imported density data is within the background density range,
Only sample data within the background density range is averaged for a predetermined number of pixels, four sample pixels in FIG. 7, and the result is output to the background removal section 2 as a detected background level DL. As a result, the detected background level DL changes following the background density as shown in FIG.

The above series of operations is repeated for each main scanning line, and at the start of the next main scanning line, the last DL value of the previous line is cleared, and the detected background level DL is cleared.
, the initial removal level IL supplied from the control unit 3 is output. The background removal unit 2 uses an offset value OFSET supplied from the control unit 3 to the input detected background level DL.
is added to obtain the background removal level, this background removal level is compared with the input density data, and if the density data is equal to or higher than the background removal level, the input density data is output as is, and the value is lower than the background removal level. If so, the density value is output as zero. This allows only the density of the background portion to be removed without changing the density of the image portion.

[0005] Also, prescanning is performed on an A3 document length, a histogram is created for each predetermined density area, and a threshold value for background removal is determined from the maximum frequency of the histogram.
A histogram background removal method is also known in which a threshold value is determined based on the density area with the next highest frequency on the higher density side than the maximum frequency of the histogram.

[0006]

[Problems to be Solved by the Invention] According to the conventional real-time background removal method, the background removal level is obtained by sequentially monitoring the density level of the original, so even if the background density is not uniform, the background removal level can be adjusted accordingly. However, since the density of the actual image is also subject to determining the background level, the background removal level varies depending on the image content, resulting in unnatural background removal. there were.

[0007] In addition, in the case of the conventional histogram background removal method, the influence of the image content is reduced, but when the threshold value is determined from the maximum value of the histogram frequency, the problem is that the background part in the density area darker than the maximum value is not removed. There is. Furthermore, the same problem occurs even if the threshold value is determined from the area with the next highest frequency on the higher density side than the maximum value. Furthermore, in these methods, when calculating the maximum value of the histogram, it is necessary to hold the maximum value up to that point, compare it with that value, and change it to the new maximum value, which makes the processing complicated. There was a problem with speeding up.

The present invention solves the above-mentioned problems and provides a background removal method for an image recording apparatus that can effectively separate the background of a document from an image to obtain an optimal background removal level. The purpose is to

[0009]

[Means and Operations for Solving the Problems] The sampling unit 13 samples image data of a document read during prescanning at predetermined intervals in the main scanning direction. At the same time, gate signals in the main scanning direction and the sub-scanning direction are generated. The sampling data and gate signal are input to the histogram creation section 14, and a histogram for each predetermined density area is created. The background removal level determination unit 15 searches the frequency of this histogram from the high density side, detects the density area that first exceeds the gray value, and determines the corresponding background removal density threshold. The determined threshold value is given to the background removal unit 11 at the time of copy scanning. Skin removal section 11
outputs the density value as zero for pixels whose density is below the threshold value, outputs the input density value as is for pixels whose density exceeds a predetermined multiple of the threshold value, and outputs the input density value as is for pixels whose density exceeds the threshold value and the threshold value. For pixels that are a predetermined multiple or less, the input density value is multiplied by a predetermined value and output. As described above, an optimal constant threshold value is determined for each document, and background removal is performed based on the threshold value, so that beautiful background removal can be performed.

0010

Embodiments Hereinafter, embodiments will be described with reference to the drawings. FIG. 1 shows the configuration of an embodiment in which the background removal method of the image recording apparatus according to the present invention is applied to a so-called plus one color copying machine that records images in two colors: black and one color other than black. In the figure, 10 is a color separation unit (hereinafter referred to as CS).
), 11 is a background removal section, 12 is a background level detection section, 13 is a sampling section, 14 is a histogram creation section,
Reference numeral 15 indicates a background removal level determining section.

G, B, and R densities output from a full color sensor (not shown in FIG. 1), subjected to predetermined signal processing such as shading correction, and digitized into a predetermined number of bits, for example, 8 bits. Data is input to CS10. CS10 is based on the input concentrations of G, B, and R.
It is determined whether the pixel is a specified color (hereinafter referred to as a chromatic color) or any other color including black (hereinafter referred to as an achromatic color), and the density data and various flags, as shown in Figure 1, are determined. Main color flag (MCF) and sub color flag (SC)
F). Here, MCF and SCF are flags indicating whether the pixel is chromatic or achromatic. For example, MCF is an achromatic pixel and the density is a predetermined value, for example, the density range is 256 steps from absolute white to absolute black. 10 to 2 in tonality
When the value is around 0, it is set as "1", and when it is less than that, it is set as "0", and the SCF is a chromatic pixel and the density is a predetermined value, for example, in the above density range, it is set as "1" or above. It is set to 1 only when the density is zero, and it is set to ``0'' when the density is zero. When recording two colors, black and one other color, the pixel is either chromatic or achromatic, so one flag indicating whether the pixel is achromatic or chromatic is sufficient, but the above The reason for using two types of flags, MCF and SCF, is as follows. That is, even a portion that appears white to humans has a density value other than zero when read by a full-color sensor, although it is an extremely small value. Therefore, when a single flag is used to distinguish between an achromatic color and a chromatic color, the color will be output at that density, and will not be absolute white but will be output at a certain non-zero density. However, this is not desirable, and it is actually desirable that pixels having achromatic colors and small density values be output as absolute white. Therefore, a flag that distinguishes between achromatic colors and chromatic colors, in this case S
CF and a flag that distinguishes whether to output absolute white or the density obtained by reading for achromatic pixels,
In this case, two types of flags, MCF, are used. Therefore, in the final image output section, a pixel for which MCF=0 actually has a certain density, but is treated as absolute white, that is, a density of zero. The density data output from the CS 10 is supplied to the background removal section 11 and the background level detection section 12, and the two flags MCF and SCF are supplied to the background removal section 11.

Now, in the present invention, detection of the background level is performed at the time of pre-scanning. Therefore, in the configuration shown in FIG. 1, when the start button (not shown) is pressed, the imaging unit including the full color sensor operates, the document image is read, and the density data, MCF, and SCF are output from the CS 10. , the background level detection section 12 performs the operation of detecting the background level. This operation is as follows.

First, the sampling unit 13 samples the input density data at predetermined intervals in the main scanning direction (FS). Sampling in the FS direction does not involve capturing density data at predetermined intervals like normal sampling, but instead monitors the density data of a predetermined number of pixels and takes the minimum density value as the density in the sampling range. Sampling is performed by capturing the data. For example, if sampling is performed at intervals of 96 pixels, the sampling unit 13 monitors the first 96 pixels of the main scanning line, captures the minimum density among them, and then samples the next 96 pixels, that is, the first 96 pixels. The operation of monitoring the 97th pixel to the 192nd pixel and capturing the minimum density value among them is repeated for each line. In addition, in the sub-scanning direction (SS), gates are performed at a predetermined number of lines, for example, at intervals of 96 lines, and the FS
Sampling is performed by gating the histogram creation unit 14 using a gate signal obtained by validating the direction sampling signal.

Although the sampling interval in the SS direction can be set arbitrarily, the set sampling interval is fixed. Therefore, if the prescan speed is changed, the SS
The timing of the signal gating the sampling lines is adjusted so that the directional sampling occurs correctly at a predetermined line spacing, eg, 96 line spacing. Specifically, it is as follows. If prescanning is always performed with the document placed on the platen, the prescanning speed will always be constant, but if an automatic document feeder or duplex automatic document feeder is used, the unit speed will be constant. In order to increase the number of copies per unit of time, an imaging unit is operated when a document is pulled in to shorten the prescan time. At this time, the relative speed is greater than when prescanning is performed with the document placed on the platen, so the timing of the signal for gating the sampling line also differs. For example, as shown in FIG. 2, the frequency division ratio setting circuit 22 includes a frequency division ratio in the platen mode, a frequency division ratio when an automatic document feeder is used, and a frequency division ratio when a double-sided automatic document feeder is used. Frequency division ratios in various modes, such as frequency ratio, are set in advance, and the frequency division ratio according to the mode instructed by the control device (not shown) is set in the SS direction frequency divider 21. . As a result, the SS direction frequency divider 21 outputs the clock CLK,
For example, since the line synchronization signal pulse can be frequency-divided to a set frequency division ratio, sampling in the SS direction can be performed at predetermined line intervals. Note that the FS direction frequency divider 20 divides the video clock VCLK at a fixed frequency division ratio. The outputs of the FS direction frequency divider 20 and the SS direction frequency divider 21 are input to an AND circuit 23, and the output thereof is supplied to the histogram creation section 14 as a gate signal.

The density data and gate signal sampled by the sampling section 13 are supplied to a histogram creation section 14, where a histogram for each predetermined density area is created. An example of a histogram is shown in FIG. In Figure 3,
When the density gradation is set to 256 gradations from 0 to 255, nine density areas from 0 to 8 are set for the density range from 0 to 143. That is, in this case, the background level has a density value of 143 or less. The density range of each density area is as shown in FIG. 3B.

It is important to set density areas with overlapping density ranges in this way. This is because, for example, if only density areas 0, 2, 4, 6, and 8 in Figure 3 are set as density areas, if the density spans both density ranges with a slight difference in density, the same density Even in manuscripts, the shape of the histogram varies depending on the case, which may result in malfunctions in detecting the background level. For example, if there is a slight difference in density, even at the same position on the same document, the density may be recognized as 11 at one time, and 12 at another, and the histogram may change depending on the situation. The shape may vary. Therefore, in order to prevent malfunctions in such cases, density areas are set so that the density ranges overlap, as shown in FIG. According to the density areas in FIG. 3, for example, density 11 is also counted as density area 0, and density area 1
It is clear that the above-mentioned malfunctions can be avoided because the number is also counted. As described above, the histogram creation unit 14 counts up the density areas corresponding to the sampling data every time the gate signal is input. However, the concentration value of the sampling data is 144.
As described above, if it is more than that, it is ignored.

When the prescan is completed, the histogram creation is completed, and the background removal level determination unit 15 determines the background removal level based on the created histogram. Its operation is as follows. First, the background removal level determining section 1
5 searches the frequency of the histogram from the high density side, detects the density area that first exceeds a preset density check value, and determines the background threshold level (DEC) belonging to the detected density area (see Figure 3B). Further, a preset offset value (OFST) is added to the background threshold level (DEC) to obtain a background removal threshold level (THRD), which is output to the background removal section 11. In addition, if there is no frequency exceeding the concentration check value, the predetermined threshold level (OHPT)
For example, the level at which the background can be effectively removed even if the original is OHP paper is output as the background removal threshold level (THRD). In the above description, if the density check value is set to a frequency of about 100 cm 2 in terms of area, the optimum background removal level can be determined for a document of a commonly used size. In this case, if the original is 100cm2 or less, such as a business card, please use OHP.
T will be set, but in order to avoid this, the density check value may be changed as necessary.

The background removal threshold level (THRD) obtained at the time of pre-scanning as described above is given to the background removal section 11, and background removal is performed at the time of copy scanning. Its operation is as follows. The background removal unit 11 sets the input/output characteristics shown in FIG. 4 using the background removal threshold level THRD. That is, if the input image data (density) is less than or equal to THRD, the output image data is set to zero, if the input image data exceeds 1.5 times THRD, it is output as is, and if the input image data exceeds THRD and If it is less than 1.5 times the THRD, the difference between the input image data and the THRD is multiplied by 3 and output. Conventionally, image data below THRD was set to zero, and image data above THRD was output as is, but in this case, the density of the background near the outline of the image tends to increase, and this phenomenon is particularly complex. This is noticeable in the case of characters such as
The image was unnatural, but like the present invention,
By providing a predetermined slope between what is output as zero and what is output as input image data,
It has been confirmed that it is possible to eliminate the unnaturalness of output images that has conventionally occurred. Although this slope has been confirmed to be preferable through various experiments, it is clear to those skilled in the art that the slope can be set to any slope other than that shown in FIG.

The background removal section 11 performs the above-mentioned background removal operation and also corrects the values of MCF and SCF. That is,
As described above, when the density of an achromatic pixel is above a predetermined level, the MCF is set to "1", and when the density of a chromatic pixel is above a predetermined level, the SCF is set to "1". However, the density may become zero or less than the predetermined level due to the above-mentioned background removal. Therefore, when the density becomes less than a predetermined level due to background removal, the background removal unit 11 corrects the MCF to zero for achromatic colors, and corrects the SCF to zero for chromatic colors.

The above is the operation of the background removal section 11 during copy scanning, but during pre-scanning, THRD
A predetermined value, for example 80, is set as the input/output characteristic shown in FIG. 5. Therefore, during pre-scanning, background removal is performed according to the input/output characteristics shown in FIG. 5, and the image data from which the background has been removed is supplied to a means disposed after the background removal means, such as a document size detection section.

Although one embodiment of the present invention has been described above, it is clear to those skilled in the art that the present invention is not limited to the above embodiment, and that various modifications can be made.

[0022]

[Effects of the Invention] As is clear from the above description, according to the present invention, it is possible to determine the optimal background removal level for each document, and therefore it is possible to eliminate the unnaturalness of images that conventionally occurs. can.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of an embodiment in which a background removal method of an image recording apparatus according to the present invention is applied to a plus one color copying machine.

FIG. 2 is a diagram showing a configuration example of a sampling section of a background level detection section.

FIG. 3 is a diagram for explaining creation of a histogram.

FIG. 4 is a diagram illustrating an example of input/output characteristics set in a background removal section during copy scanning.

FIG. 5 is a diagram illustrating an example of input/output characteristics set in the background removal section during pre-scanning.

FIG. 6 is a diagram showing a configuration example of a conventional background removal method.

FIG. 7 is a diagram for explaining the operation of conventional background removal level detection.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10... Color separation section, 11... Background removal section, 12... Background level detection section, 13... Sampling section, 14... Histogram creation section, 14... Background removal level determination section.

Claims (2)

[Claims] Claim 1: A histogram of pixel density is created by sampling the image data of the document read during pre-scanning at predetermined intervals in the main scanning direction and the sub-scanning direction, and the frequency of the histogram is searched from the high density side to obtain a predetermined value. A background removal method for an image recording apparatus, characterized in that a background removal density threshold is determined based on a density value of a density area that first exceeds a value of . 2. A histogram of pixel density is created by sampling the image data of the document read during pre-scanning at predetermined intervals in the main scanning direction and the sub-scanning direction, and the frequency of the histogram is searched from the high density side to obtain a predetermined value. A threshold value for background removal density is determined based on the density value of the density area that first exceeds the value of For pixels exceeding a predetermined multiple of the threshold value, the input density value is output as is, and for pixels whose density exceeds the threshold value and is less than a predetermined multiple of the threshold value, the input density value is multiplied by a predetermined value and output. Background removal method for image recording devices.