JP2016116135A - Image processing apparatus, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent false recognition of thick lines in characters as pattern parts to reduce consumption of toner.SOLUTION: An image processing apparatus includes a down-sampling circuit 21 that converts input image data into low-resolution data having a lower resolution than the resolution of the input image data, and a character/pattern determination circuit 22 that determines attribute information on pixels in the low-resolution data. The character/pattern determination circuit 22 determines the attribute information on the pixels of the low-resolution data and reflects the determined attribute information on the input image data.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an image processing apparatus, method, and program, and more particularly, to toner consumption reduction.

  Conventionally, in the technical field of copying machines, it has been desired to express characters with high resolution and images with high gradation in the scanned image. Also desirable in terms. In order to realize this, it is effective to perform separate processing for the character part and the picture part, and a technique for distinguishing the character part and the picture part for that purpose is known.

  As a technique for discriminating whether a pixel of interest is a character portion or a picture portion, if the pixel of interest and a detected character edge are included in a predetermined mask, there is a technique for determining a pixel of interest as a character portion ( For example, Patent Document 1). In addition, there is a technique that counts the number of consecutive pixels that are determined to be black pixels, and determines that a point of interest is a character portion according to the count result (for example, Patent Document 2).

  However, even with the technique disclosed in Patent Document 1, when the line forming the character itself is thick like a thick character, the character edge and the pixel of interest in the character (such as the center of the line. May not be included in the mask, and the pixels in the character may be misrecognized as a picture part. However, there is a problem that the target pixel in the character is processed with four colors of CMYK, resulting in low image quality and a large amount of toner consumption.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to prevent erroneous recognition of the inside of a character as a picture portion and to reduce toner consumption.

  To achieve the above object, the present invention discriminates attribute information of each pixel in the low resolution data, low resolution conversion means for converting the input image data into low resolution data having a resolution lower than the resolution of the input image data. Attribute determination means for determining the attribute information of each pixel of the low-resolution data, and reflecting the determined attribute information in the input image data.

  According to the present invention, it is possible to prevent a character from being erroneously recognized as a picture part, and to reduce toner consumption.

1 is a block diagram illustrating a schematic configuration of an image forming apparatus 1 according to a first embodiment. It is a block diagram which shows the structure of the character and picture separation circuit 20 of FIG. FIG. 3 is a block diagram showing a configuration of a character / picture determination circuit 22 in FIG. 2. 6 is a diagram illustrating an example of an edge detection filter in Embodiment 1. FIG. It is a figure for demonstrating the effect of the downsampling in Embodiment 1. FIG. It is a figure which shows the example of the multistage control mask in Embodiment 2. FIG. It is a block diagram which shows the structure of the character and picture separation circuit 20 of Embodiment 3. FIG. FIG. 10 is a block diagram illustrating a configuration of a character / picture separation circuit 20 according to a fourth embodiment. FIG. 10 is a block diagram illustrating a configuration of a character / picture separation circuit 20 according to a fifth embodiment. It is a block diagram which shows the structure of the downsampling circuit 21 of Embodiment 6.

  The embodiment described below is characterized by performing resolution conversion on an input image and performing edge determination and black pixel determination on a reduced image in the process of determining the character part and the pattern part of the image. There is. Hereinafter, this feature will be described in detail with reference to the drawings.

<Embodiment 1>
FIG. 1 shows a schematic configuration of an image forming apparatus 1 according to the present embodiment. The image forming apparatus 1 can provide a function as a copying machine, and as illustrated, an image reading unit 10, a first image processing unit 11, a second image processing unit 12, and an image output unit. 13 and a character / picture separation circuit 20.

  The image reading unit 10 has a function of outputting a color image read by a scanner or the like. The character / picture separation circuit 20 has a function of separating a character area and a picture area from an image read by the image reading unit 10 and outputs the determination result as a control signal. The object on which the character / picture separation circuit 20 performs image processing is not limited to an image read using the image reading unit 10.

  The first image processing unit 11 performs image processing that is performed in parallel with the processing of the character / picture separation circuit 20 and performs processing that does not depend on the output result of the character / picture separation circuit 20. The second image processing unit 12 performs image processing, for example, filter processing, on the output image obtained by the first image processing unit 11 in accordance with a control signal output from the character / picture separation circuit 20.

  Image processing performed by the second image processing unit 12 in the present embodiment includes image processing that generates black and black for pixels that are determined to be “black characters” in the character / picture separation circuit 20. The character / picture separation circuit 20 outputs a control signal in which the density of the K signal is designated for each pixel of black characters. In response to the control signal, the second image processing unit 12 generates black ink on pixels determined to be black character pixels in the output image of the first image processing unit 11. By this image processing, a high-quality output image can be obtained in the image output processing in the image output unit 13 at the subsequent stage. At the same time, color toner is not wasted, so the toner consumption is reduced. Note that the same effect is obtained with respect to the color character by performing the same processing as that for the black character.

  The image output unit 13 forms an image based on the output image on which the image processing by the second image processing unit 12 has been performed, and outputs the formed image. In this embodiment, the specific mode of output is based on an electrophotographic system using toner.

  FIG. 2 is a diagram for explaining the configuration of the character / picture separation circuit 20 of FIG. As shown in FIG. 2, the character / picture separation circuit 20 includes a downsampling circuit 21, a character / picture determination circuit 22, and a multistage control signal generation circuit 23.

  The downsampling circuit 21 has a function of reducing the image size with respect to input image data. The image reduction rate of the downsampling circuit 21 can be switched according to the resolution of the image, for example, by preparing a plurality of threshold values.

  The character / picture determination circuit 22 has a function of determining attribute information (black character, color character, pattern, etc.) of each pixel and determining a character part and a picture part in the image. It has become.

  The edge detection circuit 221 in FIG. 3 applies an edge detection filter circuit to the low-resolution image obtained by the downsampling circuit 21, and uses the obtained signal as an edge amount. The edge detection circuit 221 has a function of dividing the edge amount into a plurality of values by a predetermined threshold. When only one threshold value is used, it is divided into two types of edge pixels and non-edge pixels. When a larger number of threshold values is used, the pixels are classified according to the value of the edge amount. As an example, a Laplacian filter shown in FIG. 4 is used for the edge detection filter circuit. Thereby, the response of the pixel position corresponding to the edge of the image is output high.

  The black pixel detection circuit 222 in FIG. 3 has a function of dividing the low resolution image obtained by the downsampling circuit 21 with a predetermined threshold. When only one threshold value is used, it is divided into two types of black pixels and non-black pixels. When a threshold value higher than that is used, the pixels are classified according to the degree of black.

  The input image data is, for example, data read from a scanner using a CCD element or the like, and is data of 8 bits for each pixel after the image is read and A / D converted. Further, although the input image data is assumed to be a color signal (RGB), a monochrome image or a G signal may be used as a representative.

  The character / picture determination circuit 22 having the above-described configuration uses a 5 × 5 mask, for example, and if there is at least one edge pixel in the mask and the target pixel is a black pixel, It has a function of judging pixels as characters. However, when the attribute of each pixel is determined without performing the processing by the downsampling circuit 21, an erroneous determination as described below may occur.

  FIG. 5 is a diagram in which Katakana “d” is reduced to 1/3 by downsampling. In the illustrated example, since the distance between the upper and lower character edges is 30 pixels before reduction, the character cannot be determined because the center of the character does not cover the character edge with a 5 × 5 mask. However, if the distance between the upper and lower character edges is 10 pixels by performing the downsampling, the character can be determined because the mask has a character edge even at the center of the character.

  The multi-stage control signal generation circuit 23 shown in FIG. 2 receives the result of the determination by the character / design determination circuit 22 for each pixel, and the control signal output from the determination result of black character, color character, or design. Present for each pixel. Since this control signal is generated from an image that has been down-sampled, a plurality of pixels are set to the same control signal for one pixel after down-sampling in order to cope with the original resolution. For example, considering that the image size is down-sampled to 50% main and sub, the control signal generated for one pixel after down-sampling is the one pixel in the image output by the first image processing unit 11. Are generated for four pixels corresponding to.

  In the case where the pattern is determined, the multistage control signal generation circuit 23 passes through and outputs it as it is. In other words, the control signal generation circuit 23 does not generate a control signal for a pixel after downsampling that has been determined to be a pattern by the character / picture determination circuit 22.

  On the other hand, for a pixel determined to be a black character, the multistage control signal generation circuit 23 changes the pixel data in multiple stages according to the attribute information of the pixels around the pixel of interest, which is “black character multistage control”. Perform the process. For color characters, the same processing as that performed for pixels determined to be black characters is performed.

  The black character multi-stage control will be described. The multistage control signal generation circuit 23 has a function of changing the ratio of the K signal of the target pixel in accordance with the number of pixels between the target pixel and the edge pixel. As an example, when a 5 × 5 mask is considered, when the target pixel is a black character and an edge pixel (the number of pixels between the target pixel and the edge pixel is 0), the ratio of the K signal is set to 100%, When the number of pixels between the target pixel and the edge pixel is 2 for black characters, processing is performed so that the ratio of the K signal is 60%.

  Note that the number of pixels between the target pixel and the edge pixel is the number of pixels (including the edge pixel itself) between the black character edge pixel closest to the target pixel.

  By performing the above-described black character multi-stage control, the black rate gradually decreases from the character edge toward the middle of the character, and even when the character determination result and the pattern determination result are greatly different, the visual discomfort is reduced. At the same time, toner consumption can be reduced.

  In addition, the present embodiment performs resolution conversion on the input image, and performs edge determination and black pixel determination on the reduced image, so that the distance between the edge of the character and the inside is close even for a thick character. Thus, since the character can be accurately determined as a character, according to the present embodiment, it is possible to reduce the amount of toner consumed as compared with the conventional technology. In addition, in order to prevent CMY signals from being mixed into black characters, it is possible to create a high-quality image as compared with the prior art.

<Embodiment 2>
Next, another means of the multistage control signal generation method in the multistage control signal generation circuit 23 will be described. The present embodiment is partially different from the first embodiment in the contents of black character multi-stage control performed on pixels determined to be black characters.

  In the present embodiment, a mask as shown in FIG. 6 is used in the black character multi-stage control. FIG. 6 uses a 5 × 5 size mask. In the first embodiment, the ratio of the K signal is controlled according to the number of pixels between the target pixel and the black character edge (2 in the case illustrated in FIG. 6). However, in the present embodiment, the black pixels in the mask are further added. Note the number of

  The black pixels existing in the mask are counted, and (in this case, 10) the ratio of the K signal is increased if the count value is large, and the ratio of the K signal is decreased if the count value is small. By controlling the ratio of the K signal according to the number of pixels between the target pixel and the edge pixel and controlling the ratio of the K signal according to the number of black pixels, the ratio of the K signal can be optimized. You will be able to adjust to.

  As an example, in the case where the number of pixels between the black character edge is 1 but the number of black pixels existing in the mask is 3, the distance from the black character edge is close, but the number of black pixels is small. It is considered that the probability that the pixel is a black pixel is low. In such a case, control is performed to reduce the ratio of the K signal.

  Further, when the edge detection circuit and the black pixel detection circuit 222 in FIG. 3 have a plurality of threshold values, the ratio of the K signal can be controlled more finely. As an example, if the black character edge amount is divided into five stages using four thresholds, the edge amount is 2 (the edge amount is small) and the edge even if the number of pixels between the target pixel and the edge pixel is the same. When the amount is 4 (the edge amount is large), the ratio of the K signal of the target pixel is increased when the edge amount is 4 than when the edge amount is 2.

<Embodiment 3>
In the present embodiment, the structure of the character / picture separation circuit 20 is as shown in FIG. As shown in the figure, the configuration is different from the configuration of the first embodiment shown in FIG. The downsampling circuit 21, the character / picture determination circuit 22 and the multistage control signal generation circuit 23 in the upper stage of FIG. 7 perform the same processing as in FIG.

  The lower part of FIG. 7 performs the same processing as the upper part without reducing the input image data. In the configuration of the present embodiment, in the case of a fine character, there is a possibility that the character disappears by down-sampling. Therefore, a circuit that does not perform down-sampling is added.

  In the present embodiment, separation determination processing is performed to separate input image data into processing that lowers the resolution (upper part in FIG. 7) and processing that does not lower the resolution (lower part in FIG. 7). By performing separation determination processing before and after downsampling, the multistage control signal generation circuit 23 performs expansion processing for thick characters, and the multistage control signal generation circuit 23a performs expansion processing for thin characters. .

  The selector 26 selects whether the target pixel is bold or thin. For example, when the multi-stage control signal generation circuit 23 has a high possibility that the ratio of the K signal is equal to or higher than the threshold, that is, a black character, the determination condition of the selector 26 is the output result of the multi-stage control signal generation circuit 23. Otherwise, the output result of the multistage control signal generation circuit 23a is output. That is, the selector 26 selects the output result having the highest K signal ratio for each pixel.

<Embodiment 4>
FIG. 8 shows the configuration of the character / picture separation circuit 20 of the present embodiment. As shown in the figure, this embodiment is obtained by replacing the multistage control signal generation circuit 23 a of FIG. 7 with an expansion filter circuit 27. In FIG. 7, the multi-level control is performed on the K signal for the fine character, and the process of adaptively switching the ratio of the K signal is performed. However, in the case of the fine character, a 5 × 5 mask is used. Then, control can be performed only up to a pixel located at a position about two pixels away from the character edge. Therefore, considering that there is almost no visual influence, it is not meaningful to perform multi-level control. Therefore, the circuit scale can be reduced by not performing multi-level control for fine characters.

<Embodiment 5>
FIG. 9 shows the configuration of the character / picture separation circuit 20 of this embodiment. As shown in the figure, this embodiment has a configuration in which the number of times of downsampling is increased in the circuit of FIG.

  In the description so far, downsampling has been performed only once. However, there is a problem that the character size cannot be applied to various manuscripts. Since the size of characters existing in one image is not always the same, for example, when downsampling is performed at a certain reduction rate, even if it is effective for a certain character size, If there is a larger character in this area, one reduction ratio may not be sufficient.

  Therefore, the above problem can be solved by preparing a plurality of downsampling circuits 21 for the input image. FIG. 9 shows an example in which two downsampling circuits 21 are used. For example, the downsampling circuit 21 downsamples the input image to ¼, and the downsampling circuit 21b downsamples the input image to ½. In the selector circuit, the one having the highest K signal ratio among the outputs of the multistage control signal generation circuit 23, the multistage control signal generation circuit 23b, and the expansion filter circuit 27 is selected.

Although two downsampling circuits 21 are used in FIG. 9, naturally, a larger number of downsampling circuits 21 may be used.
By performing the above processing, various character sizes can be handled.

<Embodiment 6>
The present embodiment is characterized by the configuration of the downsampling circuit 21. FIG. 10 shows a configuration of the downsampling circuit 21 of the present embodiment.
The description so far has been made on the assumption that downsampling in the main scanning and sub-scanning directions is performed at the same ratio. However, when the resolutions in the main scanning and sub-scanning directions are different, the same resolution after downsampling is easier in performing processing. This is because there is a reason that it is not necessary to change the size of the filter.

  Therefore, in the present embodiment, the main scanning direction down-sampling 211 and the sub-scanning direction down-sampling 212 are provided, and down-sampling is performed independently in the main scanning and sub-scanning directions. For example, when the main scanning direction resolution is 600 dpi and the sub-scanning direction resolution is 300 dpi, the main scanning direction down-sampling 211 sets the main scanning direction to 1/6, and the sub-scanning direction down-sampling 212 sets the sub-scanning direction. 1/3 (100 dpi).

  Further, assuming that the performance of edge separation is very high, it can be considered that a character is always surrounded by edge pixels. Therefore, even if the resolution reduction by the downsampling circuit 21 is performed only in one direction (for example, the main scanning direction), the determination in the character can be sufficiently performed.

  In this way, the circuit scale can be reduced by omitting down-sampling in the sub-scanning (main scanning) direction according to the edge separation performance.

20 Character / Picture Separation Circuit 21 Down-sampling Circuit 211 Down-sampling in the Main Scanning Direction 212 Sub-Scanning Down-Sampling 22 Character / Picture Determination Circuit 221 Edge Detection Circuit 222 Black Pixel Detection Circuit 23 Multi-stage Control Signal Generation Circuit 26 Selector 27 Expansion Filter Circuit

JP 2002-199212 A JP 2005-033482 A

Claims (10)

Low resolution conversion means for converting input image data into low resolution data having a resolution lower than the resolution of the input image data;
Attribute determining means for determining attribute information of each pixel in the low resolution data;
With
The image processing apparatus, wherein the attribute determining unit determines attribute information in each pixel of the low resolution data, and reflects the determined attribute information in the input image data. Black pixel determination means for determining whether the pixel of interest is a black pixel or a non-black pixel, and determining color information of each pixel of the low resolution data;
Edge detection means for detecting edge pixels of the low-resolution data,
The attribute determination unit is configured to determine the attribute of each pixel in the low resolution data based on the color information determined by the black pixel determination unit and the positional relationship between the edge pixel and the target pixel detected by the edge detection unit. The image processing apparatus according to claim 1, wherein information is discriminated. The attribute determination unit reflects the attribute information of each pixel of the low-resolution data corresponding to each pixel of the input image data to each pixel of the input image data. The image processing apparatus described.   The multi-stage control means for changing the color information of the target pixel according to the number of pixels between the target pixel and the edge pixel, according to any one of claims 1 to 3, Image processing device.   The multi-stage control means changes color information of a pixel of interest according to the number of pixels determined as black pixels by the black pixel determination means included in a specific pixel range. 4. The image processing apparatus according to 4.   The image processing apparatus according to claim 4, comprising a plurality of sets of the low resolution conversion unit, the attribute determination unit, and the multi-stage control unit, which have different resolutions used for resolution reduction.   The image processing apparatus according to claim 6, further comprising a selector that selects color information having the darkest density among color information of a target pixel output from the other-stage control unit of each set.   8. The image processing according to claim 1, wherein the low-resolution conversion unit reduces the resolution of the input image data at a different ratio in the main scanning direction and the sub-scanning direction. 9. apparatus. A low resolution conversion step of converting the input image data into low resolution data having a resolution lower than the resolution of the input image data;
An attribute determination step of determining attribute information of each pixel in the low resolution data;
Have
In the attribute determining step, attribute information in each pixel of the low resolution data is determined, and the determined attribute information is reflected in the input image data. In the image processing device,
Low resolution conversion processing for converting the input image data into low resolution data having a resolution lower than the resolution of the input image data;
Attribute determination processing for determining attribute information of each pixel in the low-resolution data;
And execute
In the attribute discrimination process, an attribute information in each pixel of the low resolution data is discriminated, and the discriminated attribute information is reflected in the input image data.