JP2006050548A - Image processing device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make a software process to be performed at a higher speed by simplifying an operation process for acquiring a color to be daubed when producing a base image for making a color image to be multilayered, and to make a hardware process to be performed at a low cost. <P>SOLUTION: A main image such as characters etc. to be drawn in relatively low brightness within an original image is to be extracted from the original image, and the base image is to be produced by daubing a portion within the original image corresponding to an image line portion of the main image by a desired color to be daubed on the basis of image data of its neighboring pixels. At this time, the color to be daubed is to be determined in accordance with a brightness component value determined on the basis of the brightness component value of the neighboring pixel and a color difference component value which is previously determined. Especially, the color difference component value when determining the color to be daubed is previously set to an intermediate value, and the color to be daubed is made to be colorless. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus and an image processing method for generating a so-called multi-layer image in which a color image is divided into a plurality of layers, and more particularly to a main image such as characters and a background image as a background thereof. The present invention relates to an image processing apparatus and an image processing method that perform compression processing using different compression methods.

  In order to digitize documents, a method of reading a document with a scanner and converting it to image data has become widespread, and when distributing a document created with document creation software on a computer, it can be viewed without any problems at the distribution destination. It is desirable to convert the document into image data, and in order to convert such a document into image data, various compression methods for reducing the amount of data due to demands such as saving storage capacity and shortening communication time are available. It's being used.

However, in the case of a color document, the reproduction quality may be degraded by color compression processing due to the difference in brightness between the character that is usually drawn at low brightness and its surroundings. There is known a technique of performing compression processing using different compression methods separately for the background image and the background image. (See Patent Document 1). In this technique, a character image is extracted by binarizing a color image, and a base image is created by filling the character image portion with the same color as the periphery of the character image.
JP 2003-244447 A

  However, in the conventional technique, when obtaining the fill color when creating the background image, the values of the pixels serving as the background around the character image are obtained by averaging, and R, G, B Since the average value is calculated for each component value, the calculation process becomes complicated. When software processing is performed, a long time is required for processing, and when hardware processing is performed, the circuit configuration becomes complicated and the manufacturing cost is increased. There is a difficult point.

  The present invention has been devised to solve such problems of the prior art, and its main purpose is to obtain a fill color when generating a background image for multi-layering a color image. The present invention provides an image processing apparatus and an image processing method that are configured to simplify the calculation process for the software, to increase the speed in software processing, and to reduce the cost in hardware processing is there. Furthermore, the present invention provides an appropriate filling process even when a main image such as a character or a line drawing figure is drawn with higher brightness than a background image (background image) in a color original image, such as a white character. It is also possible to prevent the deterioration in the compression efficiency and the deterioration of the image quality of the background image caused by the outline portion remaining in the background image so as to surround the main image portion. It is said.

  In order to solve such a problem, in the image processing apparatus according to the present invention, as shown in claim 1, main image generating means for extracting from the original image a main image drawn with a luminance different from that of the background in the original image; , A background image generating means for generating a background image by filling a portion in the original image corresponding to the main image with a required fill color based on image data of the surrounding pixels, and the main image and the background image obtained thereby Image compression means for compressing each image data by different compression methods, and the base image generation means is configured to determine a luminance component value determined based on luminance component values of peripheral pixels and a preset color. The filling color is determined based on the degree component value.

  According to this, since it is only necessary to calculate the luminance component value to obtain the fill color, the calculation process is greatly simplified, so that the processing time is shortened in software processing and the hardware processing is performed. Thus, the circuit configuration can be simplified.

  In this case, the portion of the original image corresponding to the main image is filled with a different fill color from the surrounding portion. However, since the fill color has a small difference in luminance from the peripheral portion, the reproduction image quality is improved by the compression process. There is no significant drop.

  When the main image is drawn in a plurality of colors, the main image may be sorted for each display color and stored in a file by dividing into a plurality of layers for each display color. Further, the file format for storing the main image such as characters and the background image divided into a plurality of layers is not particularly limited, but PDF, TIFF, XML, and the like are preferable as highly versatile. Further, the main image generation means can extract the main image drawn with relatively low luminance from the original image by binarizing the luminance component value with a predetermined threshold value.

  In the image processing apparatus, as described in claim 2, the main image may be at least one of a character, a ruled line, and a line drawing figure. According to this, since characters, ruled lines, and lines and points constituting a line drawing figure are drawn with relatively low luminance, the reproduction image quality can be improved by performing compression processing separately from the base image.

  In the image processing apparatus, as described in claim 3, the chromaticity component value used when determining the fill color in the background image generation means is preset to an intermediate value, and the fill color is an achromatic color. Can do. According to this, by setting the chromaticity component value of the filled color as an intermediate value, when the luminance component value is determined based on the luminance component value of the surrounding pixels, the background portion that is often white and the filled color Will be the same white color for optimal results.

  In the image processing apparatus, as described in claim 4, the image compressing unit compresses the background image by JPEG, and the sampling ratio Y: Cb of the luminance component value and the chromaticity component value at the time of the JPEG compression. : Cr can be any of 4: 1 :: 1, 4: 2: 2, and 4: 2: 0.

  In the image processing apparatus, as described in claim 5, the painting process by the base image generation unit can be performed in a predetermined block unit which is a processing unit in the JPEG compression process. According to this, the filling process and the JPEG compression process can be handled in the same processing unit.

  In the image processing apparatus according to the present invention, as described in claim 6, main image generation means for extracting from the original image a main image drawn in the original image with a luminance different from the background, and the original image corresponding to the main image A background image generating means for generating a background image by filling the inside portion with a required fill color based on the image data of the surrounding pixels, and the image data of the main image and the background image obtained thereby are compressed into different compression methods. Image compression means for performing compression processing, and when the base image generation means is a normal image in which the main image has a lower luminance than the surroundings, the luminance component value determined based on the luminance component values of the surrounding pixels; The chromaticity component value determined on the basis of the chromaticity component values of the surrounding pixels in the case where the main image is an inverted image having a higher brightness than the surroundings, by determining the fill color based on a preset chromaticity component value And pre- And it shall determine the fill color by the set luminance component value.

  According to this, even when the main image is an inverted image surrounded by a chromatic color with low luminance, an appropriate fill color can be acquired to improve the image quality.

  In the reverse image, the luminance difference between the main image portion and the surrounding background portion increases so that it can be clearly seen. However, as with normal images, the luminance component value based on the luminance component value of the surrounding pixels is set in advance. When the fill color is determined based on the chromaticity component values, the difference in chromaticity component values between the main image portion and the surrounding background portion becomes large, and the main image portion is subjected to compression processing (for example, JPEG). In some cases, the filled color (for example, gray) oozes out of the main image portion and degrades the image quality. On the other hand, when configured as in the present invention, the image quality can be prevented from degrading.

  In the image processing apparatus according to the present invention, as described in claim 7, main image generation means for extracting from the original image a main image drawn in the original image with a luminance different from the background, and the original image corresponding to the main image Underground image generation means for generating the background image by filling the inside part with the required main image fill color based on the image data of the surrounding pixels, and the respective image data of the main image and the background image obtained thereby are compressed differently An image compression means for compressing the image by the method, wherein the background image generation means obtains a contour portion existing in the original image so as to surround the main image portion from the pixel values of the outer region. The process of filling with the contour fill color was performed.

  According to this, since the background image from which the outline portion is removed can be obtained, a reduction in compression efficiency and a deterioration in image quality when the background image is compressed can be avoided, and a high speed can be achieved by software processing. It is possible to simplify the circuit configuration in the wear processing.

  In the image processing device, as described in claim 8, the background image generation means can extract a contour fill color for each pixel by a window region having a predetermined number of pixels centering on the target pixel. . According to this, since the contour fill color is determined for each pixel, the memory capacity required for processing can be reduced. Here, it is not necessary to hold the image data of the entire page, and it is sufficient to hold the image data of the number of lines corresponding to the size (number of pixels) of the window area, for example.

  In this case, when there are pixels (for example, black pixels on the binary image) that constitute the main image portion in the window area, it is determined that the target pixel constitutes the contour portion, and the target pixel is filled. By doing so, pixels from the boundary pixel of the main image to a predetermined distance can be filled as a contour portion.

  In particular, the contour portion extends in a band shape so as to surround the outer periphery of the main image portion, and the width of the contour portion varies depending on the characteristics of the image reading device and the like, so the size of the window depends on the state of the original image. The size (number of pixels) may be changed, so that an appropriate filling process according to the width of the contour portion can be performed.

  The extraction of the outline fill color may be performed based on the value of the pixel in the window region. At this time, the pixels recognized as the background portion in the window region, that is, the constituent pixels of the main image portion and the contour portion are extracted. The contour fill color may be determined based on the values of peripheral pixels that are regarded as not corresponding to any of the constituent pixels.

  At this time, when there are a plurality of pixels that can be recognized as the background portion in the window region, it is preferable to average the values of those pixels to obtain a contour fill color, which makes it more preferable than a case based on a single pixel value. An appropriate contour fill color can be obtained. Further, when there is no pixel that can be recognized as a background part in the window area, it is preferable to adopt a configuration in which the contour fill color obtained by the process for the pixel immediately before the target pixel is adopted. If the pixel fill color value is compared with the current value of the target pixel, and if there is a color difference that exceeds a predetermined value, the contour fill color of the previous pixel is not adopted and the fill process is not performed. It is good to have a configuration, and it is possible to avoid a problem that the background color changes so that it is filled with an incorrect color.

  In the image processing method according to the present invention, the main image drawn in the original image with a luminance different from the background is extracted from the original image, and a portion in the original image corresponding to the main image is provided. Generating a background image by painting with a required fill color based on the image data of the surrounding pixels, and compressing each image data of the main image and the background image obtained thereby by different compression methods; In the step of generating the background image, the fill color is determined based on a luminance component value determined based on luminance component values of peripheral pixels and a preset chromaticity component value.

  According to this, since only the luminance component value needs to be calculated in order to obtain the fill color, the calculation process is greatly simplified, so that the processing time is shortened in software processing and the hardware processing. Thus, the circuit configuration can be simplified.

  In the image processing method according to the present invention, as described in claim 10, a step of extracting from the original image a main image drawn with a luminance different from the background in the original image, and a portion in the original image corresponding to the main image Generating a background image by painting with a required fill color based on the image data of the surrounding pixels, and compressing each image data of the main image and the background image obtained thereby by different compression methods; In the step of generating the background image, when the main image is a normal image having a lower luminance than the surroundings, the luminance component value determined based on the luminance component values of the surrounding pixels and a preset color The fill color is determined based on the chromaticity component value, and when the main image is an inverted image having higher brightness than the surroundings, the chromaticity component value determined based on the chromaticity component values of the surrounding pixels, And it shall determine the fill color by the set luminance component value.

  According to this, even when the main image is an inverted image surrounded by a chromatic color with low luminance, an appropriate fill color can be acquired to improve the image quality.

  In the image processing method according to the present invention, as described in claim 11, a step of taking out from the original image a main image drawn with a luminance different from the background in the original image, and a portion in the original image corresponding to the main image And generating a background image by painting with the required main image fill color based on the image data of the surrounding pixels, and compressing each image data of the main image and the background image obtained by different compression methods A step of generating the background image, wherein a contour portion existing in the original image so as to surround the main image portion is painted with a contour fill color obtained from the pixel values of the outer region. The crushing process was performed.

  According to this, since it is possible to obtain a background image from which the outline portion has been removed, it is possible to avoid a decrease in compression efficiency and a deterioration in image quality when the background image is compressed, and to increase the speed in software processing. It is possible to simplify the circuit configuration in the wear processing.

  In the present invention, a color system (color space) that displays a color with a luminance component value indicating brightness and a chromaticity component value indicating hue is used. For example, in the YCbCr color system, the luminance component value is Y, the chromaticity (color difference) component value is Cb · Cr.

  As described above, according to the present invention, when generating a background image for multi-layering a color image, it is possible to acquire a fill color only by calculation related to a luminance component value, and a calculation process for acquiring a fill color In software processing, the processing time can be shortened and the speed can be increased. In hardware processing, the circuit configuration can be simplified to reduce the manufacturing cost. A great effect can be obtained in the reduction. In addition, when the main image is an inverted image surrounded by a low-brightness chromatic color, the fill color is determined by the chromaticity component value based on the chromaticity component value of the surrounding pixel and the luminance component value as a fixed value. Thus, an appropriate fill color can be acquired, and a remarkable effect can be obtained in improving the image quality. Furthermore, since the background image from which the contour portion has been removed can be obtained by painting the contour portion around the main image portion with the contour fill color obtained from the pixel values of the outer region, the background image can be obtained. It is possible to avoid a reduction in compression efficiency and a deterioration in image quality during the compression process, increase the speed in software processing, and simplify the circuit configuration in hardware processing.

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

  FIG. 1 is a block diagram showing a schematic configuration of a multifunction machine to which an image processing apparatus according to the present invention is applied. This multi-function machine includes a CCD 1 that reads an image of a document, a read data processing unit 2 that processes image data from the CCD 1 so as to obtain an appropriate read result, and a read data processing unit that obtains an appropriate print result. 2 has a print data processing unit 3 that processes image data from 2, and a printer engine 4 that records an image on recording paper based on the image data from the print data processing unit 3, and copies the original image onto the recording paper. It has a copy function.

  In addition, this multifunction device has a scanner function for reading an image of a document and creating an output image file, and image data obtained through the CCD 1 and the read data processing unit 2 is sent to the multi-layer image generation unit 5. In this case, the read image data is separated into a base image and a character image, compressed by a different compression method, and stored in a PDF (portable document format) file format (multi-layer image file). ) Is created.

  Furthermore, this multifunction device has a data conversion function for creating a multilayer image file from print data of a document created by document creation software on a computer such as a PC, and print data in a page description language such as a postscript. Is received by the network control unit 6, and the print data is converted into RBG image data by the data conversion unit 7, and then sent to the multi-layer image generation unit 5 to create a multilayer image file.

  The read data processing unit 2 includes an A / D / offset adjustment unit 11 that performs A / D conversion and offset adjustment of the output signal of the CCD 1, a shading / black correction unit 12 that corrects sensitivity unevenness of the CCD 1, and a line sensor. The line-to-line correction unit 13 corrects a shift in the sub-scanning direction caused by the CCD 1 and the gamma correction unit 14 performs gamma correction according to the characteristics of the CCD 1.

  The print data processing unit 3 includes a color conversion unit 16 that performs color space conversion of RGB image data to CMYK image data, a scaling unit 17 that expands and contracts to an image size corresponding to the output resolution of the printer engine 4, and a printer. A printer gradation correction unit 18 that corrects gradation so as to conform to the output characteristics of the engine 4 and characters in order to appropriately perform the processing of the color conversion unit 16 and the printer gradation correction unit 18 according to the characteristics of the image. And an image area determination unit 19 for detecting an area such as a photograph.

  FIG. 2 is a block diagram illustrating a configuration of the multi-layer image generation unit illustrated in FIG. The multi-layer image generation unit 5 extracts a main image such as a character or a line drawing figure drawn from the original image with a lower luminance in the original color image than the background image (background image). Generation means) 21, a background image generation section (background image generation means) 22 for generating a background image by painting a portion of the original image corresponding to the image line portion of the main image with a required fill color, and The main image compression unit (image compression unit) 23 and the background image compression unit (image compression unit) 24 that compress the main image and the background image using different compression methods, and the main image and the background image obtained thereby Each of the compressed image data is divided into a plurality of layers to be filed.

  The main image generation unit 21 includes a luminance generation unit 31 that calculates luminance for each pixel from RGB image data, and a binarization process that binarizes the luminance image obtained thereby based on a predetermined threshold value. An image for extracting a main image such as a character or a graphic based on the area information obtained by the area determination unit 33 for determining an area such as a character or a graphic in the binary image obtained thereby; An extraction unit 34, a main image color extraction unit 35 that extracts a display color at the time of reproduction of the main image obtained thereby from the original image, and stores the main image in a file in a separate layer for each display color obtained thereby In order to do this, it has a color-by-color layer dividing unit 36 that creates data of main images by color.

  The area determination unit 33 determines each area of the character in the binary image, the table separated by the ruled line, the figure drawn by the line drawing, and the bitmap image (photograph, etc.), and the binary of each area obtained here Based on the area information regarding the position (coordinates) on the image, the image extracting unit 34 extracts characters, ruled lines, and line drawing figures as main images. In the character area, a process of cutting out an image of one character unit through line recognition is performed. In the table area, the table structure may be recognized and a ruled line image and a cell unit image may be cut out.

  The main image compression unit 23 compresses main image data such as characters and line drawing figures by a compression method suitable for binary image compression, for example, MMR or JBIG. In the background image compression unit 24, the data of the background image is compressed by a compression method with excellent multi-value image compression efficiency, for example, JPEG. The compressed image data of the main image and the background image obtained in this manner is filed as a separate layer in the file generation unit 25, so that when the original image is directly JPEG compressed, the luminance is lower than that of the surrounding background portion. It is possible to avoid the occurrence of mosquito noise in the surrounding background portion due to the main images such as characters and line drawings drawn on the screen.

  In assigning the compressed data of the main image to the color-specific layers, a predetermined number (for example, eight) of color-specific layers is set in advance, and the main image color extracting unit 35 determines the number of colors corresponding to the number of layers. Therefore, a color reduction process for unifying the approximate colors is performed. The main image color extraction unit 35 generates color palette information indicating the correspondence between the display color of the main image (binary image) separated by color and the identification number of each color layer. The file generation unit 25 stores the multi-layer image file.

  The base image generation unit 22 represents a portion of the binary main image obtained by the image extraction unit 34 of the main image generation unit 21, that is, a portion in the original image corresponding to a pixel that is a black pixel on the binary image. A background image is generated by painting with a paint color similar to the surrounding color, and a paint color extracting unit 37 that extracts a paint color for painting from the original image and an image extracting unit 34 of the main image generating unit 21. A paint processing unit 38 is provided for painting a portion in the original image corresponding to the image line part of the obtained main image with the paint color obtained by the paint color extracting unit 37.

  The fill color extraction unit 37 determines the fill color with reference to the pixel values of the part in the original image corresponding to the non-image line part around the image line part of the main image obtained by the image extraction unit 34. In particular, here, after the RGB image data is converted to YCbCr image data, the luminance component value Y of the fill color is determined based on the luminance component value of the pixel in the portion of the original image corresponding to the non-image portion. On the other hand, the color difference component values Cb · Cr of the fill color are preset to Cb = Cr = 128 when the intermediate value, for example, the color difference component value is expressed in 256 levels, and the fill color is an achromatic color.

  In the background image compression unit (image compression means) 24, the data of the background image is JPEG-compressed, and in this JPEG compression, downsampling for greatly compressing the color difference component value with respect to the luminance component value is performed. The sampling ratio Y: Cb: Cr of the luminance component value Y and the color difference component values Cb · Cr may be any one of 4: 1: 1, 4: 2: 2, 4: 2: 0.

  FIG. 3 is a conceptual diagram showing a processing procedure in the multilayer image generation unit shown in FIG. A binary image obtained by binarizing the original image is obtained through the luminance generation unit 31 and the binarization processing unit 32 of the main image generation unit 21, and further, the characters, tables, and figures are obtained through the region determination unit 33 and the image extraction unit 34. The image data of the main image for each area type is obtained, and the image of the character area is segmented in units of one character by a rectangle circumscribing the character in the binary image to obtain a character image for each character.

  Extraction of the color of the main image performed by the main image color extraction unit 35 of the main image generation unit 21, and extraction of the fill color and fill processing performed by the fill color extraction unit 37 and the fill processing unit 38 of the base image generation unit 22. Is performed for each block of 8 × 8 pixels, which is a processing unit in JPEG compression processing.

  First, in the main image color extracting unit 35 of the main image generating unit 21, the color of the pixel having the same coordinate as that of the pixel constituting the image line unit which is a black pixel on the binary image is acquired from the data of the multi-valued image (original image). The average value of each pixel value on the multi-valued image is calculated, that is, the RGB values of each pixel are averaged to calculate the average value for each RGB, and the image line portion of the main image is displayed. A color (eg, red) is determined.

  In addition, the fill color extraction unit 37 of the base image generation unit 22 has luminance component values of pixels having the same coordinates as the pixels constituting the non-image part that is a white pixel around the image line part that is a black pixel on the binary image. Is obtained from the data of the multivalued image subjected to the YCbCr color system conversion, the average value of the luminance component values for each pixel on the multivalued image is calculated, and the average luminance component value and the intermediate value (128) are calculated. The fill color is determined based on the color difference component value set in advance. For example, if the portion in the multi-valued image corresponding to the non-image area of the binary image is blue, the portion corresponding to the image area of the binary image is surrounded by the background image obtained by the fill processing unit 38. The color scheme is filled with gray with the same brightness as the blue color.

  When acquiring the display color of the main image, the display color is determined for a plurality of pixels selected according to an appropriate rule, not for all the pixels constituting the image line portion in the processing block. It is good to do. Similarly, when obtaining the fill color of the background image, the fill color may be determined for a plurality of pixels selected according to an appropriate rule from the pixels constituting the non-image portion. In addition, pixels located on the outer periphery of the character image for each character can be used as a reference for determining the fill color of the base image.

  Although the background image is created as described above, since the color difference component value of the fill color is fixed here, the difference in the color difference component value between the fill pixel and its surrounding pixels is increased in some cases, Although it may take time to calculate the background image compression, when downsampling is performed in which the color difference component value is greatly compressed with respect to the luminance component value as in JPEG compression, the color difference component value is compared with the luminance component value. Since the calculation load is relatively reduced, the overall calculation load including the acquisition of the fill color is reduced.

  4 and 5 are flowcharts showing a processing procedure in the multilayer image generation unit shown in FIG. Here, an example in which characters are used as a main image is shown.

  First, image data of an RGB original image stored in the memory is acquired in the main image generation unit 21 (step 101), and then image data based on YCbCr is generated by color system (color space) conversion processing (step 102). After the luminance image by the luminance component value Y is created by the luminance generation unit 31, the binarization processing unit 32 performs a process of binarizing the luminance image (step 103). A layout analysis process for detecting areas such as characters and graphics in the binary image is performed by the area determination unit 33, and attributes relating to the area type and the position (coordinates) of the area on the original image are set for each area. Is created (step 104).

  Next, the process proceeds to a process of acquiring an image of one character unit (steps 105 to 108). Here, a process of cutting out a character image for each character from the binary image in the character region is first performed by the image extraction unit 34 (step 106) Next, a process of adding the storage address of the character image for each character cut out to the area list is performed (step 107).

  Next, the display color of the character image is acquired for each character, and the process proceeds to the step of compressing the image by assigning it to the character layer by color (steps 109 to 116). Here, the character image belonging to the character region is first referred to the region list. The processing for obtaining the coordinates for each processing block is performed, and a character list describing the information regarding the processing block thus obtained is created (step 110). Then, the process of acquiring the display color of the character image in units of processing blocks described in the character list is performed by the main image color extraction unit 35 (step 112), and then the process of sorting the image in units of processing blocks for each display color is performed. This is performed by the color-specific layer dividing unit 36 (step 113). Further, when the display color acquisition and sorting process for all characters belonging to the target character area is completed (step 114), the main image compression unit 23 performs a process of compressing the binary image for each display color (step 115). ).

  Next, as shown in FIG. 5, the process proceeds to a process (steps 117 to 123) of obtaining and painting the fill color of the background image in units of processing blocks. Here, first, information on the processing block for each character is obtained as described above. After the described character list is created (step 118), a process of acquiring a fill color for each processing block described in the character list is performed by the fill color extracting unit 37 (step 120), and then the acquired fill color is used. The painting process is performed by the painting processing unit 38 in units of processing blocks (step 121).

  When the filling process for one page is completed (step 123), the background image compression unit 24 performs the sub-sampling of the filled multi-value image and JPEG compression (steps 124 and 125).

  When the compressed data of the character image and the background image is acquired in this way, the process proceeds to the process of creating the multilayer image file, and the process of adding the header information to each compressed data of the character image and the background image and writing to the file is performed. This is performed by the generation unit 25 (steps 126 and 127).

  FIG. 6 is a block diagram illustrating another example of the multi-layer image generation unit illustrated in FIG. Here, as in the above example, the main image generation unit 21 performs processing for determining regions such as characters and graphics in the binary image obtained through the luminance generation unit 31 and the binarization processing unit 32. A process of extracting a main image such as a character or a figure based on the area information obtained here is performed by the determining unit 33, and the image extracting unit 34 performs the process. In contrast, the image extraction unit 34 discriminates between a normal character in which the character that is the main image has a lower luminance than the surroundings and a white character (inverted character) in which the characters that are the main image have a higher luminance than the surroundings. The discrimination result (outlined character information) is sent to the fill processing unit 38 of the base image generation unit 22.

  Here, a normal character is composed of black pixels on a binary image and is surrounded by white pixels, whereas a white character (reverse character) is a white pixel on a binary image. Configured, and the periphery thereof is covered with black pixels. Note that here, for characters, the normal image whose main image has lower brightness than the surroundings and the reverse image whose main image has higher brightness than the surroundings are identified and processed differently, but for ruled lines and line drawing figures The same processing is possible.

  FIG. 7 is a block diagram showing in detail the configuration of the background image generation unit shown in FIG. In the fill color extraction unit 37 of the base image generation unit 22, after the RGB image data is converted into Y, Cb, and Cr image data, the original image corresponding to the surrounding pixels of the main image obtained by the image extraction unit 34. The fill color is determined with reference to the pixel value of the inner part. In particular, here, the luminance component value Y and the color difference component value Cb are based on the pixel value of the part in the original image corresponding to the surrounding pixels of the main image. -Each value of Cr is obtained.

  Based on the outline character information from the image extraction unit 34, the fill processing unit 38 switches the image signal from the fill color extraction unit 37 to the signal switching units 42 a, 42 b, and Y for each image signal of Y, Cb, and Cr. 42c is provided, and when the outline character information indicates a normal character (0), the luminance component value Y obtained by the fill color extraction unit 37 is sent to the main image partial fill processing unit 41 as it is. The color difference component values Cb and Cr are sent to the main image partial filling processing unit 41 as fixed values. On the other hand, when the white character information indicates a white character (1), a fixed value of the luminance component value Y is sent to the main image partial fill processing unit 41, and the color difference component value Cb / Cr is a solid color extraction. What is obtained in the section 37 is sent to the main image partial filling processing section 41.

  The fixed value for determining the color difference component value Cb · Cr in the case of a normal character is set to Cb = Cr = 128 when the intermediate value, for example, the color difference component value is expressed in 256 levels, as in the above example. The fill color is achromatic. On the other hand, the fixed value for determining the luminance component value Y in the case of a white character (inverted character) is set to a value smaller than the intermediate value, for example, Y = 100 when the luminance component value is expressed in 256 levels. This is due to the fact that the surrounding area is usually drawn at a relatively low luminance so that the character can be clearly seen in the case of white characters. Can be acquired.

  In the background image compression unit (image compression means) 24, the background image data is JPEG compressed as described above, and in this JPEG compression, downsampling is performed to greatly compress the color difference component value with respect to the luminance component value. The sampling ratio Y: Cb: Cr of the luminance component value Y and the color difference component values Cb · Cr may be any one of 4: 1: 1, 4: 2: 2, 4: 2: 0.

  FIG. 8 is a conceptual diagram showing a processing procedure in the base image generation unit 22 shown in FIG. In the case of a white character (inverted character), processing for inverting white pixels and black pixels is performed on the binary image, and the character as the main image is black on the inverted binary image obtained as a result. In the main image color extraction unit 35 of the main image generation unit 21, the color of the pixel having the same coordinates as the pixel that is the black pixel on the inverted binary image is a multi-value image (original image). The average value of each pixel value on the multi-valued image is calculated, that is, the RGB value of each pixel is averaged to calculate the average value for each RGB, and the main image is displayed. A color (eg, light gray) is determined.

  In addition, the fill color extraction unit 37 of the base image generation unit 22 has a color difference component of pixels having the same coordinates as the pixels constituting the non-image part that is a white pixel around the image line part that is a black pixel on the inverted binary image. The value Cb · Cr is acquired from the data of the multivalued image subjected to the YCbCr color system conversion, the average value of the color difference component values Cb · Cr for each pixel on the multivalued image is calculated, and the average color difference component value And the fill color is determined by the luminance component value Y preset to the predetermined value (100). For example, if the portion in the multi-valued image corresponding to the non-image portion of the inverted binary image is red, the portion corresponding to the image portion of the inverted binary image is included in the background image obtained by the fill processing unit 38. The color scheme is filled with red having the same color difference as the surrounding red.

  Further, as shown in FIG. 7, as described above, the filling processing unit 38 of the base image generation unit 22 includes a portion in the original image corresponding to the main image obtained by the image extraction unit 34 of the main image generation unit 21. In addition to the main image portion fill processing unit 41 that fills with the fill color obtained by the fill color extracting unit 37, the contour portion existing in the original image so as to surround the main image portion is displayed in the outer region. A contour portion fill processing unit 43 is provided for filling with a fill color determined based on the pixel value. The Y, Cb, and Cr image signals that have undergone the fill process are converted into Y, Cb, and Cr images. The signal is sent to the background image compression unit 24 through the OR circuits 44a, 44b, and 44c for each signal.

  FIG. 9 is a conceptual diagram showing an outline of processing in the contour partial filling processing unit shown in FIG. Depending on the characteristics of the image reader and other factors, the original image may have a contour that remains thin in the drawing color of the main image so as to surround the main image. Since the efficiency is lowered and the image quality is deteriorated, the contour portion fill processing unit 43 performs a process of painting the contour portion with the color of the background portion outside the contour portion, thereby obtaining a background image from which the contour portion is removed. it can.

  FIG. 10 is a conceptual diagram showing the outline of processing in the contour partial filling processing shown in FIG. When the image signal for each pixel is input, the contour part fill processing unit 43 performs a process of determining whether or not the pixel is in the contour part and extracting a fill color one by one. Is performed in units of a predetermined window area centered on the target pixel.

  Whether or not the target pixel is in the contour portion is determined by whether or not there is a black pixel that constitutes a character (main image) in the window area, excluding the target pixel. If it is determined that the target pixel is in the contour portion, then a fill color extraction process is performed.

  In the fill color extraction process, it is first determined whether or not there are peripheral pixels that form the background portion in the window area. Specifically, in the window area, the remaining pixels excluding adjacent pixels adjacent to the black pixel are determined. It is determined whether or not there are peripheral pixels. If it is determined that there are peripheral pixels, the values of the peripheral pixels excluding the black pixels and adjacent pixels are averaged to determine the fill color of the target pixel. Is done. On the other hand, when there is no peripheral pixel, the fill color acquired by the process for the pixel immediately before the target pixel is adopted.

  For example, in the examples shown in (I) and (II), since there are peripheral pixels in the determination area, the fill color of the target pixel is determined by averaging the values of the peripheral pixels. On the other hand, in the example shown in (III), since there is no surrounding pixel, the filling color is not extracted in the determination region, and the filling color acquired by the process for the pixel immediately before the target pixel is adopted. In particular, since the fill color in the determination region is not extracted even in the immediately preceding pixel and the previous pixel, the fill color acquired in the process (II) is actually used. At this time, since there is usually no color change inside and outside the character portion, an appropriate filled color can be obtained.

  Furthermore, when using the fill color acquired in the process for the previous pixel, the contour fill used for the previous pixel is used so that the wrong color is not filled in the portion where the background color changes. The color value is compared with the current value of the target pixel, and when there is a color difference exceeding a predetermined value, the contour fill color of the immediately preceding pixel is not adopted and the fill process is not performed.

  Note that the contour partial filling process is sequentially performed for each input pixel, and the process of the pixel located behind is performed based on the value after filling. Further, here, for convenience of explanation, the main image portion which is a black pixel on the binary image is displayed superimposed on the multi-value image, but the actual image data of the main image portion is a multi-value image. Of the pixel value.

  FIG. 11 is a conceptual diagram showing an example when the size of the window area shown in FIG. 10 is changed. The window area is a square of N × N (N is the number of vertical and horizontal pixels, 3 or more), and the example shown in (A) is a window area of 3 × 3. A window region is composed of a total of nine pixels adjacent in the left, right, and diagonal directions. In the example shown in (B), the window area is 5 × 5, and the window area is configured by a total of 25 pixels.

  The size (number of pixels) of the window area can be changed according to the state of the original image, specifically, the width of the contour portion. In the determination processing whether the target pixel is in the contour portion, When it is determined whether or not there is a black pixel except for the target pixel, as shown in (B), if the window area is set large according to the width of the contour portion, the character portion ( Pixels from the boundary pixels (black pixels) of the main image portion) to a predetermined distance can be determined to be in the contour portion and can be filled, and the accuracy of the filling processing can be improved.

  FIG. 12 is a flowchart showing a processing procedure in the multi-layer image generation unit shown in FIG. This is a continuation of the flow shown in FIG. 4, and in the process of obtaining and filling the fill color of the background image in units of processing blocks, a character list describing information about the processing blocks for each character was created. Later (step 118), a process of filling the contour portion existing around the character portion (main image portion) is first performed by the contour portion fill processing unit 43 (step 131), and then the processing block unit described in the character list Then, the process of acquiring the fill color of the character part is performed in the fill color extracting unit 37 (step 132), and the process of filling the character part with the acquired fill color is performed in the process block unit in the fill processing unit 38 (step). 133). The rest is the same as the example shown in FIG.

  FIG. 13 is a flowchart showing a procedure of the contour portion filling process shown in FIG. Here, in the manner shown in FIG. 10, the contour portion where the target pixel exists around the character portion (main image portion) based on the appearance state of the black pixel in the N × N window region centered on the target pixel. If the target pixel is an outline part, a process of extracting a fill color from the values of surrounding pixels located at a position separated from the character part in the window area is performed. The process of filling the target pixel with the outline filling color obtained by the above is performed.

  First, in the contour portion filling processing unit 43, the target pixel constitutes the contour portion depending on whether or not a black pixel exists in the N × N window region centered on the target pixel except for the target pixel at the center. Is determined (step 201). When it is determined that a black pixel exists in the window area and the target pixel constitutes a contour portion, the pixel of the main image is next displayed in the window area. (Black pixel) and a process for determining whether there is a peripheral pixel that does not correspond to any of the adjacent pixels adjacent to it (Step 202), so that the peripheral pixel exists in the window area When the determination is made, a process for determining the contour fill color by averaging the pixel values of the surrounding pixels is performed (step 203), and a fill process for replacing the value of the target pixel with the value of the contour fill color obtained here is performed. Be called Step 204).

  On the other hand, in the determination of whether or not the target pixel is in the contour portion based on the presence or absence of the black pixel in the window region (step 201), there is no black pixel in the window region and the target pixel does not constitute the contour portion. If determined, the painting is not performed. If it is determined in the step 202 whether or not there are peripheral pixels in the window area, it is determined that there are no peripheral pixels in the window area. Step 203) is not performed, and a fill process is performed in which the value of the target pixel is replaced with the value of the outline fill color acquired immediately before (step 204).

  FIG. 14 is a flowchart showing a procedure of the character filling color extraction process shown in FIG. Here, it is determined whether or not the target character portion constitutes a white character, and in the case of a normal character that is not a white character, the fill color is determined in the manner shown in FIG. In the case of white characters, the fill color is determined in the manner shown in FIG.

  First, the values (Y, Cb, Cr) of pixels located on the outer periphery of the character image for each character segmented by the image extracting unit 34 are searched from the multi-valued image, and character peripheral color candidates are determined from the pixel values. The required processing is performed by the fill color extracting unit 37 (step 301). Then, the filling processing unit 38 determines whether or not the target character portion is a white character (step 302). If it is determined that the character portion is a white character, the character surrounding color acquired by the painting color extracting unit 37 is determined. A process of obtaining a fill color from the candidate color difference component values Cb and Cr and the luminance component value Y set to a predetermined fixed value is performed (step 303). On the other hand, if it is determined that the character is not a white character (step 302), a process for obtaining a fill color from the luminance component value Y of the character surrounding color candidate and the color difference component value Cb / Cr set to a predetermined fixed value is performed (step 302). 304).

  FIG. 15 is a flowchart showing a procedure of the character part filling process shown in FIG. Here, it is first determined whether or not black pixels are present in the binary image (step 401). If it is determined that black pixels are present in the binary image, the filling shown in FIG. Processing is performed to fill the pixels on the multi-valued image at the same position (coordinates) as the character (main image portion) with the fill color determined in the color extraction processing (step 402). On the other hand, when there is no black pixel in the binary image, the painting process is not performed.

  The image processing apparatus and the image processing method according to the present invention have an effect of simplifying a calculation process for obtaining a fill color when generating a background image for multi-layering a color image. It is useful as an image processing apparatus and an image processing method that perform compression processing using different compression methods separately for main images such as characters and background images.

1 is a block diagram showing a schematic configuration of a multifunction machine to which an image processing apparatus according to the present invention is applied. The block diagram which shows the structure of the multi-layer image production | generation part shown in FIG. The conceptual diagram which shows the point of the process in the multilayer image generation part shown in FIG. The flowchart which shows the procedure of the process in the multilayer image generation part shown in FIG. The flowchart which shows the procedure of the process in the multilayer image generation part shown in FIG. The block diagram which shows another example of the multi-layer image production | generation part shown in FIG. The block diagram which shows in detail the structure of the base image generation part shown in FIG. FIG. 7 is a conceptual diagram showing a processing procedure in the base image generation unit shown in FIG. The conceptual diagram which shows the outline | summary of the process in the outline partial filling process part shown in FIG. The conceptual diagram which shows the point of the process in the outline partial filling process shown in FIG. Conceptual diagram showing an example when the size of the window area shown in FIG. 10 is changed The flowchart which shows the procedure of the process in the multi-layer image production | generation part shown in FIG. The flowchart which shows the procedure of the outline partial filling process shown in FIG. The flowchart which shows the procedure of the character filling color extraction process shown in FIG. The flowchart which shows the procedure of the character part filling process shown in FIG.

Explanation of symbols

5 Multilayer image generation unit 21 Main image generation unit (main image generation means)
22 Background image generation unit (background image generation means)
23 Main image compression unit (image compression means)
24 Base image compression unit (image compression means)
37 Fill Color Extraction Unit 38 Fill Processing Unit

Claims (11)

A main image generating means for extracting from the original image a main image drawn with a luminance different from that of the background in the original image;
A background image generation means for generating a background image by painting a portion in the original image corresponding to the main image with a required fill color based on the image data of the surrounding pixels;
Image compression means for compressing each image data of the main image and the background image obtained thereby by different compression methods;
The image processing apparatus, wherein the background image generation unit determines the fill color based on a luminance component value determined based on luminance component values of peripheral pixels and a preset chromaticity component value.   The image processing apparatus according to claim 1, wherein the main image is at least one of a character, a ruled line, and a line drawing figure.   The image processing apparatus according to claim 1, wherein a chromaticity component value for determining a fill color in the base image generation unit is preset to an intermediate value, and the fill color is an achromatic color.   The image compression means compresses the background image by JPEG, and the sampling ratio Y: Cb: Cr of the luminance component value and the chromaticity component value at the time of the JPEG compression is 4: 1: 1, 4: 2: 2. And the image processing apparatus according to claim 1, wherein the image processing apparatus is any one of 4: 2: 0.   The image processing apparatus according to claim 1, wherein the filling process by the base image generation unit is performed in a predetermined block unit which is a processing unit in the JPEG compression process. A main image generating means for extracting from the original image a main image drawn with a luminance different from that of the background in the original image;
A background image generation means for generating a background image by painting a portion in the original image corresponding to the main image with a required fill color based on the image data of the surrounding pixels;
Image compression means for compressing each image data of the main image and the background image obtained thereby by different compression methods;
In the case where the base image is a normal image in which the main image has a lower luminance than the surroundings, the background image generating means uses the luminance component value determined based on the luminance component value of the surrounding pixels and the chromaticity component value set in advance. When the fill color is determined and the main image is an inverted image having higher brightness than the surroundings, the chromaticity component value determined based on the chromaticity component values of the peripheral pixels and the preset luminance component value An image processing apparatus characterized by determining a fill color. A main image generating means for extracting from the original image a main image drawn with a luminance different from that of the background in the original image;
A background image generating means for generating a background image by painting a portion in the original image corresponding to the main image with a required main image fill color based on image data of surrounding pixels;
Image compression means for compressing each image data of the main image and the background image obtained thereby by different compression methods;
The background image generating means performs a process of filling a contour portion existing in the original image so as to surround the periphery of the main image portion with a contour fill color acquired from the pixel value of the outer region. An image processing apparatus.   The image processing apparatus according to claim 7, wherein the base image generation unit extracts a contour fill color for each pixel using a window region having a predetermined number of pixels centered on the target pixel. Extracting from the original image a main image drawn with a brightness different from that of the background in the original image;
Painting a portion in the original image corresponding to the main image with a required fill color based on image data of the surrounding pixels to generate a background image;
Compressing each image data of the main image and the background image obtained thereby by different compression methods,
In the step of generating the background image, the fill color is determined based on a luminance component value determined based on luminance component values of peripheral pixels and a preset chromaticity component value. . Extracting from the original image a main image drawn with a brightness different from that of the background in the original image;
Painting a portion in the original image corresponding to the main image with a required fill color based on image data of the surrounding pixels to generate a background image;
Compressing each image data of the main image and the background image obtained thereby by different compression methods,
In the step of generating the background image, when the main image is a normal image having lower luminance than the surroundings, a luminance component value determined based on the luminance component values of the surrounding pixels, a preset chromaticity component value, The fill color is determined by the above, and when the main image is an inverted image having higher brightness than the surroundings, the chromaticity component value determined based on the chromaticity component values of the surrounding pixels, and a preset luminance component value An image processing method, wherein the fill color is determined by: Extracting from the original image a main image drawn with a brightness different from that of the background in the original image;
Painting a portion in the original image corresponding to the main image with a required main image fill color based on image data of surrounding pixels to generate a background image;
Compressing each image data of the main image and the background image obtained thereby by different compression methods,
In the step of generating the background image, a process of filling a contour portion existing in the original image so as to surround the periphery of the main image portion with a contour fill color obtained from the pixel value of the outer region is performed. A featured image processing method.

Images