JP2010004491A - Image processor, image forming apparatus, image processing method, image processing program, and computer readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor capable of identifying a local background region and a photographic region having the same color as the local background region. <P>SOLUTION: In a region separation processing part 15 of the image processor 1, an overlapping determining part 157 determines the presence of overlapping of accumulation object sections in an accumulation histogram up to a prior stage following a processing sequence and in a sectional histogram of a processing object stage, and an accumulating part 154 accumulates the accumulation object sections determined to have overlapping, to generate the accumulation histogram up to the processing object stage. An image identification value assigning part 155 assigns image identification values to pixels belonging to the accumulation object sections determined to have no overlapping. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a digital image feature separation method and system.

  In order to perform an image quality improvement process on a digital image typified by a scanned image of a document document, it is common to identify an image area and apply an optimum image quality improvement process to each identified image area. . Since the effect of the image quality improvement processing largely depends on the identification accuracy of the image area, development of a technique for identifying the image area is underway. Examples of the image area to be identified include a character area, a halftone dot area, a background area, and a photograph area.

  FIG. 11 shows an example of a general document image. The document image shown in FIG. 11 is composed of a character area, a halftone dot area, and a page background (background) area (tag number 6 in FIG. 11). More specifically, the dot area is a local background area (tag numbers 1 and 2 in FIG. 11), a photo area (tag number 7 in FIG. 11), and a graphic area (tag numbers 3, 4, and 5 in FIG. 11). Composed.

  As a conventional technique for identifying an image region, for example, there is an image processing apparatus disclosed in Patent Document 1. Patent Document 1 is a technique for identifying background information including a page background and a local background from a digital image.

  FIG. 12 is a diagram illustrating a processing flow of the image processing apparatus described in Patent Document 1. In this image processing apparatus, first, an input digital image is divided into a plurality of sections that do not overlap (S201). Next, a segment histogram is generated for each segment divided in S201 (S202). The feature amount of the segment histogram is calculated from the segment histogram generated in S202 (S203), and the feature amount is accumulated (S204). Finally, a significant value corresponding to each local background region is assigned based on the feature accumulation result of the segmented histogram (S205).

  FIG. 13 is a diagram showing in more detail the processing contents of S201 and S202 in the processing flow shown in FIG. As shown in FIG. 13, the image processing apparatus described in Patent Document 1 divides an input digital image into N sections for each band. The section histogram 301 shown in FIG. 13 corresponds to the first section histogram corresponding to the first section, the section histogram 302 corresponds to the second section histogram corresponding to the second section, and the section histogram 303 corresponds to the Nth section. An Nth segment histogram is shown.

  FIG. 14 is a diagram showing in more detail the processing contents of S204 and S205 in the processing flow shown in FIG. FIG. 14 shows a feature accumulation result obtained by accumulating the feature amounts calculated from the respective segment histograms, and for each local background region indicated by the feature accumulation result, image identification that is a significant value corresponding to each local background region Image identification value 4 is assigned from value 1.

With the above processing, the image processing apparatus described in Patent Document 1 can know local background distribution information for a page more accurately than an apparatus that simply determines a local background area from a histogram of an entire digital image. The background can be detected. Specifically, since the count value of the feature accumulation result increases as the feature amount in each section reaches the entire page, it can be determined that there is a high possibility that it is the background of the entire page. Moreover, it can be determined that the lower the count value of the feature accumulation result, the higher the possibility of a local background.
JP 2007-235948 A (published September 13, 2007)

  However, in Patent Document 1, when the same color as the color determined as the local background is also used in the photographic region, it cannot be accurately separated. For example, in the manuscript shown in FIG. 11, the local background is the portion indicated by the tag numbers 1 and 2, but the local background indicated by the tag number 1 and the photo area indicated by the tag number 7 have the same color. If this is the case, there is a problem that it is determined to be a local background area even though it is a photographic area.

  The present invention has been made in view of the above problems, and its purpose is to improve the determination accuracy of the local background area and to identify the local background area and the photographic area having the same color as the local background area. An image processing apparatus, an image forming apparatus, an image processing method, an image processing program, and a computer-readable recording medium are provided.

  The present invention is to solve the above-mentioned problem, adaptively extract a divided section of an input image, and image identification values in a page background area and a local background area based on the feature amount of the extracted histogram in the divided section. Assign. For example, in the process of generating segmented histograms in units of bands sequentially from the top to the bottom of the input image and extracting the feature values of the segmented histogram, the latest segmented feature extraction result is compared with the previous accumulated feature information, and the accumulated features In the information, if the existing local background area does not exist in the latest segmented feature extraction result, a new image identification value is assigned to the area corresponding to the local background area that no longer exists, and the new image identification value by the assigned image identification value Add to. By repeating the above processing, an image identification value corresponding to the local background or page background information is finally assigned to all images. In addition, in this processing, when generating a segmented histogram, the accuracy of determining the local background region is improved by calculating the histogram excluding pixels that are considered to be photographic regions and pixels that are supposed to be character pixels. For the purpose.

  In order to solve the above problems, an image processing apparatus according to the present invention is an image processing apparatus that performs region separation of an input image, and includes a dividing unit that divides the input image into a plurality of sections, and each of the divided sections. In addition, for the characteristic value related to the image, a classification histogram generating means for creating a classification histogram indicating the number of pixels having the characteristic value as a frequency, and for each of the classification histograms, an accumulation target section in which the frequency is equal to or greater than a predetermined threshold is extracted. A classification histogram feature extraction unit that performs the determination, a duplication determination unit that determines whether or not the accumulation target section overlaps in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage, and the determination that the duplication exists Accumulating means for accumulating the accumulation target section and generating a cumulative histogram up to the processing target stage; The pixels in the cumulative target section is characterized by comprising an image identification value assigning means for assigning the image identification value.

  According to the above configuration, the image is divided into a plurality of sections so as not to overlap, and for each section, a section histogram indicating the number of pixels having the characteristic value for the characteristic value regarding the image is obtained. Then, for each of the classification histograms, an accumulation target section in which the frequency is equal to or greater than a predetermined threshold is extracted. Then, the presence or absence of the overlap of the accumulation target section in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage is determined, and the accumulation target section determined to be overlapped is accumulated to be processed. A cumulative histogram up to a stage is generated, and an image identification value is assigned to pixels belonging to the accumulation target section determined to have no overlap. When generating a cumulative histogram, based on the above determination, remove the non-overlapping section from the cumulative histogram up to the previous section, accumulate the overlapping section in the cumulative histogram up to the previous stage, and Generate a cumulative histogram up to the segment. As described above, the generation of the cumulative histogram is repeated, and an image identification value is assigned whenever a non-overlapping section appears.

  Here, depending on the document, there is a possibility that the same color as the local background is used for the photographic area. However, the above configuration of the present invention improves the detection accuracy by limiting the position where each image identification value can exist and detecting the local background area in consideration of the layout information (position information) of the document. It is possible to prevent the photographic region from being misidentified as a local background. Therefore, according to the above configuration of the present invention, it is possible to identify a photographic region and a local background region having the same color as the local background region.

  In addition to the above-described configuration, the image forming apparatus according to the present invention further includes an image identification signal generation unit that generates an image identification signal that separates pixels in the local background region from the input image based on the image identification value. It may be. According to this configuration, the pixels in the local background area can be separated from the input image by the image identification signal.

  In the image processing apparatus according to the present invention, in addition to the above-described configuration, the section histogram generation unit may generate a section histogram using pixels other than character pixels. According to this configuration, segmented histograms are generated using only pixels that are likely to be local background regions, and are used to determine overlapping sections and assign image identification values, improving image identification accuracy. can do.

  In the image processing apparatus according to the present invention, in addition to the above configuration, the segmented histogram generation unit may generate a segmented histogram using pixels in a flat region. According to this configuration, segmented histograms are generated using only pixels that are likely to be local background regions, and are used to determine overlapping sections and assign image identification values, improving image identification accuracy. can do.

  In the image processing apparatus according to the present invention, in addition to the above configuration, the segmented histogram generation unit may calculate entropy of a block including a plurality of pixels including a target pixel in order to select a pixel in a flat region. Good. According to this configuration, the density distribution shape by the entropy of a block (local histogram) composed of a plurality of pixels including the target pixel, that is, the density distribution information indicating the width of the density distribution can be accurately calculated. Local background information can be calculated using pixels that are likely to be background regions. The shape of the histogram is calculated using entropy. The lower the entropy is, the flatter area (the distribution is narrower), and the higher the entropy, the tendency to be the photographic area (wider distribution). can do.

  In the image processing apparatus according to the present invention, in addition to the above configuration, the segmented histogram generation means includes a plurality of target pixels generated using pixels other than the character pixels in order to select pixels in the flat region. You may calculate the entropy of the block which consists of pixels. According to this configuration, it is possible to calculate the histogram distribution information (the entropy of a block (local histogram) including a plurality of pixels including the target pixel) ignoring the character pixel. The density distribution information can be calculated with high accuracy.

  That is, in the image processing apparatus according to the present invention, when creating a segmented histogram, the entropy of a local histogram (different from the segmented histogram) is calculated, so that only flat region pixels are targeted. Thereby, it can suppress that a photograph area pixel is counted in a division | segmentation histogram. By suppressing to some extent, the photographic region pixels can be ignored by the comparison process between the segmented histogram and the predetermined threshold.

  An image processing method according to the present invention is an image processing method for performing region separation of an input image in order to solve the above-described problem, wherein a division step for dividing the input image into a plurality of divisions, and for each of the divided divisions In addition, a segment histogram generation step for creating a segment histogram indicating the number of pixels having the characteristic value of the image as a frequency, and a cumulative target section in which the frequency is equal to or greater than a predetermined threshold is extracted for each segment histogram. A classification histogram feature extraction step, a duplication determination step for determining whether or not there is duplication of the accumulation target section in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage, and the determination that there is duplication An accumulation step for accumulating the accumulation target section to generate a cumulative histogram up to the processing target stage; Without the the pixels belonging to the determined said accumulated target section is characterized in that it comprises an image identifying value assignment step of assigning image identification value.

  According to the above method, the same effect as that of the image processing apparatus can be obtained, and a photographic region having the same color as the local background region can be identified in the digital image.

  In order to solve the above problems, an image forming apparatus according to the present invention includes any one of the image processing apparatuses described above. Since the image forming apparatus according to the present invention includes the image processing apparatus according to the present invention, the local background area and the photographic area having the same color can be identified, and therefore processing corresponding to each area is performed. And an image forming apparatus capable of outputting a high-quality image.

  The image processing apparatus according to the present invention may be realized by a computer. In this case, the image processing apparatus is realized by the computer by causing the computer to operate as each unit in the image processing apparatus. A processing program and a computer-readable recording medium on which the image processing program is recorded also fall within the scope of the present invention.

  According to these configurations, the same operational effects as those of the image processing apparatus can be realized by causing the computer to read and execute the image processing program.

  As described above, the image processing apparatus according to the present invention divides the input image into a plurality of sections, and for each of the divided sections, the number of pixels having the characteristic value for the characteristic value related to the image is set to a frequency. Section histogram generation means for creating a histogram shown as follows, section histogram feature extraction means for extracting the accumulation target section whose frequency is equal to or greater than a predetermined threshold for each section histogram, and cumulative histogram up to the previous stage according to the processing order And an overlap determination means for determining whether or not the accumulation target section overlaps with the processing target stage division histogram, and a cumulative histogram up to the processing target stage is generated by accumulating the accumulation target section determined to have overlap. And an image identification value for assigning an image identification value to the pixels belonging to the accumulation target section determined to have no overlap. And means against Ri, and a.

  Depending on the manuscript, the same color as the local background may be used for the photo area. However, the above configuration of the present invention improves the detection accuracy by limiting the position where each image identification value can exist and detecting the local background area in consideration of the layout information (position information) of the document. It is possible to prevent the photographic region from being misidentified as a local background. Therefore, according to the above configuration of the present invention, it is possible to identify a photographic region and a local background region having the same color as the local background region.

  First, a characteristic configuration of the image processing apparatus according to the present embodiment will be described. The entire image processing apparatus, image forming apparatus, and image reading apparatus according to this embodiment will be described later.

  FIG. 1 is a diagram illustrating a region separation processing unit 15 included in the image processing apparatus 1 and the image processing apparatus 10 according to the present embodiment described later. The region separation processing unit 15 includes a local background region extraction unit 150 that is a characteristic configuration of the image processing apparatus according to the present embodiment.

  The local background region extracting unit 150 includes an image dividing unit (dividing unit) 151, a segmented histogram generating unit (segmented histogram generating unit) 152, a segmented histogram feature extracting unit (segmented histogram feature extracting unit) 153, and an accumulating unit (accumulating). Means) 154, an image identification value assigning unit (image identification value assigning unit) 155, an image identification signal generation unit (image identification signal generation unit) 156, and an overlap determination unit (duplication determination unit) 157.

  The image dividing unit 151 is a block that divides an input digital image (input image) into a plurality of sections.

  The segment histogram generation unit 152 is a block that creates a segment histogram indicating, for each segment divided by the image segmentation unit 151, the number of pixels having the characteristic value of the characteristic value related to the image as the frequency. In the present embodiment, the characteristic value relating to the input digital image (horizontal axis of the classification histogram) is a density value (tone value, pixel value). However, the characteristic value is not limited to this, and for example, luminance, saturation, hue, and the like may be used.

  The segment histogram feature extraction unit 153 is a block that calculates a feature amount for each segment histogram and extracts information indicating a section having the calculated feature amount.

  The accumulating unit 154 is a block that accumulates the accumulation target sections determined as having the overlap and generates a cumulative histogram up to the processing target stage.

  The overlap determination unit 157 is a block that determines whether or not the accumulation target section overlaps in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage.

  The image identification value assigning unit 155 is a block that assigns an image identification value to pixels belonging to the accumulation target section determined to have no overlap.

  The image identification signal generation unit 156 is a block that generates an image identification signal from the input digital image based on the image identification value.

  FIG. 2 is a diagram showing a flow of local background region extraction processing performed by the region separation processing unit 15. First, the variable L and the variable n are initialized to 1, and the cumulative histogram is initialized to 0 (S501). However, the variable L is a variable representing an image identification value assigned to each local background, and the variable n is a variable representing a division number when dividing an image into divided regions. The cumulative histogram is a histogram obtained by accumulating the sections extracted by the segmented histogram feature extraction unit 153.

  Next, a segmented histogram of the nth segmented region (nth segment) of the input digital image is generated (S502). Then, the feature amount of the section histogram is calculated, and the section feature indicating the section having the calculated feature amount is extracted (S503). Here, in the present embodiment, the segment features are the start density value and the end density value of the local background region. Then, the extracted segmented feature is compared with the accumulated feature information accumulated in the accumulated histogram, and it is determined whether or not an image identification value is assigned to the local background region existing in the accumulated histogram (S504). If there is a local background area to which the image identification value is assigned (YES in S504), the image identification value L is assigned to the corresponding local background area (S505), and the feature information of the assigned local background area is deleted from the cumulative histogram (S506). ). The variable L is incremented by the necessary number, that is, by the number of new image identification values assigned at the time of the n-th segment processing so that the variable L represents the image identification value to be assigned next (S507). If there is no local background region to which the image identification value is assigned (NO in S504), the segment feature calculated in the nth segment is added to the cumulative histogram (S508). It is determined whether the processing has been completed for all the sections (S509). If not completed (S509niteNO), 1 is added to the variable n (S510), and processing for the next section area is started (return to S502). . When the process of the last section is finished (YES in S509), an image identification value is assigned to the local background area remaining in the cumulative histogram (S511), and the process is finished. Through the above processing, background information corresponding to each local background region is adaptively calculated, and an image identification signal is generated from the background information and the input digital image.

Next, the processing flow shown in FIG. 2 will be described in more detail. FIG. 3 is a detailed block diagram of the n-th segment histogram calculation process (S502) of FIG. 2 (or may be a detailed block diagram of the segment histogram generation unit 152 of FIG. 1). First, when a digital image of the nth section is input, a histogram calculation target pixel selection process is performed (S601). This process selects a histogram calculation target pixel, which is a pixel to be counted as a frequency when creating a section histogram for the nth section digital image. A condition for selecting a histogram calculation target pixel includes uniformity of the target pixel. Considering that the density of the background area is uniform compared to the density of the photo area or the like, it is determined whether or not the target pixel is a flat pixel. When it is determined that the target pixel is a flat pixel, the target pixel is selected (counted) as a histogram calculation target pixel. Here, as a feature for determining whether or not the target pixel is a flat pixel, for example, there is a variance value of I × J window data centered on the target pixel. The variance value σ ² of I × J window data can be calculated by the following equation (1).

  here,

Represents the value (for example, density value) of the input digital image data in the i-th row and j-th column,

Represents an average value (for example, density average value) of input digital image data in I × J window data.

The smaller the variance value σ ^{2, the} higher the possibility that the target pixel is a flat pixel, and the higher the variance value σ ² , the higher the possibility that the target pixel is a non-flat pixel.

  Another feature for determining whether or not the target pixel is a flat pixel is the degree of spread of the density distribution in the I × J window data centered on the target pixel. One method for calculating the spread of the density distribution is the entropy of a histogram of I × J window data. The entropy E of the histogram of I × J window data can be calculated by the following equation.

Here, h represents a histogram of I × J window data, and h (i) represents a frequency in BIN i. However, the histogram h of I × J window data is normalized so that the total frequency is 1. The lower the entropy E, the narrower the density distribution. That is, there is a high possibility that the target pixel is a flat pixel. The higher the entropy E, the higher the possibility that the target pixel is a non-flat pixel.

Moreover, a character pixel is mentioned as conditions for selecting another histogram calculation object pixel. It is determined whether or not the target pixel is a character pixel. If it is determined that the target pixel is not a character pixel, the target pixel is selected (counted) as a histogram calculation target pixel. Here, as a feature for determining the character pixel, there is a dispersion value used for determining whether or not the target pixel is a flat pixel. Target pixel as the variance value sigma ² is large is likely to be character pixels are likely to be non-character pixels smaller the variance sigma ^2.

  Another feature for determining whether the pixel of interest is a character pixel is an edge amount. As a method for calculating the edge amount, for example, Sobel filter processing can be mentioned. The edge amount G calculated by the Sobel filter process can be calculated by the following equation (3).

  However, Gx and Gy are the calculation results of the edge operator processing in the horizontal direction and the vertical direction, respectively, and are calculated by the following equations (4) and (5), respectively.

  In the n-th segment histogram calculation process of S502 of FIG. 2, a segment histogram calculation process (S602) that is a process of counting the pixels selected in the histogram calculation target pixel selection process (S601) is performed to generate a segment histogram. . Through the above processing, a segmented histogram can be generated only for a uniform region, so that the accuracy of local background region detection can be improved.

  Next, FIG. 4A is a detailed block diagram of the segment feature calculation processing (S503) of FIG. 2 (or may be called a detailed block diagram of the segment histogram feature extraction unit 153 of FIG. 1). . First, a BIN (section) for which a segment feature is to be calculated is selected for the nth segment histogram. The simplest method for selecting the target BIN is threshold comparison processing. As shown in FIG. 4B, the frequency (here, the number of pixels) in each BIN of the classification histogram is compared with a predetermined threshold, and a BIN having a frequency greater than the predetermined threshold is selected as the classification feature calculation target BIN. Next, as shown in FIG. 4C, a segment feature is calculated for a BIN having a frequency greater than a predetermined threshold. Among BINs having a frequency greater than a predetermined threshold, a continuously distributed BIN group is treated as one local background region, and the start position and end position of the local background region are calculated. Here, for a BIN having a frequency greater than a predetermined threshold, the following processing is performed in order to treat a continuously distributed BIN group as one local background region. First, the frequency in each BIN (density value) of the histogram is compared with a predetermined threshold value. Next, among BINs (density values) larger than a predetermined threshold value, BINs continuously larger than the predetermined threshold value are handled as one local background region.

  In FIG. 4C, there are two local background areas, and the start position of the first local background area LP0 is represented by LP0.st and the end position is represented by LP0.end. Similarly, the start position of the second local background region LP1 is represented by LP1.st, and the end position is represented by LP1.end.

With the above processing, the local background region information of the segment histogram can be calculated with high accuracy by simple processing. Further, although not shown in the figure, the histogram smoothing process is performed before, after, or before and after the threshold comparison process in order to improve the accuracy of detecting the local background area. The influence of artifacts caused by the tone screen can be reduced.

5 and 6 are diagrams for explaining the determination process in S504 of FIG. 2 in more detail. FIG. 5 is a diagram for explaining processing when there is a local background region to which an image identification value is assigned to the cumulative histogram (when YES is determined in S504 in FIG. 2). The determination process in S504 of FIG. 2 is a conditional branch process that returns a determination result by comparing the local background area of the cumulative histogram and the local background area of the nth segment histogram. FIG. 4A shows a cumulative histogram when the n-th segmented histogram is processed, and each of the three local background areas including the first local background area P0, the second local background area P1, and the third local background area P2. Has a background area. However, in order to confirm overlap, it is necessary to calculate the segment feature for the cumulative histogram as well, and the local background area of the cumulative histogram can be calculated by the same process as the segment feature calculation process shown in FIG.
FIG. 5B shows an n-th section histogram, and has two local background areas each composed of a first local background area LP0 and a second local background area LP1. First, the degree of overlap of the first local background region P0 of the cumulative histogram with the first local background region LP0 of the nth segment histogram is calculated. Using the width (BIN number) of the local background region of the first local background region P0 of the cumulative histogram as a reference value, the ratio between the width (BIN number) of the local background region overlapping between P0 and LP0 and the reference value is When larger than the predetermined threshold, LP0 is determined to belong to P0. When the width (number of BINs) of the local background region overlapping between P0 and LP0 is d, the determination formula is as the following formula (6).

  Next, for the first local background region P0 of the cumulative histogram, the degree of overlap with the second local background region LP1 of the nth segment histogram is calculated. Similarly, it is determined whether or not each local background region of the cumulative histogram overlaps with the local background region included in the nth section histogram. Note that in the determination processing of S504 in FIG. 2, since it is only determined whether or not there is an overlapping local background region, the first local background region P0 of the cumulative histogram is a plurality of nth segment histograms. Even when it is determined that the local background region overlaps, it is simply determined that there is an overlapping local background region. According to FIGS. 5A and 5B, the first local background region P0 of the cumulative histogram is the first local background region LP0 of the nth segment histogram, and the third local background region P2 of the cumulative histogram is the nth segment. Although it overlaps with the second local background region LP1 of the histogram, the second local background region P1 of the cumulative histogram does not overlap with any local background region included in the nth segment histogram. Note that the determination of whether to assign an image identification value to the local background area existing in the cumulative histogram, which is the process in S504 of FIG. 2, is specifically, the local background area of the cumulative histogram is the nth segment histogram. It is determined whether or not any local background region overlaps. Therefore, if they do not overlap, it is determined in S504 of FIG. 2 that there is a local background area to be allocated (YES in S504). In accordance with the processing flow shown in FIG. 2, a new image identification value L is assigned to the second local background region P1 of the cumulative histogram (S505), the second local background region P1 information is deleted from the cumulative histogram (S506), and finally Then, a cumulative histogram shown in FIG.

  On the other hand, FIG. 6 is a diagram for explaining processing when there is no local background region to which an image identification value is assigned to the cumulative histogram (when NO is determined in S504 in FIG. 2). FIG. 6A shows a cumulative histogram when processing the (n + 1) -th section histogram, and has two local background areas each composed of a first local background area P0 and a second local background area P1. FIG. 6B represents the (n + 1) th segment histogram, and has two local background regions each including a first local background region LP0 and a second local background region LP1. In the same manner as the processing described with reference to FIG. 5, it is determined whether or not each local background region of the cumulative histogram overlaps with the local background region included in the (n + 1) th section histogram. According to FIGS. 6A and 6B, the first local background region P0 of the cumulative histogram is the first local background region LP0 of the (n + 1) th segment histogram, and the second local background region P1 of the cumulative histogram is the (n + 1) th. Since it overlaps with the second local background region LP1 of the segmented histogram, in S504 of FIG. 2, it is determined that there is no local background region to be allocated (NO in S504), and the cumulative histogram is compared with the cumulative histogram according to the processing flow shown in FIG. The segment features of the n + 1 segment histogram are added (S508).

  FIG. 7 is a diagram for explaining the processing of S511 in FIG. 2 in detail. FIG. 7 shows a cumulative histogram (final cumulative histogram) when the processing is completed for all sections, and has two local background areas each composed of a first local background area P0 and a second local background area P1. In the process of S511 in FIG. 2, an image identification value L is assigned to the first local background region P0, and an image identification value L + 1 is assigned to the second local background region P1.

  In the present embodiment, the background area is also treated as a local background area. Therefore, an area that exists over the entire image is determined as a base area. Here, the first local background region P0 and the second local background region P1 are sections accumulated up to the end, and one of them existing over the entire image is determined as the background region.

  FIG. 8 is a diagram illustrating an example of a configuration of information corresponding to each image identification value assigned in the local background region extraction process.

  In order to efficiently generate an image identification signal from the image identification value and the input digital image, local background information and classification information corresponding to each image identification value are held as image identification information.

  The local background information is information indicating the start position and the end position of the local background area. The category information is information indicating whether or not an assigned image identification value exists in each category. In the classification information shown in FIG. 8, “1” indicates that an image identification value exists, and “0” indicates that no image identification value exists. The image identification value assigning unit 155 in FIG. 1 compares the value (for example, density value) of the input digital image data located in each section with the local background information, and determines whether or not to assign an image identification value for each pixel. In FIG. 8, the image identification value 1 consisting of the density values 0 to 16 exists only in the first section, and even if the same density value exists in the second section and the Nth section, the image identification value 1 is not assigned. Means. Since the image identification value 2 exists in the first division and the second division, but does not exist in the Nth division, the image identification value 2 is set even if pixels having density values of 32 to 48 exist in the Nth division. It means not assigning.

  The image identification value 6 compares the data value (for example, density value) of the entire input digital image with the local background information, and determines whether or not the image identification value 6 is assigned to each pixel. However, a pixel determined not to be a local background is identified as being a local background region or a non-local background region by assigning 0. That is, an image identification value 0 is assigned to a pixel to which image identification values 1 to 6 are not assigned, indicating a non-local background region.

  Through the above processing, local background information can be extracted from the input digital image with high accuracy. Further, for the original shown in FIG. 1, an image identification signal (region identification signal) can be generated by assigning a unique image identification value to each local background region. Based on this image identification signal, it is determined that the pixel of interest is a pixel in the local background region.

  As described above, in this embodiment, when creating a segmented histogram, the entropy of the local histogram is calculated to target only the flat region pixels. Thereby, it can suppress that a photograph area pixel is counted in a division | segmentation histogram. By suppressing to some extent, the photographic region pixels can be ignored by the comparison process between the segmented histogram and the predetermined threshold. Even if the segment histogram is not counted from the photographic area pixels, the same color as the local background area may be used for the photographic area depending on the document. However, in the present embodiment, it is possible to suppress detection of pixels in the photographic region as a local background by limiting the positions where each image identification value can exist.

  FIG. 9 is a block diagram showing a configuration of an image forming apparatus (digital color multifunction peripheral) 2 including an image processing apparatus (color image processing apparatus) 1 according to an embodiment of the present invention. A color image input device 3, a color image output device 4, a transmission device 5 and an operation panel 6 are connected to the image processing device 1, thereby constituting a digital color multifunction device 2 as a whole.

  The image processing apparatus 1 includes an A / D (analog / digital) conversion unit 11, a shading correction unit 12, an input tone correction unit 13, a compression processing unit 14, a region separation processing unit 15, a color correction unit 16, and a black generation lower color. The configuration includes a removal unit 17, a spatial filter processing unit 18, an output tone correction unit 19, and a tone reproduction processing unit 20.

  The operation panel 6 includes a setting button, a numeric keypad, a display unit such as a liquid crystal display, and the like for controlling the operation of the digital color MFP 2.

  The color image input device (image reading means) 3 is composed of, for example, a scanner unit equipped with a CCD (Charge Coupled Device), and the reflected light image from the original is converted into RGB (R: red, G: green, B: blue). Is read as an analog signal by a CCD and input to the image processing apparatus 1.

  The analog signal read by the color image input device 3 passes through the image processing device 1 within the A / D conversion unit 11, the shading correction unit 12, the input tone correction unit 13, the compression processing unit 14, and the region separation processing unit 15. The color correction unit 16, the black generation and under color removal unit 17, the spatial filter processing unit 18, the output tone correction unit 19, and the tone reproduction processing unit 20 are sent in this order, and are output as a stream to the color image output device 4. .

  An A / D (analog / digital) conversion unit 11 converts an RGB analog signal into a digital signal, and a shading correction unit 12 applies color to the digital RGB signal sent from the A / D conversion unit 11. Processing for removing various distortions generated in the illumination system, imaging system, and imaging system of the image input apparatus 3 is performed.

  The input tone correction unit 13 adjusts the color balance of the RGB signal (RGB reflectance signal) from which various distortions have been removed by the shading correction unit 12, and at the same time, is used in the image processing apparatus 1 such as a density signal. The signal is converted into a signal that can be easily handled by the image processing system.

  The compression processing unit 14 performs compression processing on the RGB signal that has been subjected to the input tone correction processing. The compressed image data is temporarily stored in a storage means such as a hard disk. For example, when the scan to e-mail mode is selected on the operation panel 6, it is sent to a mail from the transmission device 5 such as a network card or a modem. It is sent to the destination set as an attachment. When the compression processing is not performed, the compression processing unit 14 outputs the RGB signal input from the input tone correction unit 13 to the subsequent region separation processing unit 15 as it is.

  The region separation processing unit 15 separates each pixel in the input image into one of a character region, a halftone dot region, and a photograph region from the RGB signal. As described above, the region separation processing unit 15 further separates the local background region. Then, based on the separation result, the region separation processing unit 15 generates an image identification signal (or region identification signal) indicating which region the pixel belongs to, a black generation and under color removal unit 17, a spatial filter processing unit 18, The RGB signal output from the compression processing unit 14 is output to the subsequent color correction unit 16 as it is.

  The color correction unit 16 performs a process of removing color turbidity based on spectral characteristics of CMY (C: cyan, M: magenta, Y: yellow) color materials including unnecessary absorption components in order to realize faithful color reproduction. .

  The black generation and under color removal unit 17 generates a black (K) signal from the three color signals of CMY after color correction by the color correction unit 16, and subtracts the K signal obtained by black generation from the original CMY signal. To generate a new CMY signal. The CMY three-color signal is converted into a CMYK four-color signal by processing in the black generation and lower color removal unit 17.

  As an example of the black generation process, there is a method (general method) for generating black by skeleton black. In this method, the input / output characteristic of the skeleton curve is y = f (x), the input data is C, M, Y, the output data is C ′, M ′, Y ′, K ′, UCR (Under Color When the removal rate is α (0 <α <1), the black generation and under color removal processing is expressed by the following equation (7).

  The spatial filter processing unit 18 performs spatial filter processing by a digital filter on the image data of the CMYK signal input from the black generation and under color removal unit 17 based on the region identification signal, thereby correcting the spatial frequency characteristics. Processing is performed to prevent blurring of the output image and deterioration of graininess. Similar to the spatial filter processing unit 18, the gradation reproduction processing unit 20 performs predetermined processing on the image data of the CMYK signal based on the region identification signal.

  For example, the region separated into characters by the region separation processing unit 15 has a high frequency enhancement amount in the sharp enhancement processing in the spatial filter processing by the spatial filter processing unit 18 in order to improve the reproducibility of black characters or color characters in particular. Increased. At the same time, the gradation reproduction processing unit 20 selects binarization or multi-value processing on a high-resolution screen suitable for high frequency reproduction.

  Further, with respect to the region separated into halftone dots by the region separation processing unit 15, the spatial filter processing unit 18 performs low-pass filter processing for removing the input halftone component. The output tone correction unit 19 performs an output tone correction process for converting a signal such as a density signal into a halftone dot area ratio that is a characteristic value of the color image output device 4. Then, gradation reproduction processing (halftone generation) is performed so that the image is finally separated into pixels and each gradation is reproduced.

  For the region separated into photographs by the region separation processing unit 15, binarization or multi-value processing is performed on the screen with an emphasis on gradation reproducibility.

  The image data subjected to the above-described processes is temporarily stored in the storage means, read out at a predetermined timing, and input to the color image output device 4.

  The color image output device 4 outputs image data onto a recording medium (for example, paper). Examples of the color image output device 4 include a color image output device using an electrophotographic method or an inkjet method, but are particularly limited. It is not a thing.

  The above processing is controlled by a CPU (Central Processing Unit) (not shown).

  When sending a facsimile, if a modem is used to perform a transmission procedure with the other party and the transmission is ready, image data compressed in a predetermined format (image data read by a scanner) Is read from the memory, subjected to necessary processing such as changing the compression format, and sequentially transmitted to the other party via the communication line. When receiving a facsimile, the CPU receives image data transmitted from the other party while performing a communication procedure, and inputs the received image data to the image processing apparatus 1. The image processing apparatus 1 receives the received image data (not shown). The compression / decompression processing unit performs decompression processing. The decompressed image data is subjected to rotation processing and resolution conversion processing as necessary, subjected to output tone correction and tone reproduction processing, and is output from the color image output device 4. The digital color multifunction device 2 performs data communication with a computer connected to the network and other digital multifunction devices via a LAN cable with the transmission device 5 such as a network card.

  In the above, a color multifunction peripheral has been described, but a monochrome multifunction peripheral may be used.

  FIG. 10 is a block diagram illustrating a configuration of an image reading apparatus (flatbed scanner) 100 according to an embodiment of the present invention. As shown in FIG. 10, a color image processing apparatus (image processing apparatus) 10 includes an A / D conversion unit 11, a shading correction unit 12, an input tone correction unit 13, and a compression processing unit 14. In addition, the color image input device 3 and the operation panel 6 are connected to constitute an image reading device 100 as a whole.

  The color image input device (image reading means) 3 is composed of, for example, a scanner unit equipped with a CCD (Charge Coupled Device), and the reflected light image from the original is converted into RGB (R: red, G: green, B: blue). The analog signal is read by the CCD and input to the color image processing apparatus 10.

  An analog signal read by the color image input device 3 is converted into an A / D conversion unit 11, a shading correction unit 12, an input tone correction unit 13, a compression processing unit 14, and a region separation processing unit in the color image processing device 10. Sent in order of 15.

  The A / D converter 11 converts RGB analog signals into digital signals, and the shading correction unit 12 applies the color image input device 3 to the digital RGB signals sent from the A / D converter. The process which removes the various distortion which arises in this illumination system, an imaging system, and an imaging system is performed. The shading correction unit 12 adjusts the color balance.

  The input tone correction unit 13 adjusts the color balance of the RGB signal (RGB reflectivity signal) from which various distortions have been removed by the shading correction unit 12 and at the same time supplies the color image processing apparatus 10 with a density signal and the like. A process of converting the signal into an easy-to-handle signal of the employed image processing system is performed.

  The compression processing unit 14 performs the above-described compression processing on the RGB signal that has been subjected to the input tone correction processing. The compressed image data is temporarily stored in storage means such as a hard disk. For example, when scan to e-mail mode is selected on the operation panel, the destination set by attaching to the mail from the network card Sent to. When the compression processing is not performed, the compression processing unit 14 outputs the RGB signal input from the input tone correction unit 13 as it is.

  The region separation processing unit 15 separates each pixel in the input image into one of a character region, a halftone dot region, and a photograph region from the RGB signal. As described above, the region separation processing unit 15 further separates the local background region. Then, based on the separation result, the region separation processing unit 15 outputs an image identification signal (or region identification signal) indicating which region the pixel belongs to. Here, when an RGB signal and a region identification signal are output from the image reading apparatus 100, these signals are input to a multifunction device, a printer, and a computer connected via a network and processed. In this case, it is necessary to configure the multifunction apparatus, the printer, and the computer to recognize the area identification signal and perform processing according to the area identification signal.

  A digital camera may be used as the image reading apparatus 100.

  As another embodiment of the present invention, the above-described local background region extraction processing is performed on a computer-readable recording medium in which program codes (execution format program, intermediate code program, source program) of a program to be executed by a computer are recorded. A recording of the image processing method to be performed.

  As a result, it is possible to provide a portable recording medium on which a program code for performing the image processing method for performing the local background region extraction process is recorded.

  In the present embodiment, as the recording medium, a memory (not shown) such as a ROM itself may be a program medium because processing is performed by a microcomputer. However, it may be a program medium provided with a program code reading device as an external storage device and readable by inserting a recording medium therein.

  In any case, the stored program may be configured to be accessed and executed by the microprocessor, or in any case, the program code is read and the read program code is stored in the microcomputer. The program code may be downloaded to a program storage area (not shown) and the program code is executed. It is assumed that this download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, such as a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a flexible disk or a hard disk, a CD-ROM / MO / MD / DVD, or the like. Semiconductors such as optical discs, IC cards (including memory cards) / optical cards, etc., or mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash ROM, etc. It may be a medium that carries a fixed program including a memory.

  In the present embodiment, since the system configuration is such that a communication network including the Internet can be connected, it may be a medium that fluidly carries a program so as to download a program code from the communication network. When the program code is downloaded from the communication network in this way, the download program may be stored in the main device in advance, or may be installed from another recording medium. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  The recording medium is read by a program reading device provided in a digital color image forming apparatus or a computer system, whereby the above-described image processing method is executed.

  The computer system includes an image input device such as a flatbed scanner, a film scanner, and a digital camera, a computer that performs various processes such as the above image processing method by loading a predetermined program, and a CRT display that displays the processing results of the computer. An image display device such as a liquid crystal display and a printer that outputs the processing results of the computer to paper or the like. Furthermore, a network card, a modem, and the like are provided as communication means for connecting to a server or the like via a network.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and the invention can be obtained by appropriately combining the technical means disclosed in each embodiment. Embodiments are also included in the technical scope of the present invention.

  The present invention can be applied to a color copying machine, a color printer, a monochrome copying machine, a monochrome printer, an MFP (Multi Function printer), a scanner, a personal computer, and the like having an image processing function.

It is a figure which shows the area | region separation process part with which the image processing apparatus which is one Embodiment of this invention is provided. It is a figure which shows the flow of the local background area | region extraction process performed in the area | region separation process part of the image processing apparatus which is one Embodiment of this invention. It is the figure which showed the processing content of S502 in the processing flow shown in FIG. 2 in detail. (A)-(c) is a figure which shows the processing content of S503 in the processing flow shown in FIG. 2 in detail. (A)-(c) is a figure for demonstrating in detail the determination process of S504 in the processing flow shown in FIG. (A)-(c) is a figure for demonstrating in detail the determination process of S504 in the processing flow shown in FIG. It is a figure for demonstrating in detail the process of S511 in the processing flow shown in FIG. It is a figure which shows an example of a structure of the information corresponding to each image identification value allocated by the local background area | region extraction process in the image processing apparatus of this embodiment. 1 is a diagram illustrating a configuration of an image forming apparatus including an image processing apparatus according to an embodiment of the present invention. 1 is a diagram illustrating a configuration of an image reading apparatus according to an embodiment of the present invention. It is a figure which shows the example of a general document image. It is a figure which shows the processing flow of the conventional image processing apparatus. It is the figure which showed in detail the processing content of S201 and S202 in the processing flow shown in FIG. It is the figure which showed the processing content of S204 and S205 in the processing flow shown in FIG. 12 in detail.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 2 Image forming apparatus 15 Area | region separation process part 100 Image reader 150 Local background area | region extraction part 151 Image separation part (dividing means)
152 Sectional histogram generation unit (partition histogram generation means)
153 Segment histogram feature extraction unit (segment histogram feature extraction means)
154 Accumulation part (accumulation means)
155 Image identification value assigning unit (image identification value assigning means)
156 Image identification signal generator (image identification signal generator)
157 Duplicate judgment unit (duplication judgment means)

Claims (10)

An image processing apparatus that performs region separation of an input image,
Dividing means for dividing the input image into a plurality of sections;
For each of the divided sections, a section histogram generating means for creating a section histogram indicating the number of pixels having the characteristic value as a frequency for the characteristic value related to the image;
For each of the classification histograms, a classification histogram feature extraction unit that extracts an accumulation target section in which the frequency is equal to or greater than a predetermined threshold;
A duplication judgment means for judging whether or not there is duplication of the accumulation target section in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage;
Accumulating means for accumulating the accumulation target section determined to have the overlap and generating a cumulative histogram up to the processing target stage;
An image processing apparatus comprising: an image identification value assigning unit that assigns an image identification value to pixels belonging to the accumulation target section determined to have no overlap.   The image processing apparatus according to claim 1, further comprising an image identification signal generation unit configured to generate an image identification signal that separates a pixel in a local background area from the input image based on the image identification value.   The image processing apparatus according to claim 1, wherein the section histogram generation unit generates a section histogram using pixels other than character pixels.   The image processing apparatus according to claim 1, wherein the section histogram generation unit generates a section histogram using pixels in a flat region.   The image processing apparatus according to claim 4, wherein the segmented histogram generating unit calculates entropy of a block including a plurality of pixels including a target pixel in order to select a pixel in a flat region.   The section histogram generation means calculates entropy of a block composed of a plurality of pixels including a pixel of interest generated using a pixel other than a character pixel in order to select a pixel in a flat region. 5. The image processing apparatus according to 5. An image processing method for performing region separation of an input image,
A dividing step of dividing the input image into a plurality of sections;
For each of the divided sections, a section histogram generation step of creating a section histogram indicating the number of pixels having the characteristic value as a frequency for the characteristic value related to the image;
For each section histogram, a section histogram feature extraction step for extracting a cumulative target section in which the frequency is equal to or greater than a predetermined threshold;
A duplication determination step for determining whether or not there is duplication of the accumulation target section in the cumulative histogram up to the previous stage according to the processing order and the classification histogram of the processing target stage;
An accumulation step for accumulating the accumulation target section determined as having the overlap and generating a cumulative histogram up to a processing target stage;
An image processing method, comprising: an image identification value assigning step for assigning an image identification value to pixels belonging to the accumulation target section determined to have no overlap.   An image forming apparatus comprising the image processing apparatus according to claim 1.   A program for operating the image processing apparatus according to claim 1, wherein the computer functions as each unit of the image processing apparatus.   A computer-readable recording medium on which the image processing program according to claim 9 is recorded.

Images