JP2008278372A - Image processor and processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus and method capable of applying suitable image processing to an image read from a document. <P>SOLUTION: The present invention relates to the image processor for processing an image read from a document, including a density detecting means for detecting a density of a background of the document and a bottom density in the image and a background correcting means for performing background correction on the image, wherein the background correcting means compares the density of the background with the bottom density. In the case that the bottom density is lower than the density of the background, background correction is not performed and at the same time, gradations are allocated, to a region of low density, more than those in a region of high density. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing method, and more particularly to an image processing apparatus and an image processing method for processing an image read from a document.

  In general, when a paper document is digitized (when it is read by a scanner and converted into image data), background processing is performed. This is because the background processing can remove noise and the like due to the background of the document, and can improve the readability of the image.

  On the other hand, in recent years, various paper documents can be digitized and managed by the enforcement of the e-document law. From such a background, the authenticity and visibility of documents managed electronically are regarded as important. The authenticity of a document refers to the certainty that the document has not been tampered with. In general, correction liquid, correction tape, and the like (hereinafter collectively referred to as “correction liquid”) are used for falsification of paper documents. Therefore, in order to enable detection of falsification in a state where the document exists as paper, it is required that the correction mark by the correction liquid can be identified in the digitized document.

  However, if the background processing is executed when the paper document is digitized, the correction marks are also removed. This is because the color density of the correction liquid is lower than the density of the background of a document that is generally used. Therefore, in this case, there is a problem that the correction mark cannot be identified in the document (image) after digitization, that is, the authenticity of the document cannot be guaranteed.

Therefore, for example, in the technique described in Patent Document 1, in order to leave a low density region such as a correction mark, the density statistic of the image is taken, and this is saturated for an image including an object having a higher whiteness than the background. The shading correction is performed while ensuring the dynamic range so that the brightness is corrected. Further, the correction mark can be confirmed by enhancing the local spatial contrast on the brightness-corrected image.
JP 2006-262344 A

  However, in the technique described in Patent Document 1, the same background processing is uniformly executed according to the designation by the user regardless of the presence or absence of a region having a lower density than the background of the image. Therefore, for example, a process for leaving a background is performed on an image without a correction mark in the same manner as for a document with a correction mark.

  In addition, there is a place where the expectation of the preservation cost of the document by digitizing the paper document is expected in the background of the enforcement of the e-document law. When the background is left, the document is electrified by the amount of information about the background. The amount of data (file size) tends to increase, and the document storage cost increases.

  In addition, if the background is left, there is a problem that the amount of color material (toner) used at the time of printing an electronic document also increases.

  Furthermore, when the background is left, the reproducibility of a portion brighter than the background is enhanced, so that the background portion looks dark and tends to be difficult to see as a whole.

  SUMMARY An advantage of some aspects of the invention is that it provides an image processing apparatus and an image processing method capable of performing appropriate image processing on an image read from a document.

  Therefore, in order to solve the above problems, the present invention is an image processing apparatus that processes an image read from a document, and a density detection unit that detects a density and a minimum density of the background of the document in the image, A background correction unit that performs background correction on the image, and the background correction unit does not remove the background according to a comparison between the density of the background and the minimum density.

  Such an image processing apparatus can perform appropriate image processing on an image read from a document.

  ADVANTAGE OF THE INVENTION According to this invention, the image processing apparatus and image processing method which can perform an appropriate image process with respect to the image read from the original document can be provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a hardware configuration example of an image forming apparatus according to an embodiment of the present invention. In FIG. 1, an image forming apparatus 10 includes a reading device 101, an image data processing device 102, a bus control device 103, an image data processing device 104, an HDD 105, a CPU 106, a memory 107, a plotter I / F device 108, a plotter device 109, and an operation. A panel 110, a line I / F device 111, a network / F device 112, a south bridge 113, a ROM 114, and the like are provided.

  The reading apparatus 101 generates and outputs, for example, RGB 8-bit digital image data (hereinafter simply referred to as “image data”) from the density information of the original obtained by scanning the original (paper document).

  The image data processing device 102 performs image processing described later on the image data input from the reading device 101 and outputs the processed image data.

  The bus control device 103 controls exchange (communication) of various data such as image data and control commands in the image forming apparatus 10 via a bus.

  The image data processing device 104 performs image processing described later on the image data processed by the image data processing device 102 and outputs the processed image data.

  An HDD (Hard Disk Drive) 105 is a storage device for accumulating image data and accompanying information of the image data.

  The CPU 106 is a microprocessor that controls the operation of each hardware of the image forming apparatus 10.

  The memory 107 is a volatile memory used for temporarily storing a program for causing the CPU 106 to control the image forming apparatus 10, intermediate processing data, and the like.

  The plotter I / F device 108 receives CMYK image data sent from the CPU 106 and outputs it to the dedicated I / F of the plotter device 109.

  Upon receiving the CMYK image data, the plotter device 109 outputs (prints) an image on a transfer paper (printing paper) by an electrophotographic process using a laser beam.

  The South Bridge 113 is a general-purpose circuit having a bus bridge function.

  The ROM 114 is a storage device that stores a program for causing the CPU 106 to control the image forming apparatus 10.

  The operation panel 110 is a part that functions as an interface between the image forming apparatus 10 and the user. The operation panel 110 includes an LCD (Liquid Crystal Display) and key switches, and displays various states and operation methods of the image forming apparatus 10 on the LCD. Or detecting a key switch operation by the user.

  The line I / F device 111 is an interface for connecting to a telephone line. The image forming apparatus 10 exchanges image data with a fax machine 50 outside the apparatus via a telephone line.

  The network / F device 112 is an interface for connecting to a network such as a LAN (Local Area Network) or the Internet. The image forming apparatus 10 performs various communications and input / output of image data via a network with a PC (Personal Computer) 60 or the like outside the apparatus.

  FIG. 2 is a diagram illustrating a functional configuration example of the image data processing apparatus 102 according to the first embodiment. 2, the image data processing apparatus 102 includes a γ conversion unit 21, a gradation correction unit 22, a filter processing unit 23, a background correction unit 24, a color conversion unit 25, a compression processing unit 26, a density detection unit 27, and the like. .

  The γ conversion unit 21 performs γ (gamma) conversion on RGB 8-bit read image data, and converts gradation characteristics from reflectance linearity to density linearity.

  The density detection unit 27 detects the background density and the minimum density of the RGB image data output from the γ conversion unit 21. A known technique may be used to detect the background density and the minimum density.

  FIG. 3 is a diagram for explaining a method for detecting the background density and the minimum density. In FIG. 3, the horizontal axis indicates the image density (the density increases as it goes to the right), and the vertical axis indicates the density frequency (number of appearances) (the frequency increases as it goes up (dark)). That is, in FIG. 3, a histogram of the image of the image data (image density distribution) is taken, the most frequent density (Tr) is the background density, and the lowest (bright) density that forms the frequency peak. Concentration (Tm) is the minimum concentration. FIG. 3 shows an example in which two peaks are formed at the concentration Tr and the concentration Tm. However, when a peak is formed at a concentration lower than the concentration Tm, the lower concentration is the lowest concentration. It is said.

  The gradation correction unit 22 performs gradation correction based on density value information (information indicating the background density and the minimum density) detected by the density detection unit 27. For example, when the minimum density is the background density (when Tr = Tm in FIG. 3), the gradation correction unit 22 does not perform any special correction. This is because there is a low possibility that a correction mark exists. On the other hand, when the minimum density is lower than the background density (in the case of Tm <Tr in FIG. 3), the gradation correction unit 22 performs the gradation of the highlight portion with respect to the image data output from the γ conversion unit 21. Gradation correction is performed so that a large amount of gradation is taken (that is, more gradation is assigned to a relatively low density region (part) than to a high density region). In this case, it can be considered that the minimum density is due to the presence of the correction mark, and the correction mark and the background etc. can be obtained by relatively increasing the number of gradations in the low density area where the correction mark can exist. This is because the contrast can be enhanced.

  The filter processing unit 23 performs filter processing such as emphasizing the character portion and smoothing the pattern portion.

  The background correction unit 24 performs background correction processing on the image data subjected to the filter processing by the filter processing unit 23 based on density information detected by the density detection unit 27. Details of the background correction processing will be described later.

  The color conversion unit 25 converts the image data on which the background correction has been performed by the background correction unit 24 into image data in a color space having a predetermined characteristic. This color space is, for example, an Adobe (registered trademark) -RGB space defined by Adobe (registered trademark).

  The compression processing unit 26 compresses the image data output from the color conversion unit 25. The compression processing unit 26 performs lossless compression such as a PNG (Portable Network Graphics) format, and stores the compressed image data in the HDD 105. The image data stored in the HDD 105 is subjected to image processing by the image data processing device 104.

  FIG. 4 is a diagram illustrating a functional configuration example of the image data processing device 104 according to the first embodiment. In FIG. 4, the image data processing apparatus 104 includes an expansion processing unit 41, a color conversion unit 42, a scaling processing unit 43, a compression processing unit 44, and the like.

  The decompression processing unit 41 decompresses image data output from the image data processing apparatus 102 and stored in the HDD 105.

  The color converter 42 converts the color space of the image data according to the output application. For example, if it is used for distribution, it is converted to a color space of sRGB (standard RGB), and when it is output by the plotter device 109, it is converted to a color space of CMYK.

  The scaling processing unit 43 performs scaling processing (enlargement or reduction) of the image data according to the setting content designated by the user via the operation panel 110.

  The compression processing unit 44 performs compression processing according to the setting content designated by the user via the operation panel 110, and outputs compressed image data. For example, a compression format such as JPEG or PDF is used.

  When the image data generated by the image data processing device 104 is used for distribution, the image data is distributed to an external PC or the like via the network / F device 112. The PC adds an electronic signature or a time stamp to the image data as necessary. When the image data is used for printing, the image data is output to the plotter device 109.

  Next, the background correction processing by the background correction unit 24 of the image data processing apparatus 102 will be described in detail. Based on the background density (referred to as “Tr”) detected by the density detection unit 27 and the minimum density (referred to as “Tm”), the background correction unit 24 sets Tr−Tm> α (α is defined as “Tr”). In the case of a predetermined threshold value (a positive integer)) (that is, assuming that the minimum density Tm is regarded as the density of the correction mark, if the difference between the background density and the correction mark density is greater than α), the background is left (not removed). ), And when Tr−Tm ≦ α (that is, when the minimum density Tm is regarded as the density of the correction mark, the difference between the background density and the density of the correction mark is equal to or less than α), a predetermined correction is made to Tr The background correction is performed by multiplying the coefficient εβ (appropriate value of 0 <εβ <1) and the density indicated by Tr × εβ (that is, the density lower than the background density) as a reference (the background correction is “background correction A”). That said.) This is a correction for not removing the background completely.

  FIG. 5 is a diagram for explaining the background processing when the difference between the minimum density and the background density is smaller than a predetermined threshold. In FIG. 5, the horizontal axis indicates the density value of the 8-bit RGB input image (that is, the image before the background correction process) to the background correction unit 24. The vertical axis represents the density value of the output image from the 8-bit RGB background correction unit 24 (that is, the image after the background correction processing). In both the horizontal and vertical axes, 0 corresponds to a highlight portion (bright portion), and 255 corresponds to a shadow portion (dark portion). That is, the larger the value, the higher the density.

  In the figure, Th is an input value corresponding to Tr × εβ. As shown in the figure, the background correction unit 24 performs background correction so that the output value becomes the minimum 0 for input values equal to or less than Th. This indicates that all regions having a density of Tr × εβ or less (that is, regions having a density lower than the background) are all expressed as the same (lowest) density region. Therefore, it is possible to avoid the removal of the background and correction marks. However, since the background correction is performed to such an extent that the background is not removed, it is possible to output an overall bright and easy-to-view image rather than using the input value as the output value as it is.

  As described above, the image forming apparatus 10 according to the first embodiment automatically executes appropriate background processing that leaves an area having a lower density than the background density according to the density information of the image. Accordingly, it is possible to automatically perform background processing that can leave a correction mark.

  Further, the user can obtain an image with a predetermined image quality and cost consciousness without requiring complicated settings when digitizing a paper document.

  Next, a second embodiment will be described. In the second embodiment, the background correction processing by the background correction unit 24 of FIG. 2 is different from the first embodiment. Other points may be the same as in the first embodiment.

  In the second embodiment, the background correction unit 24 leaves (does not remove) the background if Tr−Tm> α based on the background density Tr and the minimum density Tm detected by the density detection unit 27. When Tr−Tm ≦ α, the ground is corrected by multiplying Tr by a predetermined correction coefficient εβ (however, an appropriate value of εβ> = 1) and using the density indicated by Tr × εβ as a reference (such as The background correction is referred to as “background correction B”).

  That is, in the second embodiment, the value that the correction coefficient εβ can take is different from that in the first embodiment. Setting εβ to a value of 1 or more means performing background correction so as to remove at least the background. That is, in this case, in FIG. 5, Th (input value corresponding to Tr × εβ) is the same density as the background of the paper or an input value having a higher density than the density of the background of the paper. It becomes the lowest level value. Therefore, according to the background correction B, unnecessary information such as paper stains is deleted, and the image quality of the output image is improved. In addition, since the uniform density region increases in the output image, continuous data increases, and as a result, the compression rate of the output image is improved.

  Incidentally, “normal mode” and “special mode” may be provided as processing modes of the image data processing apparatus 102. The processing mode may be switched as desired by the user via the operation panel 110.

  In the “normal mode”, the density detection unit 27 does not operate, undergoes filter processing, the background correction unit 24 performs background correction at a constant (predetermined) background level, and the color conversion unit 25 and the compression processing unit 26 are After that, RGB image data is output. That is, in the “normal mode”, the input image data read by the reading device 101 is not conscious of the presence or absence of a density region lower than the background. Therefore, the “normal mode” is suitable for image data of general documents such as characters and photographs.

  On the other hand, the “special mode” is a processing mode suitable for highly important documents that can be stored electronically by the e-document method, such as national tax documents. The “special mode” has “high” and “low” as security levels. The security level is selected by the user via the operation panel 110 according to the importance of the document, for example. When “high” is selected as the security level, the background correction unit 24 performs background correction processing using the background correction A. On the other hand, when “low” is selected as the security level, the background correction unit 24 performs the background correction processing by the background correction B.

  Therefore, for example, if you want to obtain an image that is faithful to the original without leaking information such as paper tears, wrinkles, and dirt, regardless of the presence or absence of correction marks, select security level `` High '' and if there are correction marks If you do not want to be able to check, the state at the general document level is sufficient, but if you want to emphasize the storage cost, you should select the security level “low”.

  According to the second embodiment, the user does not need to individually separate a large amount of paper documents or perform complicated settings when digitizing, and can obtain an image with the requested image quality and save it. Cost benefits.

  Next, a third embodiment will be described. In the third embodiment, the configuration of the image data processing apparatus 102 is different from that of the first embodiment. In the third embodiment, the image data processing apparatus 102 is referred to as “image data processing apparatus 102a”. Other points may be the same as in the first embodiment.

  FIG. 6 is a diagram illustrating a functional configuration example of the image data processing apparatus 102a according to the third embodiment. 6, the image data processing apparatus 102a includes a γ conversion unit 31, a background density detection unit 32, a low density region determination unit 33, a gradation correction unit 34, a filter processing unit 35, a background correction unit 36, a color conversion unit 37, And a compression processing unit 38 and the like. Among these, the γ conversion unit 31, the filter processing unit 35, the color conversion unit 37, and the compression processing unit 38 may be the same as the γ conversion unit 21, the filter processing unit 23, the color conversion unit 25, and the compression processing unit 26 in FIG. Therefore, the description is omitted.

The background density detection unit 32 detects the background density of the RGB image data output from the γ conversion unit 31. The detection method may be the same as in the first embodiment. The low concentration region determination unit 33 determines whether or not there is a low concentration region (region having a lower concentration than the background) based on the background concentration value detected by the background concentration detection unit 32. judge. The determination method will be described later. The “region” refers to an area having a certain extent in the image and composed of a plurality of pixels.

  The gradation correction unit 34 performs gradation correction on the RGB image data output from the γ conversion unit 31 according to the determination result of the presence or absence of the low density region of the image by the low density region determination unit 33. For example, the gradation correction unit 34 does not perform special correction when it is determined that there is no low density region, and takes a large gradation in the highlight portion of the image data when it is determined that there is a low density region. Gradation correction is performed as described above. In this case, it can be considered that the low density area is due to the presence of the correction mark, and the correction mark and the background, etc. are increased by relatively increasing the number of gradations of the low density area where the correction mark can exist. This is because the contrast can be enhanced.

  The background correction unit 36 performs background correction processing based on the determination result of the presence or absence of the low density region by the low density region determination unit 33 and the background density Tr detected by the background density detection unit 32. If it is determined that there is a low density region, the background correction unit 36 leaves the background so that the density Tr value is not saturated according to the Tr value. When it is determined that there is no low density region, the Tr value is multiplied by a predetermined correction coefficient εβ (appropriate value of 0 <εβ <1), and the background correction is performed using the density indicated by Tr × εβ as a reference. This is a correction for not removing the background completely. The background correction by the background correction unit 36 in the third embodiment is referred to as “background correction C”.

Next, a method for determining the presence / absence of a low concentration region by the low concentration region determination unit 33 will be described in detail.
The low density area determination unit 33 divides the image of the image data into a plurality of blocks composed of a plurality of images, and determines the presence or absence of the low density area in units of blocks. For example, for RGB image data output from the highly accurate γ converter 31, the image is divided into arbitrary blocks, and the histogram for each block is compared with the background density detected by the background density detector 32. The presence or absence of a lower density region (low density region) is determined. If the density value of the background of the document is equal to the density value of the low density area, the density becomes the background density, so it is determined that there is no low density area of the image. In this case, the minimum density is set to the background density. Therefore, there is no loss of images because the background is left behind.

  As an example, a specific method for determining the presence or absence of a low density region by the low density region determination unit 33 will be described. First, the image of the image data is divided into a plurality of blocks (divided areas) each composed of n × m (n and m are arbitrary integers) pixels. In the following, one block is assumed to be an n-by-m matrix P, the density value of the n-by-m pixel is expressed by p_ {n, m}, and the n-by-m column of the matrix Q having the same size as the matrix P An element value is represented by q_ {n, m}.

  Subsequently, the background density Tr is compared with p_ {k, l}. If Tr> p_ {k, l}, the value of q_ {k, l} is set to 1. If Tr ≦ p_ {k, l}, the value of q_ {k, l} is set to 0 [k: 1, 2,... N, l: 1, 2,.

In this case, in the block, the frequency of the portion (pixel) having a lower density than the background density is
Σq_ {k, l}
Is calculated by

The average density value in the block is
Σp_ {h, i} × q_ {h, i} / Σq_ {k, l}
[H = 1, 2,..., N i = 1, 2,.

  As described above, the statistic of the entire image is calculated based on the frequency and average density value of the low density portion in each block, and the presence / absence of the low density region is determined based on the statistic.

  When the number of blocks is z, the frequency of an arbitrary block is H_b, and the average density value is I_b [b: 1, 2,..., Z], Σ {H_b × (Tr−I_b)} is the statistics of the entire image. Amount.

  This is based on the assumption that the density near the background density may include the measurement error part of the RGB image data read from the reading device 101, and that the lower density part is weighted. It is a statistic.

  When satisfying Σ {H_b × (Tr−I_b)}> εα> 0 (εα is a threshold determined according to the environment such as the performance of the reading apparatus 101), it is determined that there is a low concentration region, and when not satisfied, It is determined that there is no low concentration area.

  According to the third embodiment, appropriate background processing can be automatically performed according to the determination of the presence or absence of the low-density region. Therefore, it is possible to accurately perform background processing that can leave a correction mark.

  Further, the user can obtain an image with a predetermined image quality and cost awareness without having to make complicated settings when digitizing a paper document.

  Next, a fourth embodiment will be described. In the fourth embodiment, the background correction processing by the background correction unit 36 of FIG. 6 is different from the third embodiment. Other points may be the same as in the third embodiment.

  In the fourth embodiment, the background correction unit 36 performs background correction according to the background density Tr detected by the background density detection unit 32 and the determination result of the presence or absence of the low density region of the image by the low density region determination unit 33. Do. That is, when it is determined that there is a low density region, a process for leaving the background is performed, and when it is determined that there is no low density region, Tr is multiplied by a predetermined correction coefficient εβ (an appropriate value of εβ> = 1). Then, the background correction is performed based on the density indicated by Tr × εβ (such background correction is referred to as “background correction D”). That is, in the fourth embodiment, the possible value of the correction coefficient εβ is different from that in the third embodiment. Setting εβ to a value of 1 or more means performing background correction so as to remove at least the background.

  Incidentally, “normal mode” and “special mode” may be provided as processing modes of the image data processing apparatus 102a. The processing mode may be switched as desired by the user via the operation panel 110.

  In the “normal mode”, the low density area determination unit and the background density detection unit 32 do not operate, and the background correction unit 36 performs background correction at a constant background level through the gradation correction unit 34 and the filter processing unit 35. That is, in the “normal mode”, the input image data read by the reading device 101 is not conscious of the presence or absence of a density region lower than the background. Therefore, the “normal mode” is suitable for image data of general documents such as characters and photographs.

  On the other hand, the “special mode” is a processing mode suitable for highly important documents that can be stored electronically by the e-document method, such as national tax documents. The “special mode” has “high” and “low” as security levels. The significance of the security level is the same as in the second embodiment.

  When “high” is selected as the security level, the background correction unit 36 performs background correction processing using the background correction C. On the other hand, when “low” is selected as the security level, the background correction unit 36 performs the background correction processing by the background correction D.

  According to the fourth embodiment, the user does not need to individually separate a large amount of paper documents or perform complicated settings when digitizing, and obtains or saves an image with the requested image quality. Cost benefits.

  The processing by the image data processing apparatus 102 (102a) and the image data processing apparatus 104 may be executed by the CPU 106 based on a program recorded in the ROM 114.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to such specific embodiment, In the range of the summary of this invention described in the claim, various deformation | transformation・ Change is possible.

1 is a diagram illustrating an example of a hardware configuration of an image forming apparatus according to an embodiment of the present invention. It is a figure which shows the function structural example of the image data processing apparatus 102 in 1st embodiment. It is a figure for demonstrating the detection method of a background density | concentration and a minimum density | concentration. It is a figure which shows the function structural example of the image data processing apparatus 104 in 1st embodiment. It is a figure for demonstrating the background process in case the difference of minimum density and background density is smaller than a predetermined threshold value. It is a figure which shows the function structural example of the image data processing apparatus 102a in 3rd embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image forming apparatus 21, 31 γ conversion unit 22, 34 Tone correction unit 23, 35 Filter processing unit 24, 36 Background correction unit 25, 37 Color conversion unit 26, 38 Compression processing unit 27 Density detection unit 32 Background density detection unit 33 Low-density area determination unit 41 Decompression processing unit 42 Color conversion unit 43 Scaling processing unit 44 Compression processing unit 50 FAX device 101 Reading device 102 Image data processing device 103 Bus control device 104 Image data processing device 105 HDD
106 CPU
107 Memory 108 Plotter I / F Device 109 Plotter Device 110 Operation Panel 111 Line I / F Device 112 Network / F Device 113 South Bridge 114 ROM

Claims (16)

An image processing apparatus for processing an image read from a document,
A density detecting means for detecting the background density and the minimum density of the original in the image;
Background correction means for performing background correction on the image,
The image processing apparatus according to claim 1, wherein the background correction means does not remove the background according to a comparison between the density of the background and the minimum density.   The density detecting means detects the most frequent density in the density distribution of the image as the density of the background, and the lowest density among the densities forming a peak in the density distribution of the image is the lowest density. The image processing apparatus according to claim 1, wherein:   2. The background correction unit according to claim 1, wherein the background correction unit performs background correction on the image based on a density lower than the density of the background according to a comparison between the density of the background and the minimum density. 2. The image processing apparatus according to 2.   When the lowest density is lower than the background density, the image processing apparatus has gradation correction means for performing gradation correction on the image so that the number of gradations is increased in a relatively low density portion. The image processing apparatus according to any one of claims 1 to 3. An image processing apparatus for processing an image read from a document,
Low density area determination means for determining the presence or absence of an area having a density lower than the background density of the document in the image;
Background correction means for performing background correction on the image,
The image processing apparatus according to claim 1, wherein the background correction unit does not remove the background when it is determined that the low density region exists.   The low density region determination means compares the product of the frequency of pixels having a lower density than the background density and an average value of the density of the low density pixels in the image with a predetermined threshold value. 6. The image processing apparatus according to claim 5, wherein the presence or absence of an area is determined.   7. The background correction unit according to claim 5 or 6, wherein when it is determined that the density area does not exist, the background correction is performed on the image based on a density lower than the density of the background. The image processing apparatus described.   And a gradation correction unit that performs gradation correction on the image so that the number of gradations is increased in a relatively low density portion when it is determined that the density region does not exist. The image processing apparatus according to claim 5. An image processing method by an image processing apparatus for processing an image read from an original,
A density detection procedure for detecting a background density and a minimum density of the document in the image;
A background correction procedure for performing background correction on the image,
The image processing method according to claim 1, wherein the background correction procedure does not remove the background according to a comparison between the density of the background and the minimum density.   In the density detection procedure, the most frequently occurring density in the density distribution of the image is detected as the density of the background, and the lowest density among the densities that form a peak in the density distribution of the image is the lowest density. The image processing method according to claim 9, wherein:   10. The background correction procedure according to claim 9, wherein the background correction is performed on the image based on a density lower than the density of the background according to a comparison between the density of the background and the minimum density. The image processing method according to 10.   A gradation correction procedure for performing gradation correction on the image so that the number of gradations is increased at a relatively low density portion when the lowest density is lower than the background density; An image processing method according to any one of claims 9 to 11. An image processing apparatus for processing an image read from a document,
A low density region determination procedure for determining the presence or absence of a region having a density lower than the background density of the document in the image;
A background correction procedure for performing background correction on the image,
The image processing method according to claim 1, wherein the background correction procedure does not remove the background when it is determined that the low density region exists.   The low-density area determination procedure compares the frequency of pixels having a lower density than the density of the background in the image with the average value of the density of the low-density pixels by comparing the product with the predetermined threshold value. The image processing method according to claim 13, wherein the presence or absence of an area is determined.   15. The background correction process according to claim 13 or 14, wherein when it is determined that the density region does not exist, the background correction is performed on the image based on a density lower than the density of the background. The image processing method as described.   A gradation correction procedure for performing gradation correction on the image so that the number of gradations is increased in a portion having a relatively low density when it is determined that the density region does not exist; The image processing method according to claim 13.

Images