JP4787776B2 - Image processing apparatus, image forming apparatus including the same, and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus having a reading device that reads both the front surface and the back surface of an original, and in particular, an image processing apparatus that detects characters or patterns in image data and performs appropriate image processing, and the image. The present invention relates to an image forming apparatus including a processing device.

  Conventionally, a scanned original image is separated into image areas and classified into a plurality of image areas, for example, a character area, a picture area, and a halftone dot area, and image quality is improved by executing appropriate image processing according to the classification. Various image processing techniques have been proposed.

  As a conventional technique for improving the image quality of the front and back surfaces by reading images formed on the front surface and the back surface of the document, Japanese Patent Application Laid-Open No. 2005-133867 discloses that post-processing of the read front surface image data and back surface image data is the same. It has been proposed to reduce the difference in image quality between the front and back image data by applying it in the processing unit.

  In Patent Document 2, a CIS (Contact Image Sensor) that directly reads an image with a linear sensor and a CCD image sensor are used to read images on both sides of the document by conveying the document to the conveyance path once. Is disclosed. According to Patent Document 2, it is shown that interpolation of image data in a gap of a line sensor configured by arranging a plurality of light receiving chips like CIS can be performed accurately.

Patent Document 2 discloses an image processing technique for improving image quality by separating an image area of a read original image to facilitate separation of small characters and halftone dots.
JP 2006-13924 A JP 2005-217467 A JP 2003-46772 A

  2. Description of the Related Art Conventionally, an image processing apparatus that reads an original image, makes a determination based on the characteristics of the image data, and performs image area separation that classifies the image area into a plurality of image areas is known, for example, in Patent Document 3 described above. In such a conventional image processing apparatus, by switching the appropriate image processing for each pixel based on the result of the image area separation, the image quality can be improved in a document in which characters / pictures, chromatic / achromatic, etc. are mixed. Improvements can be made.

  However, when the original to be read is a double-sided document, that is, when information is drawn on both the front and back sides, when separating the image area on one side, there is an error due to the influence of the show-back from the other side. Judgment may occur. For example, let us consider a case where characters are written on the front side of a document as shown in FIG. 10A and the back side of the document has a high-density pattern as shown in FIG. 10B. At this time, the image of the surface read by the image reading device is affected by the show-back and becomes as shown in FIG. Although only characters are actually drawn on the front surface, an image influenced by the pattern on the back surface is also read. As a result, an area that is originally determined as a character on the surface may be erroneously determined as a picture. The degree of influence on the back surface increases as the paper thickness decreases and the back pattern density increases.

  Thus, when one side of a document image is read, what is read at medium to low density may be the effect of show-through, but is it read by the effect of show-through? Or it cannot be judged whether it originally exists on the surface. In particular, when the paper is thin, the influence of the show-through is large and the possibility of erroneous determination is increased.

  The above-mentioned patent document 1 does not use the data on the back surface when correcting the data on the front surface, and the image processing shown in the patent document 2 also performs the interpolation of the image data using both the data on the front and back surfaces. In addition, Patent Document 3 uses information on only one side of an original when performing image area separation, and lacks consideration of the influence of the above-described offset.

  The present invention reduces image area separation misjudgment due to the influence of show-through and reduces image quality by using both image information on one side and the other side when performing image area separation on one side of a document. The main purpose is to provide simple images.

In order to solve the above problems, the present invention mainly adopts the following configuration.
An image reading device that reads both one surface of the document and the other surface, and an image area separation device that classifies the read document image into a plurality of image regions, and the read document image includes: An image of a double-sided document, on which information to be read is drawn on one side and the other side of the document, and the plurality of image areas classified by the image area separation device are at least a character area and a picture area Including
When the image area separation device performs image area separation on one side of a document, the character area is determined by using both read image information on the one side and image information on the other side. A white area extraction unit that determines whether the white area of the document is based on density information read from the images of the one side and the other side, and the white area extraction unit When it is determined that there is a character area, the image area separating apparatus determines that the white area of the one surface is extracted, and the density information of the other surface is used to extract the white area of the one surface. When changing the white area extraction criterion and changing the white area extraction criterion of the one surface, if the density of the other surface is higher than a predetermined value, the one surface is determined to be a white region. The white region extraction reference is changed so that it is easily performed .

Moreover, Te the image processing apparatus odor, said to change the white area extracted reference on one surface, a configuration in which change only when a specific intermediate density range of the concentration of the one surface.

  According to the present invention, when image area separation is performed on one (one) surface of an original, both image information on the surface and the other (opposite) surface is used, so that image region separation due to the effect of show-through is achieved. It is possible to provide a high-quality image. Furthermore, in particular, it is possible to reduce erroneous image area separation determination between characters and pictures.

  An image processing apparatus according to an embodiment of the present invention will be described in detail below with reference to FIGS. FIG. 1 is a block diagram showing a schematic configuration of a digital full-color image processing apparatus according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an image flow from the first image reading unit and the second image reading unit to the memory control unit through the image receiving unit and the image area separation preprocessing unit in the image processing apparatus according to the present embodiment. It is. FIG. 3 is a diagram showing a detailed configuration of the image area separation pre-processing unit in the image processing apparatus according to the present embodiment. FIG. 4 is a diagram showing 5 × 5 pixels referred to in the high density pattern matching unit of the image area separation preprocessing unit according to the present embodiment.

  FIG. 5 is a diagram showing a detailed configuration of the scanner correction unit in the image processing apparatus according to the present embodiment. FIG. 6 is a diagram showing a detailed configuration of the printer correction unit in the image processing apparatus according to the present embodiment. FIG. 7 is a diagram showing a detailed configuration of the image area separation unit in the scanner correction unit according to the present embodiment. FIG. 8 is a diagram showing a detailed configuration of the white area extraction unit in the image area separation unit according to the present embodiment. FIG. 9 is a block diagram showing another configuration example of the digital full-color image processing apparatus according to the embodiment of the present invention.

  In FIG. 1, when operating as a full-color copying machine, the first image reading unit 1 and the second image reading unit 2 read image data color-separated into R (red), G (green), and B (blue) from a document. The image data (analog signal) is converted into digital data and output. Here, each of the generated image data is input to the image data receiving unit 3.

  The received image data is input to the memory control unit 5 and the image area separation preprocessing unit 4. The image area separation pre-processing unit 4 generates data X1 and X2 to be referred to when image area separation is performed by an image area separation unit 65 (see FIG. 5) included in the scanner correction unit 6 described later (FIG. 2, FIG. 2). (See FIG. 3).

  In the memory control unit 5, the data X1 and X2 generated by the image area separation preprocessing unit 4 and the image data read by the first image reading unit 1 and the second image reading unit 2 are stored in the memory while being synchronized. The memory control unit 5 has a storage (not shown) for the purpose of handling a large amount of image data. The image data and the data X1 and X2 generated by the image area separation preprocessing unit 4 are stored in the storage, and the next input image is received or the image data stored for hard copy is stored in the scanner correction unit 6. Or send it to. At this time, image data storage processing and image data extraction from the storage are controlled using an image memory control (not shown). In the present embodiment, the image data from the first reading unit 1 is processed first, and then the image data from the second reading unit 2 is processed. When reading continuously, the image data read by the first image reading unit 1 and the image data read by the second image reading unit 2 are alternately sent to the scanner correction unit 6.

  As will be described later, the scanner correction unit 6 performs scanner γ correction processing on the RGB image data (digital data) read by the scanner and classifies the image area into characters, line drawings, patterns, etc. (image area separation). Depending on the determination result of image area separation, image processing such as emphasizing the character portion of the image and smoothing the image portion of the image is performed to correct the characteristics of the scanner.

  The compression processing unit 7 compresses the multi-valued image data after the scanner correction and sends the data to the general-purpose bus 8. The compressed image data is sent to the controller 9 through the general-purpose bus 8. The controller 9 has a semiconductor memory (not shown) and accumulates sent data. The accumulated data is written to a hard disk drive (HDD) 10 which is a large capacity storage device as needed. This is because paper is jammed at the time of printout, so that it is possible to avoid re-reading the original even if the output does not end normally, to perform electronic sorting to rearrange multiple original image data, and to store the read original This is to re-output when necessary. In this embodiment, the image data is compressed. However, if the bandwidth of the general-purpose bus 8 is sufficiently wide and the capacity of the HDD to be stored is large, the data may be handled in an uncompressed state.

  Next, the controller 9 sends the image data of the HDD 10 to the decompression processing unit 13 via the general-purpose bus 8. The decompression processing unit 13 decompresses the compressed image data to the original multi-value data and sends it to the printer correction unit 14. The printer correction unit 14 performs printer γ correction processing and gradation processing, and performs image correction processing, dithering processing, etc. according to the gradation characteristics of the plotter 15 and the gradation characteristics and image area separation results of the plotter 15. Data quantization is performed. The plotter 15 is a transfer paper printing unit using a laser beam writing process. The plotter 15 draws image data as a latent image on a photoconductor, and forms a copy image on the transfer paper after image formation / transfer processing with toner.

  When operating as a distribution scanner that distributes image data to a PC via a network, the processing flow is the same as when operating as a copier until compression processing is performed. Thereafter, an image is sent to the controller 9. The controller 9 performs format processing. In the format processing, general-purpose image format conversion to JPEG, TIFF, or BMP format is performed. Thereafter, the image data is distributed to the external PC terminal 12 via the NIC (network interface controller) 11.

  The first image reading unit 1 and the second image reading unit 2 shown in FIG. 1 can be realized, for example, by using a known document feeding device and image reading device as disclosed in Patent Document 2 above. If the apparatus shown in Patent Document 2 is adopted, CIS is simultaneously performed (meaning about the time of the same original conveyance, not the exact time coincidence) at the time of conveyance of the original for reading the first side of the original by the CCD image sensor. The second side of the document can be read with (Contact Image Sensor). That is, by using the CCD image sensor and the CIS, the images on both the front and back sides of the original can be read by one conveyance of the original to the conveyance path.

  In the present embodiment, the portion that reads the first surface of the document by the CCD image sensor corresponds to the first image reading portion 1, and the portion that reads the second surface using the CIS is the second image reading portion 2. It corresponds to. Here, CIS is a CIS (Contact Image) that uses an LED (Light Emitting Diode) having a small shape as a light source and directly reads an image with a linear sensor via a lens for the purpose of downsizing the apparatus. (Sensor). When reading a color image, a combination of LED light sources of three colors R (red), G (green), and B (blue) is used for the LED, and a set of three columns for RGB three colors is used as the line sensor unit. good.

  Next, an image flow from the first image reading unit 1 and the second image reading unit 2 to the memory control unit 5 through the image receiving unit 3 and the image area separation preprocessing unit 4 will be described with reference to FIG. To do. A full color RGB image is read from the first image reading unit 1 and the second image reading unit 2. Here, the data read by the first image reading unit 1 is (r1, g1, b1), and the image data read by the second image reading unit 2 is (r2, g2, b2). These image data enter the image receiving unit 3. This data is sent to the memory control unit 5, but is also sent to the image area separation preprocessing unit 4.

  In the image area separation pre-processing unit 4, data X1 and X2 that are referred to by the image area separation unit 65 (see FIG. 5) existing in the subsequent process are generated. Data generated based on the data read by the first image reading unit 1 is X1, and data generated based on the data read by the second image reading unit 2 is X2. The image area separation unit 65 refers to X2 when performing image area separation of the data read by the first image reading unit 1, and performs image area separation of the data read by the second image reading unit 2. Reference is made to X1. With these references, when image area separation is performed on one side (one side) of the document, information on the opposite side (the other side) can be referred to.

  Next, details of the image area separation preprocessing unit 4 will be described with reference to FIG. In the following description, it is assumed that the larger the numerical value of the image data, the darker the image. The image area separation preprocessing unit 4 includes a back surface binarization unit 41, a high density pattern matching unit 42, and an expansion unit 43.

  The back surface binarization unit 41 generates an active signal r_h1 when all the RGB signals (r1, g1, b1) are larger than the threshold value r_th1. Further, the active signal r_h2 is generated when all the RGB signals (r2, g2, b2) are larger than the threshold value r_th2. The high density pattern matching unit 42 refers to 5 × 5 pixels as shown in FIG. 4, and when all of the 5 × 5 pixels referred to by the threshold r_th1 are on (ON), the target pixel (FIG. 4). Active signal R_H1 is generated at the center pixel). When all the 5 × 5 pixels referred to by the threshold value r_th2 are on (ON), the active signal R_H2 is generated for the target pixel.

  When the active signal R_H1 is generated in the target pixel, the expansion unit 43 generates the active signal X1 in the range of 7 × 7 pixels centering on the active signal R_H1. Further, when the active signal R_H2 is generated in the target pixel, the active signal X2 is generated in the range of 7 × 7 pixels with the center as the active signal R_H2. X1 and X2 detect a high density area of the document. Since there is a high probability of show-through on the opposite surface (the other surface) of the high density portion, such an area is referred to by the image area separation unit 65 described later.

  Next, the configuration of the scanner correction unit 6 will be described below with reference to FIG. The scanner correction unit 6 includes an image area separation unit 65 that determines whether an image area (image area) of a document is a character area, a halftone dot area, a pattern area, or the like based on RGB image data input from the scanner, and a scanner. A scanner γ correction unit 61 that converts a digital value of RGB image data based on characteristics of the image into a digital value proportional to lightness, and a sharpening process or a smoothing process is applied to the image data based on the image area separation result. Filter processing unit 62 and image data color-separated into R, G, and B, C (cyan), M (magenta), Y (yellow), and K (Bk) (black) recording colors, which are different color spaces. It comprises a color correction unit 63 that converts color image data including information, and a scaling unit 64 that outputs the input image with the size in the main scanning direction enlarged or reduced. The details of the function and operation of the image area separation unit 65 will be described later.

  Next, the configuration of the printer correction unit 14 will be described with reference to FIG. The printer correction unit 14 includes a printer γ correction unit 71 that performs γ correction on the CMYK image data that has undergone the expansion processing unit 13 illustrated in FIG. 1 according to the frequency characteristics of the plotter, and a dither process and an error diffusion process. A gradation processing unit 72 that performs quantization, and outputs CMYK image data in which printer characteristics are corrected. This output data is sent to the plotter 15, and the plotter 15 creates an image based on the sent image data, and forms the image by transferring the created image onto a sheet.

  The printer γ correction unit 71 performs γ correction to add contrast when the determination result of the image area separation is a character determination area, and reproduces gradation when the determination result of the image area separation is a pattern. Γ correction that places importance on The gradation processing unit 72 performs halftone processing that places importance on character sharpness when the determination result of the image area separation is a character determination area, and when the determination result of the image area separation is a pattern, A halftone process that emphasizes smoothness is performed.

  Next, the image area separation unit 65 included in the scanner correction unit 6 shown in FIG. 5 will be described with reference to FIG. As shown in FIG. 7, the image area separation unit 65 includes a filter unit 81, an edge extraction unit 82, a halftone dot extraction unit 83, a white region extraction unit 84, a color determination unit 85, and an overall determination unit 86. About each component, except the white area extraction part 84, it is completely the same as what is disclosed by the said patent document 3 (Unexamined-Japanese-Patent No. 2003-46772) known.

  The filter unit 81 corrects the G image data generated by the scanner mainly for extracting the edge of the character. Since the data read by the scanner may be blurred due to the performance of the lens, an edge enhancement filter is applied. However, it is necessary not to emphasize the line pattern for gradation expression widely used in copying machines. The image data value processed by the filter unit 81 of the pixel of interest that is the current pixel to be processed is given to the edge extraction unit 82 and the white area extraction unit 84.

  The character region has many high-level density pixels (black pixels) and low-level density pixels (white pixels), and these black pixels and white pixels are continuous at the edge portion. The edge extraction unit 82 detects a character edge based on the continuity of such black pixels, white pixels, and white pixels.

  The halftone dot extracting unit 83 detects the pixels (halftone peak pixels) that form part of the halftone dots from the pixel density information in the two-dimensional local region having a predetermined size using the G image data. A halftone dot detector and a second halftone peak detector that detects halftone dots using B image data and has the same function as the first halftone peak detector (G image data in the first halftone peak detector) Because it is used, it reacts to most colors, but does not react to Y. Therefore, it aims to detect halftone dot peaks of Y using B image data) and 65 below 100 lines. A third halftone peak detector for detecting lines (newspaper halftone dots) or more, and a peak and valley net detected by either the first halftone peak detector or the second halftone peak detector. The point peak pixel is counted for each two-dimensional small area of a predetermined size and the count value is A first halftone dot area detector that determines a halftone dot area, a second halftone dot area detector that receives a detection result of the third halftone dot peak detector and determines a halftone dot area, And a temporary storage means for temporarily storing the determination result by the halftone dot area detection unit. The present embodiment is not limited to this configuration, and any configuration that extracts halftone dots may be used.

  The color determination unit 85 will be described. When detecting color (chromatic) pixels and black (achromatic) pixels in a document, there is a relative reading shift of R, G, and B due to sampling of each color image data and mechanical accuracy. . The color determination unit 85 for finding a chromatic color region converts the input data R, G, B into C, M, Y and W (white) signals for color determination in the hue division unit, and performs color pixel determination. To do.

  The white area extraction unit 84 will be described with reference to FIG. 8 to be described later. Broadly speaking, the white region extraction unit 84 includes a binarization unit and an RGB white extraction unit. The binarization unit includes image density data (G image data of the filter unit 81). 2) is binarized with a threshold value to generate a binarized white determination signal for generating white data. The RGB white extraction unit performs RGB white background detection, color pixel detection, and gray pixel detection to determine whether the image data is a white area or a gray area (medium density area).

  The overall determination unit 86 in the image area separation unit 65 includes a character determination unit, a character determination unit, an expansion processing unit, and a decoding unit. In the character determination unit, the result of the edge extraction unit 82 is an edge, and halftone extraction is performed. When the result of the part 83 is no halftone dot and the result of the white area extracting part 84 is a white area, it is determined that the character edge. Otherwise, it is determined as a non-character edge (a pattern or a character), and the result is output to the expansion processing unit. By performing the expansion process, the separation result is different and the character area is not thinned. The C / P signal finally output by the decoding unit is a character edge signal. When the C / P signal is 1, it is a character edge region. A B / C signal is output from the color determination unit 85 and is a signal that is referred to when determining whether the original is a color document or a monochrome document.

  The image area separation unit 65 generates an image area separation signal C / P signal and a B / C signal as outputs. The C / P signal is a 2-bit signal. When C / P = 0, the picture area, when C / P = 1, the character area, when C / P = 2, the halftone area, C / P = 3. In the case of, it means a low line number halftone dot region. Here, in order to be a character region, a white region is a necessary condition. The B / C signal is a 1-bit signal. When B / C = 0, it indicates an achromatic region, and when B / C = 1, it indicates a chromatic region.

  Next, the image area separation unit 65 relating to the present embodiment is different from the known image area separation unit disclosed in Patent Document 3 (Japanese Patent Laid-Open No. 2003-46772) in the internal configuration of the image area separation unit. A white region extracting unit 84 relating to the present embodiment, which is a certain white region extracting unit, will be described below with reference to FIG.

  The white area extraction unit 84 according to the present embodiment includes an RGB white background extraction unit 93, a color pixel detection unit 92, a gray pixel detection unit 91, a gray pattern matching unit 95, a gray expansion unit 96, a white determination unit 97, and a white pattern matching unit 98. , White pattern correction unit 99, white expansion unit 100, white contraction unit 101, white correction unit 102, determination unit 103, and binarization unit 95. In this configuration, the components other than the binarization unit 94 and the RGB white background extraction unit 93 are the same as those disclosed in Patent Document 3. The functions and operations of the above-described components of the white area extraction unit 84 related to the present embodiment are as described in Patent Document 3 except for the binarization unit 94 and the RGB white background extraction unit 93. The outline will be described below.

  R, G, and B image data are input to each component of the RGB white extraction unit that includes the gray pixel detection unit 91, the color pixel detection unit 92, and the RGB white background extraction unit 93 shown in FIG. The output of the gray pixel detection unit 91 is input to the gray pattern matching unit 95, the pattern matching result of the gray pattern matching is input to the gray expansion unit 96, and is input to the determination unit 103 after performing expansion processing. In addition, the outputs of the color pixel detection unit 92, the RGB white background extraction unit 93, and the binarization unit 94 are input to the white determination unit 97, and the determination result of the white determination unit 97 is input to the white pattern matching unit 98. The matching result is input to the white pattern correction unit 99 and the white correction unit 102. The correction result of the white pattern correction unit 99 is input to the determination unit 103 after being processed by the white expansion unit 100 and the white contraction unit 101, and the processing result of the white correction unit 102 is input to the determination unit 103 as it is. The color pixel detection unit 92 detects a color pixel (color background) so as not to determine a light color as a white background.

  The white pattern matching unit 98, the white pattern correction unit 99, the white expansion unit 100, the white contraction unit 101, and the white correction unit 102 are configured to detect a border region that is not white and white. The output indicates the line width, the output of the white contraction portion 101 indicates a white area, and the output of the gray expansion portion 96 indicates medium density. The determination unit 103 determines the priority by assigning priority to these three outputs, and outputs the determination result to the subsequent stage. In this case, priority order 1 is line width information from the white correction unit 102, priority order 2 is medium density information from the gray expansion unit 96, and priority order 3 is white area information from the white shrinking unit 101. .

  The basic function of the binarization unit 94 is to binarize the edge emphasis output of the image density data (G image data) of the filter unit 81 with a threshold value thwsb, and generate white data that the white pattern matching unit 98 refers to A binarized white determination signal is generated for Here, the edge emphasis output is, for example, 256 gradations from 0 (white without density) to 255, and when the threshold thwsb is 50, if the value of the edge emphasis output is smaller than the threshold thwsb = 50, the binarization unit 94 determines “binarized white” and generates a binarized white determination signal “1”.

  In the binarization unit 94 in the present embodiment, the value of the threshold thwsb is switched depending on whether the X1 or X2 signal is active. That is, the RGB image (r1, g1, b1) read by the first image reading unit 1 is sent in synchronization with the X2 signal generated by the image area separation preprocessing unit 4, but the X2 signal is active. Increases the value of thwsb than when it is not active. The larger the value of thwsb is, the easier it is to determine as a white region. Therefore, when the X2 signal is active, that is, the image read by the second image reading unit 2 has a high density, the first image reading unit When there is a high possibility of having an influence on the image 1 as a show-through image, an area that should originally be extracted as a white area is easily determined as a white area. Therefore, it is possible to reduce the possibility that characters on the white ground will not be determined due to the influence of the show-through. For the RGB image (r2, g2, b2) read by the second image reading unit 2, when the X1 signal is active, the threshold value thwsb is set higher than when the X1 signal is not active.

  The basic function of the RGB white background extraction unit 93 is to activate a white background separation operation by extracting a white background region from R, G, and B image data. That is, the white background separation process is started. Specifically, if all of the R, G, and B image data of the pixel matrix are smaller than the threshold value thwss, the target pixel that is the central pixel of the pixel matrix is determined as a white region, and the white pattern matching unit 98 is activated. This is to extract whether there is a certain extent of white pixel region. Here, each of the R, G, and B image data has 256 gradations from 0 to 255, and the threshold value thwss <the threshold value thwsb, where the threshold value thwss is 40, all of the R, G, and B image data are threshold values. If it is smaller than thwss = 40, it is determined as “white background” and a white background determination signal “1” is generated. When any of the R, G, and B image data is greater than or equal to the threshold thwss = 40, it is determined as “white background” and a white background determination signal “0” is generated.

  In the RGB white background extraction unit 93 in the present embodiment, the value of the threshold thwss is switched depending on whether the X1 or X2 signal is active. That is, the RGB image (r1, g1, b1) read by the first image reading unit 1 is sent in synchronization with the X2 signal generated by the image area separation preprocessing unit 4, but the X2 signal is active. Increases the value of thwss than when it is not active. The larger the value of thwss, the more easily it is determined as a white region. Therefore, when the X2 signal is active, that is, the image read by the second image reading unit 2 has a high density, the first image reading unit When there is a high possibility of affecting the image 1 as a show-through image, an area that should originally be extracted as a white area is easily determined as a white area. Therefore, it is possible to reduce the possibility that characters on the white ground will not be determined due to the influence of the show-through. For the RGB image (r2, g2, b2) read by the second image reading unit 2, the threshold value thwss is set higher when the X1 signal is active than when it is not active.

  Next, an image processing apparatus according to another embodiment of the present invention will be described below. In the image processing apparatus according to another embodiment of the present invention, in addition to the X signal, the value of the G signal is also added to the condition as a condition for switching the value of the threshold thwsb in the binarization unit 94 of the white region extraction unit 84. . That is, the threshold value is higher than that when the X2 signal is active for the RGB image (r1, g1, b1) read by the first image reading unit 1 and when LowG1 <g1 <UpG1. Increase the value of thwsb. Here, LowG1 and UpG1 are constants given in advance as parameters.

  In general, in an image that is going to be image-region-separated, an erroneous determination due to the effect of show-through occurs because there is a dark pattern on the opposite side of the document, and an intermediate density portion occurs on the front side due to the show-through Is the case. Switching the threshold value thwsb in accordance with the density on the opposite surface even when not in the density range is considered to have an adverse effect due to side effects. Therefore, there is an effect of reducing erroneous determination by switching the value of the threshold value thwsb only within the density range in which the surface on which the image area separation is performed may be the influence of the show-through.

  In addition, for the RGB image (r2, g2, b2) read by the second image reading unit 2, only when the X1 signal is active and when LowG2 <g2 <UpG2 is satisfied than when it is not. The threshold value thwsb is increased. Here, LowG2 and UpG2 are constants given in advance as parameters.

  Similarly, in the RGB white background extraction unit 93 of the white region extraction unit 84, when the X2 signal is active for the RGB image (r1, g1, b1) read by the first image reading unit 1, and LowG1 Only when <g1 <UpG1, the value of the threshold thwsb is set higher than when it is not. In addition, for the RGB image (r2, g2, b2) read by the second image reading unit 2, only when the X1 signal is active and when LowG2 <g2 <UpG2 is satisfied than when it is not. The threshold value thwss is increased.

  In the embodiment of the present invention described above, the second surface of the document is read by the CIS at the same time when the document is transported to read the first surface of the document by the CCD image sensor. That is, an example is given in which images on both the front and back sides of a document are read by a single conveyance of a document to a conveyance path using a CCD image sensor and a CIS. However, embodiments of the present invention are not necessarily limited to such a case. The invention is not limited to using two reading means.

  An image processing apparatus according to an embodiment of the present invention employs a document feeding apparatus and an image reading apparatus disclosed in Patent Document 2 (Japanese Patent Laid-Open No. 2005-217467) and can be read by a switchback type image reading apparatus. For example, both the front and back sides of the document can be read only by the CCD image sensor. A method for reading both sides of a double-sided document with a switchback type image reading apparatus is also disclosed in the above-mentioned Patent Document 2. FIG. 9 shows a schematic configuration of a full-color image processing apparatus when this reading method is adopted. Show.

  FIG. 9 is a block diagram showing another configuration example of the digital full-color image processing apparatus according to the embodiment of the present invention. In another configuration example of the present embodiment illustrated in FIG. 9, the scanner correction unit 113 and the subsequent components have the same configuration as that illustrated in FIG. 1, but the processing method up to the scanner correction unit 113 is different. is doing. That is, the double-sided image reading unit 110 is a switchback type image reading device, and alternately reads both sides of a double-sided document, and sends RGB digital image data to the image area separation preprocessing unit 111. The image area separation pre-processing unit 111 receives image data of one side and performs the same processing as the configuration shown in FIG. Thereafter, the X data is output to the memory control unit 112. When the X data is active, it indicates that the document has a high density.

  When the switchback type image reading apparatus 110 is used as in the configuration example of the present embodiment, the X data on the second surface is synchronized with the RGB data on the first surface of the document and output to the scanner correction unit 113. For this reason, the RGB data and X data of the first surface are temporarily stored in a storage (not shown). Thereafter, RGB data relating to the second side of the document is read, and X data of the second side is generated by the image area separation preprocessing unit 111 and output to the memory control unit 112. The memory control unit 112 uses image memory control (not shown) to control image data accumulation processing and image data extraction from the storage.

  Therefore, the RGB data of the first surface and the X data of the second surface are synchronized and output to the scanner correction unit 113. Subsequently, the RGB data of the second surface and the X data of the first surface are synchronized and output to the scanner correction unit 113. Since the output to the scanner correction unit 113 is performed sequentially, the processing after the scanner correction unit 113 is the same as that shown in FIG. In the configuration example of the present embodiment shown in FIG. 9, the output to the scanner correction unit 113 is sequential, but two scanner correction units 113 may be provided to perform processing in parallel.

  In addition, the present invention mainly reduces image area separation misjudgment due to the effect of show-through using both image information on one surface and the other surface when performing image region separation on one surface of a document. Although this is an issue, such a problem is that a storage medium or recording medium in which a program code of software having the functions of the present embodiment is recorded is supplied to the system or apparatus, and the computer of the system or apparatus stores the storage medium in the storage medium. It is also achieved by reading and executing the programmed program code. In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the embodiment of the present invention. Further, by executing the program code read by the computer, not only the functions of the above-described embodiment are realized, but also the operating system running on the computer based on the instruction of the program code is actually A case where the function of this embodiment described above is realized by performing part or all of the processing and the processing is also included in the embodiment of the present invention.

  Furthermore, after the program code read from the storage medium is written into a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. This embodiment includes a case where the CPU or the like provided in the expansion card or the function expansion unit performs part or all of the actual processing and the functions of the present embodiment described above are realized by the processing.

  As described above, the main features of the embodiment of the present invention are as follows. That is, the present embodiment can provide a high-quality image by reducing erroneous image area separation determination of characters or pictures, and image area separation for determining whether or not the character is a white ground character when performing character determination. By using the result of reading both sides of the original when determining the result of white background extraction provided with the apparatus, it is possible to reduce erroneous determination of image area separation due to the effect of show-through. Also, when extracting the white area on one side, the criteria for extracting the white area on one side are changed based on the density information on the other side, so that the accuracy of image area separation is improved and the effect of show-through It is possible to reduce misjudgment due to.

1 is a block diagram illustrating a schematic configuration of a digital full-color image processing apparatus according to an embodiment of the present invention. It is a figure explaining the image flow from the 1st image reading part and the 2nd image reading part in the image processing apparatus which concerns on this embodiment to an image receiving part and an image area separation pre-processing part to a memory control part. It is a figure which shows the detailed structure of the image area separation pre-processing part in the image processing apparatus which concerns on this embodiment. It is a figure which shows 5x5 pixel referred in the high density pattern matching part of the image area separation pre-processing part regarding this embodiment. It is a figure which shows the detailed structure of the scanner correction | amendment part in the image processing apparatus which concerns on this embodiment. FIG. 3 is a diagram illustrating a detailed configuration of a printer correction unit in the image processing apparatus according to the present embodiment. It is a figure which shows the detailed structure of the image area separation part in the scanner correction | amendment part regarding this embodiment. It is a figure which shows the detailed structure of the white area extraction part in the image area separation part regarding this embodiment. It is a block diagram which shows the other structural example of the digital type full color image processing apparatus which concerns on embodiment of this invention. FIG. 3 is a diagram showing a document in which characters are drawn on the front side of the document and a pattern is drawn on the back side of the document. FIG. 6 is an explanatory diagram showing that a read image on the front side of the document is influenced by a back side pattern.

Explanation of symbols

1: first image reading unit, 2: second image reading unit, 3: image receiving unit, 4: image area separation preprocessing unit, 5: memory control unit, 6: scanner correction unit, 7: compression processing unit, 8 : General-purpose bus, 9: Controller, 10: HDD, 11: NIC (network interface controller), 12: External PC terminal, 13: Decompression processing unit, 14: Printer correction unit, 15: Plotter, 41: Binarization for back side Section: 42: high density pattern matching section, 43: expansion section, 61: scanner γ correction section, 62: filter processing section, 63: color correction section, 64: scaling processing section, 65: image area separation section, 71: Printer γ correction unit, 72: gradation processing unit, 81: filter unit, 82: edge extraction unit, 83: halftone dot extraction unit, 84: white area extraction unit, 85: color determination unit, 86: comprehensive determination unit, 91 : Gray pixel detection unit, 92: Color image Element detection unit, 93: RGB white background extraction unit, 94: binarization unit, 95: gray pattern matching unit, 96: gray expansion unit, 97: white determination unit, 98: white pattern matching unit, 99: white pattern correction unit , 100: White expansion unit, 101: White compression unit, 102: White correction unit, 103: Determination unit, 110: Dual screen image reading unit, 111: Image area separation preprocessing unit, 112: Memory control unit, 113: Scanner Correction unit, 114: compression processing unit, 115: general-purpose bus, 116: controller, 117: HDD, 118: NIC (network interface controller), 119: external PC terminal, 120: decompression processing unit, 121: printer correction unit, 122 : Plotter,

Claims (8)

An image reading device that reads both one surface and the other surface of the document, and an image area separation device that classifies the read document image into a plurality of image regions,
The read document image is a double-sided document image, and information to be read is drawn on one side and the other side of the document,
The plurality of image areas classified by the image area separation device includes at least a character area and a picture area,
The image area separation device, when in performing the image area separation of one surface of the document, using both of the image information of the read image information and the other surface of the one surface, it is determined the character area A white area extraction unit that determines whether the white area of the document is based on density information read from the images of the one side and the other side, and the white area extraction unit When it is determined that there is a character region, the image region separation device determines that the character region
When extracting the white region of the one surface, the density information of the other surface is used to change the white region extraction reference of the one surface,
When changing the white area extraction standard of the one surface, when the density of the other surface is higher than a predetermined value, the white area extraction standard is set so that the one surface is easily determined to be a white area. An image processing apparatus characterized in that In claim 1 ,
The image processing apparatus according to claim 1, wherein the white area extraction reference for the one surface is changed only when the density of the one surface is within a certain intermediate density range. An image forming unit that forms an image based on image data output from the image processing apparatus according to claim 1 and that forms the image by transferring the image formed on a sheet. An image forming apparatus. An image reading step for reading both the one side and the other side of the document, and an image area separating step for classifying the read document image into a plurality of image areas,
The read document image is a double-sided document image, in which information to be read is drawn on one side and the other side of the document, and a plurality of images classified by the image area separation device The image area includes at least a character area and a picture area.
In the image area separation step, when performing image area separation on one side of a document, the character area is determined using both read image information on the one side and image information on the other side. , Based on the density information read from the image on the one side and the other side, when determining whether or not it is a white area of the document, and when the white area is determined in the determining step Determining that it is a character area ,
When extracting the white region of the one surface, using the density information of the other surface, changing the white region extraction reference of the one surface,
When changing the white area extraction standard of the one surface, when the density of the other surface is higher than a predetermined value, the white area extraction standard is set so that the one surface is easily determined to be a white area. An image processing method comprising the step of changing The program which described the image processing method of Claim 4 . A storage medium on which the program according to claim 5 is recorded. An image processing apparatus including a scanner correction unit that performs predetermined processing on image data of a document read by an image reading unit and outputs the processed image data.
The original to be read is a double-sided original on which information to be read is drawn on one side and the other side,
The image reading unit includes a first image reading unit that reads the one side and a second image reading unit that reads the other side,
The scanner correction unit includes a scanner γ correction unit that performs γ correction on the RGB image data of the read original, and an image area separation unit that determines whether the image area of the read original is a character area or a picture area And at least a filter processing unit that emphasizes characters of the read image or smoothes a pattern according to a determination result of the image area separation unit,
The image area separation unit includes an edge extraction unit that detects a character edge based on the continuity of black pixels and white pixels, and a white region that is determined as a white region when all of the RGB image data of the pixel matrix is smaller than a threshold value. An area extraction unit,
The white area extraction unit inputs RGB image data, and inputs image density information of one side and the other side of the document read by the first image reading unit and the second image reading unit , When extracting the white region of one surface, the threshold value is changed based on the image density information of the other surface, and the white region of the one surface is changed from the other surface by changing the threshold value. An image processing apparatus having a function of eliminating an influence and making it easy to determine a white area. In claim 7 ,
The image processing apparatus according to claim 1, wherein the change of the threshold value in the white area extracting unit is performed only when the density of the one surface is within a predetermined intermediate density range.

Images