JP4743103B2 - Image processing apparatus, image processing method, and computer program - Google Patents

Description

  The present invention relates to an image processing apparatus, an image processing method, and a computer program for generating a binary image from a multi-valued image, and more particularly, a show-through from a multi-valued image generated by reading a double-side printed document. The present invention relates to an image processing apparatus, an image processing method, and a computer program that generate a small number of binary images.

  When a document is captured by an image capture device such as a scanner or facsimile and output as, for example, a black and white image, if the captured document is printed on both sides, it may be the opposite of the capture surface depending on the thickness and material of the paper. The image on the side surface may be transmitted, and there may be a “show-through” in which the image on the opposite side is mixed with the image on the capturing surface and input.

  In order to prevent such show-through, a conventional image processing apparatus that generates a binary image from a multi-valued image is configured as shown in FIG. The image processing apparatus 1 includes an image input unit 10 to which an image is input from the image capture device 2, a control unit 20 that performs image processing, an image output unit 30 that outputs an image to the image display device 3, and an image capture. A multi-value image storage unit 40 that holds a multi-value image captured by the apparatus 2 and a binary image storage unit 50 that holds a binary image generated by performing image processing on the multi-value image by the control unit 20. . The control unit 20 includes a frequency distribution measurement unit 21 that generates a frequency distribution table of density from the captured multi-value image, a threshold determination unit 23 that determines a threshold for binarizing the multi-value image, and a threshold determination unit. The image processing unit 25 generates a binary image from the multi-valued image based on the threshold value determined by 23.

  In such an image processing apparatus 1, as shown in FIG. 11, first, a document captured from the image capturing device 2 such as a scanner is captured as a multi-value image and stored in the multi-value image storage unit 40 (S 10). Next, a frequency distribution table indicating the characteristics of the image is generated by the frequency distribution measurement unit 21 from the multi-value image stored in the multi-value image storage unit 40 (S20). Further, based on the generated frequency distribution table, the threshold value determination unit 23 determines a threshold value for optimal binarization, for example, by discriminant analysis (S30). Thereafter, based on the threshold value determined by the binary image generation unit 25, for example, either a black pixel or a white pixel is assigned to each pixel of the multi-value image to generate a binary image (S40).

  For example, when the image processing apparatus 1 captures a manuscript with “fine characters” on the capturing surface (front surface) and “show-through” on the surface (back surface) opposite to the capturing surface, Suppose that the characters “Blown-in” as shown in FIG. At this time, in the conventional binarization process, for example, as shown in FIG. 13A, the information on the back surface is also recognized as the information on the front surface, and the information on the back surface is overlapped with the information on the front surface, making it difficult to see. There was a problem. In order to avoid the phenomenon shown in FIG. 13A, even if a threshold value that does not show through is set, for example, as shown in FIG. 13B, the information on the front surface is recognized as the information on the back surface. As a result, there was a problem that the characters on the surface would be blurred.

  In order to solve such a problem, for example, in the image forming apparatus shown in Patent Document 1, binarization processing is performed after image data conversion is performed on the front surface image data including the back image of low density to be removed and the back surface image data. Thus, an overlapping portion of the front surface image data and the back surface image data is obtained. After smoothing by calculating the histogram from this overlapping part, calculate the threshold value and remove the back side image data for the noise contained in the overlapping part, and synthesize this with the image data excluding the overlapping part data By doing so, the show-through is removed.

JP-A-6-62216

  However, the image forming apparatus disclosed in Patent Document 1 has a problem in that it takes time for processing because it is necessary to capture images on both sides.

  Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to provide a novel technique capable of preventing show-through from a double-sided printed document without damaging the information on the surface document. Another object of the present invention is to provide an improved image processing apparatus and image processing method.

  In order to solve the above-described problems, according to an aspect of the present invention, an image processing apparatus that generates a binary image from a multi-value image is provided. The image processing apparatus includes a first threshold value determination unit that determines a first threshold value for distinguishing a white pixel and an undetermined pixel in a multivalued image, and an undetermined pixel surrounded by the white pixel as an undetermined region. A representative density value determining unit for determining a representative density value of the undefined area, and a second threshold value for determining a second threshold value for determining an undefined pixel in the undefined area as a white pixel or a black pixel. The determination unit includes a pixel determination unit that compares a representative density value of each unconfirmed region with a second threshold value to determine an unconfirmed pixel as a white pixel or a black pixel.

  According to the present invention, first, the first threshold value determination unit determines the first threshold value, and the pixels constituting the multi-valued image are classified into white pixels or undefined pixels. Next, the second threshold value determination unit determines the second threshold value in order to determine the undetermined pixel as a white pixel or a black pixel. In order to determine the undetermined pixel as the white pixel or the black pixel, first, the undetermined pixel surrounded by the white pixel is determined as the undetermined area, and the representative density value in the undetermined area is determined as the representative density value. Determined by the department. Then, by comparing the representative density value of the undetermined region with the second threshold value, the undetermined pixel in the undetermined region can be determined as a white pixel or a black pixel. As described above, the first threshold value and the second threshold value are automatically obtained from the density values of the pixels constituting the multi-valued image, and it is possible to prevent show-through from appearing in the generated binary image.

  Here, the representative density value can be the density of the darkest color indicated by the undetermined pixels constituting the undetermined area, that is, the maximum density value. As a result, it can be seen that the undetermined region includes only pixels having the maximum density value.

  The image processing apparatus according to the present invention may further include a histogram creation unit that creates a histogram related to the representative density value of the undetermined region. By creating the histogram, it can be seen how the representative density values of the undetermined pixels are distributed. Here, since the show-through image usually appears in a lighter color than the surface image, the density value of the pixels constituting the show-through image is smaller than the density value of the pixels constituting the surface image. Therefore, for example, the density value corresponding to the minimum value of the minimum values of the histogram can be determined as the second threshold value by the second threshold value determination unit. At this time, if the density value corresponding to the minimum value of the minimum value is larger than the limit density value to be determined for the black pixel, the second threshold value determination unit may determine the limit density value as the second threshold value. . Thereby, even if there is no minimum value in the histogram, the second threshold value can be set appropriately.

  On the other hand, the second threshold value determination unit determines the smallest representative density value as the second threshold value among the undetermined areas occupying a predetermined ratio with respect to the total number of undetermined areas from the maximum density value side of the histogram. Also good. As a result, it is possible to prevent the determination of the density value appearing as an error as the second threshold value.

  It is also possible to provide a computer program that causes a computer to function as the image processing apparatus. The computer program is stored in a storage device included in the computer, and is read and executed by a CPU included in the computer, thereby causing the computer to function as the image processing device. A computer-readable recording medium storing a computer program can also be provided. The recording medium is, for example, a magnetic disk or an optical disk.

  In order to solve the above problem, according to another aspect of the present invention, an image processing method for generating a binary image from a multi-valued image is provided. Such an image processing method includes a first threshold value determining step for determining a first threshold value for distinguishing white pixels and undetermined pixels in a multivalued image, and the entire area of the multivalued image is surrounded by white pixels. A non-determined area determining step for determining the undetermined pixel as an undetermined area, a representative density value determining step for determining a representative density value in the undetermined area, A second threshold value determining step for determining a second threshold value for determination and a pixel for determining the undecided pixel as a white pixel or a black pixel by comparing the representative density value of the undetermined area with the second threshold value. A confirmation step.

  According to the present invention, first, the first threshold value determination unit determines the first threshold value, and the pixels constituting the multi-valued image are classified into white pixels or undefined pixels. Next, after determining the undetermined pixels surrounded by the white pixels for the entire area of the image as the undetermined areas, the representative density values in all the undetermined areas are determined. Furthermore, a second threshold value for determining uncertain pixels constituting the uncertain area into white pixels or black pixels is determined. Then, by comparing the representative density value of the undetermined region with the second threshold value, the undetermined pixel in the undetermined region can be determined as a white pixel or a black pixel. As described above, the first threshold value and the second threshold value are automatically obtained from the density values of the pixels constituting the multi-valued image, and it is possible to prevent show-through from appearing in the generated binary image.

  Here, the representative density value can be the density of the darkest color indicated by the undetermined pixels constituting the undetermined area, that is, the maximum density value. As a result, it can be seen that the undetermined region includes only pixels having the maximum density value.

  In addition, the image processing method according to the present embodiment can further include a histogram creation step of creating a histogram related to the representative density value of each undetermined region. Here, since the show-through image usually appears in a lighter color than the surface image, the density value of the pixels constituting the show-through image is smaller than the density value of the pixels constituting the surface image. Thus, in the second threshold value determining step, the density value corresponding to the minimum value among the minimum values of the histogram can be determined as the second threshold value. Here, in the second threshold value determining step, when the density value corresponding to the minimum minimum value is larger than the limit density value to be determined for the black pixel, the limit density value can be determined as the second threshold value. . Thereby, even if there is no minimum value in the histogram, the second threshold value can be set appropriately.

  On the other hand, in the second threshold value determining step, the smallest representative density value is determined as the second threshold value among the undetermined areas occupying a predetermined ratio with respect to the total number of undetermined areas from the maximum density value side of the histogram. Also good. As a result, it is possible to prevent the determination of the density value appearing as an error as the second threshold value.

  Furthermore, in order to solve the above problems, according to another aspect of the present invention, an image processing method for generating a binary image from a multi-valued image is provided. Such an image processing method includes a first threshold value determining step for determining a first threshold value for distinguishing a white pixel and an undetermined pixel in a multi-valued image, and an undetermined value including an undetermined pixel surrounded by the white pixel. A representative density value determining step in which the process of determining the representative density value in the undetermined area while determining the area is repeated for all areas of the multi-valued image, and the undetermined pixels in the undetermined area are white pixels or The second threshold value determining step for determining the second threshold value for determining the black pixel is compared with the representative density value of the undetermined region and the second threshold value, and the undetermined pixel is changed to a white pixel or a black pixel. And a pixel determining step for determining.

  According to the present invention, first, the first threshold value determination unit determines the first threshold value, and the pixels constituting the multi-valued image are classified into white pixels or undefined pixels. Next, each time an undetermined pixel surrounded by white pixels is determined as an undetermined region, a representative density value in the undetermined region is determined. This process is repeated to determine the representative density value of the undetermined area existing in the entire area of the image. Furthermore, a second threshold value for determining uncertain pixels constituting the uncertain area into white pixels or black pixels is determined. Then, by comparing the representative density value of the undetermined region with the second threshold value, the undetermined pixel in the undetermined region can be determined as a white pixel or a black pixel. As described above, the first threshold value and the second threshold value are automatically obtained from the density values of the pixels constituting the multi-valued image, and it is possible to prevent show-through from appearing in the generated binary image.

  As described above, according to the present invention, it is possible to provide an image processing apparatus and an image processing method capable of preventing show-through from a double-side printed original without losing information on the front-side original.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

  An image processing apparatus 100 according to an embodiment of the present invention, for example, as illustrated in FIG. 1, converts a multi-valued image captured from an image capturing apparatus 200 such as a single-sided scanner into an image display apparatus 300 such as a display as a binary image. Used when displaying. The document captured from the image capture device 200 is transmitted to the image processing device 100 as a multi-valued image D1. The image processing device 100 generates a binary image D3 from the received multi-valued image D1 and transmits it to the image display device 300, and the image display device 300 displays the received binary image D3. The configuration of the image processing apparatus 100 that generates the binary image D3 from the multivalued image D1 as described above will be described below.

  First, the configuration of the image processing apparatus 100 according to the embodiment of the present invention will be described with reference to FIG. Here, FIG. 2 is a block diagram showing a configuration of the image processing apparatus 100 according to the present embodiment.

  As illustrated in FIG. 2, the image processing apparatus 100 according to the present embodiment is generated by an image input unit 110 to which an image is input from the image capturing apparatus 200, a control unit 120 that performs image processing, and a control unit 120. An image output unit 130 that outputs the binary image to the image display device 300, a multi-value image storage unit 140 that stores the multi-value image D1 captured by the image capture device 200, and a threshold storage unit 150 that stores a threshold value. A representative density value storage unit 160 for each region, and a binary image storage unit 170 that stores a binary image D3 generated by image processing of the multi-valued image D1 by the control unit 120.

  The image input unit 110 is a functional unit that stores the multi-value image D1 input from the image capturing device 200 in the multi-value image storage unit 140 and supports generation of the binary image D3 from the multi-value image D1. .

  The control unit 120 includes a frequency distribution measurement unit 121 that generates a frequency distribution table of density in the multi-valued image D1 generated from the document captured by the image capturing device 200, and the multi-valued image D1 as a white pixel and an undetermined pixel. A first threshold value determination unit 122 that determines a first threshold value for classifying into a non-determined region, an undetermined region determination unit 123 that determines an undetermined region composed of undetermined pixels, and a representative density value of the undetermined region A representative density value determining unit 124 that determines the representative density value of the undetermined region, a histogram creating unit 125 that generates a histogram regarding the representative density value of the undetermined region, and a second threshold value for classifying the undetermined pixels into white pixels or black pixels. A second threshold value determination unit 126 and a binary image generation unit 127 that generates a binary image D3 from the multi-valued image D1.

  The frequency distribution measuring unit 121 measures the pixel value of each pixel of the multi-valued image, and creates a density frequency distribution table indicating the characteristics of the image. The density frequency distribution table assigns a continuous change (density value) of white to gray to black quantized to, for example, 8 bits, that is, 256 levels from 0 to 255, for each pixel of the input multi-valued image D1. A frequency distribution table is created for the assigned density values. In the present embodiment, 0 is black, 255 is white, and a value from 0 to 255 is described as corresponding to a color from black (maximum density value) to white (minimum density value). However, 0 may be white and 255 may be black. In the present embodiment, in the case of a binary image, 0 is considered black and 1 is considered white.

  The first threshold value determination unit 122 is a functional unit that determines a first threshold value for distinguishing a white pixel from an undetermined pixel (a pixel not determined as a white pixel or a black pixel). The first threshold can be calculated using, for example, a discriminant analysis method. Details of the discriminant analysis method will be described later. The first threshold value determined by the first threshold value determination unit 122 is stored in the threshold value storage unit 150.

  The undetermined area determination unit 123 is a functional unit that determines an undetermined area, which is an area formed by selecting one or more undetermined pixels classified as white pixels by the first threshold. In the undetermined region determination unit 123 according to the present embodiment, for example, an undetermined pixel surrounded by white pixels is determined as an undetermined region.

  The representative density value determining unit 124 is a functional unit that determines the representative density value of the undetermined region determined by the undetermined region determining unit 123. Here, the representative density value is a density value that characterizes an undetermined area determined based on a predetermined standard. For example, by determining the maximum density value as the representative density value, it is possible to indicate that this undetermined area includes only the density of the density at most. In the representative density value determination unit 124 according to the present embodiment, for example, after measuring the density values of the undefined pixels constituting the undefined area, the minimum value, that is, the darkest color among the measured density values of each pixel. The density value (maximum density value) of the pixel indicating is determined as the representative density value of this undetermined area. The representative density value of the determined undetermined area is stored in the area-specific representative density value storage unit 160 described later.

  The histogram creation unit 125 is a functional unit that creates a histogram related to the representative density value of the undetermined region. The histogram is a graph showing the appearance distribution of the density values of the image. The histogram creation unit 125 acquires the minimum density value of each undetermined area stored in the representative density value storage unit 160 for each area, and creates a histogram for the number of undetermined areas for the minimum density value.

  The second threshold value determination unit 126 is a functional unit that determines a second threshold value for determining an undetermined pixel as a white pixel or a black pixel. The second threshold determination unit 126 according to the present embodiment determines the second threshold using the histogram created by the histogram creation unit 125. The second threshold value determined by the second threshold value determination unit 126 is stored in the threshold value storage unit 150.

  The binary image generation unit 127 is a functional unit that performs processing for generating a binary image D3 from the multilevel image D1 stored in the multilevel image storage unit 140 using the first threshold value and the second threshold value. . The binary image generation unit 127 acquires the first threshold value and the second threshold value from the threshold value storage unit 150, and generates a binary image D3 from the multilevel image D1 stored in the multilevel image storage unit 140. The generated binary image D3 is stored in the binary image storage unit 170.

  The image output unit 130 is a functional unit that outputs the binary image D3 stored in the binary image storage unit 170 to the image display device 300, for example. The image output unit 130 may function to output to the image display device 300 according to a binary image output signal from the binary image storage unit 170, for example, and outputs in response to a request from the image display device 300. May function as follows.

  The multi-value image storage unit 140 is a storage unit that stores the multi-value image D1 input from the image capturing device 200 to the image processing device 100. The threshold storage unit 150 is a storage unit that stores the first threshold value determined by the first threshold value determination unit 122 and the second threshold value determined by the second threshold value determination unit 126. Further, the region-specific representative density value storage unit 160 is a storage unit that stores a representative density value for each unconfirmed region measured by the representative density value determination unit 124. For example, as shown in FIG. 3, the area-specific representative density value storage unit 160 stores an area number 161 for specifying an undetermined area, a maximum density value 162 in the undetermined area, and the like. The binary image storage unit 170 is a storage unit that stores the binary image D3 generated by the binary image generation unit 127. These multi-value image storage unit 140, threshold value storage unit 150, area-specific representative density value storage unit 160, and binary image storage unit 170 are configured to include a memory such as a RAM or a hard disk.

  The configuration of the image processing apparatus 100 according to the present embodiment has been described above. Next, based on FIG. 4, the image processing method by the image processing apparatus 100 according to the present embodiment will be described. FIG. 4 is a flowchart illustrating the image processing method according to the present embodiment.

  In the image processing method by the image processing apparatus 100 according to the present embodiment, as shown in FIG. 4, first, a multi-value image D1 is input to the image processing apparatus 100 (S101; multi-value image input step). For example, the image processing apparatus 100 acquires the multi-valued image D1 by capturing the document using the image capturing device 200 and inputting the multi-valued image D1 created from the captured document to the image input unit 110. In addition, the image input unit 110 stores the input multi-valued image D1 in the multi-valued image storage unit 140 and transmits a binary image creation signal to the control unit 120.

  Next, for each pixel of the multi-valued image D1 stored in the multi-valued image storage unit 140, the density distribution measuring unit 121 measures the density and creates a frequency distribution table relating to the density (S103; frequency distribution table creating step). In step S103, for each pixel of the input multi-valued image D1, for example, a quantized density value is assigned to 256 levels from 0 to 255. In this way, a frequency distribution table representing the light and shade of the multivalued image D1 is created.

  In step S103, when the frequency distribution table of the density for the multi-valued image D1 is created, the first threshold value determination unit 122 determines the first threshold value based on the frequency distribution table (S105; first Threshold determination step). The first threshold is a threshold for distinguishing a white pixel and an undetermined pixel for each pixel of the multi-valued image D1. The first threshold value can be determined by, for example, a discriminant analysis method. The discriminant analysis method is a method for determining a threshold value so that the two regions are best separated when the density value distribution of the image is divided into two regions by the threshold value. As a measure for separating the regions by the threshold value, a ratio (discriminant ratio) between the variance of the average value of the two regions and the variance of each region is used, and the threshold value is selected so that the discriminant ratio is maximized. . The discriminant analysis method is a widely used technique because a threshold can be selected even when there is no clear valley (difference) in the frequency distribution table. The first threshold value determination unit 122 stores the first threshold value selected by the discriminant analysis method in the threshold value storage unit 150 that stores the threshold value.

  Next, an undetermined area in the multi-valued image D1 is determined (S107; undetermined area determining step). The undetermined region determination unit 123 uses the first threshold value determined by the first threshold value determination unit 122 to classify the pixels constituting the multi-valued image D1 into white pixels and undetermined pixels. Then, the undetermined area determination unit 123 determines an undetermined pixel surrounded by white pixels as an undetermined area. For example, regarding the multi-valued image D1 captured by the image capturing device 200, when the pixels constituting the multi-valued image D1 are divided into white pixels and undetermined pixels by the first threshold, as shown in FIG. In addition, it is assumed that the pixel is determined to be a white pixel by the first threshold value and the other pixel that is not determined (gray and black pixels in FIG. 5A). The undetermined area determination unit 123 detects an area in which undetermined pixels are surrounded by white pixels, and determines this area as an undetermined area. In FIG. 5A, three regions, region 0, region 1, and region 2, are determined.

  When the undetermined area is determined, the representative density value of each undetermined area is determined (S109; representative density value determining step). In the present embodiment, the representative density value determination unit 124 measures the density value of the pixels constituting the undetermined area determined in step S107, and the lowest density value (the highest in the undetermined area) among the measured density values. The density value of a pixel showing a dark color) is determined as a representative density value.

  The representative density value determination unit 124 measures, for example, the density values of the undefined pixels constituting the undefined area one by one, and each time the measurement is performed, the representative density value (maximum here) Density value). The provisional representative density value is stored in, for example, a buffer (not shown) provided in the image processing apparatus 100, and the pixel value of the undetermined pixel is determined by comparison with the pixel value of the undetermined pixel. When the representative density value is larger, the representative density value is updated to the pixel value. On the other hand, when the representative density value is larger, the representative density value is not updated and is compared with the density value of the next undetermined pixel. Then, when the density values of all the undetermined pixels constituting the undetermined area have been measured, the provisional representative density value stored in the buffer can be determined as the representative density value in the undetermined area. it can.

  The representative density value of the undetermined area thus determined is stored in the area-specific representative density value storage unit 160 in association with the area number. For example, when the maximum density value of the undefined pixel constituting the undefined area is measured for the three undefined areas shown in FIG. 5A, the maximum density value of the area 0 is 160, as shown in FIG. Assume that the maximum density value of region 1 is 34 and the maximum density value of region 2 is 36.

  Thereafter, a histogram relating to the representative density value of the undetermined region determined in step S109 is created (S111; histogram creation step). The histogram creation unit 125 creates a histogram for the representative density value of the undetermined area stored in the area-specific representative density value storage unit 160 in step S109. The histogram is expressed as shown in FIG. 6, for example. The horizontal axis of this histogram indicates the representative density value of the undetermined area, and the vertical axis indicates the number of undetermined areas indicating the density value.

  Next, based on the histogram created in step S111, a second threshold value for determining an undetermined pixel constituting the undetermined region as a white pixel or a black pixel is determined (S113; second threshold value determination). Step). In the present embodiment, the second threshold value is determined as a density value corresponding to the minimum value among the minimum values in the histogram. Here, the undetermined pixel includes a pixel constituting the front image and a pixel constituting an image in which the back image is shown on the front. The show-through image usually appears in a multi-valued image captured from the image capturing device 200 with a lighter color than the surface image. That is, it is considered that the pixels constituting the show-through image have a smaller density value than the pixels constituting the surface image.

  Therefore, as shown in FIG. 6, the histogram relating to the maximum density value of the undetermined area includes a light-colored area (the peak on the right side in the histogram of FIG. 6) showing a density value with a relatively small maximum density value, and a relatively high density value. A dark region having a large density value (left mountain in the histogram of FIG. 6) appears. The second threshold value is determined as a white pixel for an undetermined pixel constituting an undetermined region considered as a show-through image, and as a black pixel for an undetermined pixel considered as an image on the front surface. It is a threshold for. Therefore, in the present embodiment, the density value when the number of regions considered to represent a show-through image in the histogram exceeds the peak is determined as the second threshold value. That is, the second threshold value of the present embodiment is a density value corresponding to the minimum value among the minimum values in the histogram.

  The determination of the second threshold value in the present embodiment is performed, for example, by scanning the histogram shown in FIG. 6 from the small density value side to the large density value side, that is, from right to left, and the density corresponding to the minimum value that appears first. The value can be determined as a second threshold. Alternatively, after all the minimum values appearing in the histogram are detected, the density value corresponding to the minimum value indicating the minimum density value may be determined as the second threshold value. As described above, the second threshold value determined by the second threshold value determination unit 126 is stored in the threshold value storage unit 150.

  When the first threshold value and the second threshold value are determined, the binary image generation unit 127 performs a process of generating a binary image (S115; binary image generation step). In the binary image generation method according to the present embodiment, first, the white pixel and the undetermined pixel are classified for the multi-valued image D1 based on the first threshold value calculated in step S105 and stored in the threshold value storage unit 150. As a result, a region to be a white pixel is determined. Next, in step S109, the maximum density value of the undetermined area stored in the area-specific representative density value storage unit 160 is compared with the second threshold value stored in the threshold value storage unit 150 to configure the undetermined area. The undetermined pixel is determined as a white pixel or a black pixel.

  Specifically, when the maximum density value of the undetermined area is less than the second threshold, the pixel is white, and when the maximum density value of the undetermined area is equal to or greater than the second threshold, the pixel is black. As described above, the maximum density value of the undetermined area indicates the density value of the pixel showing the darkest color in the undetermined area, and means that there is only a pixel of that density at most in the undetermined area. To do. On the other hand, as described above, the second threshold value is determined as follows: the undecided pixels that constitute the indeterminate area considered to be a show-through image are white pixels, and the undecided pixels that constitute the undetermined area that is considered to be the front image are This is a threshold value for determining a black pixel. Therefore, by determining whether or not the maximum density value in the undetermined area is greater than or equal to the second threshold value, the pixels constituting the undetermined area considered as a show-through image are changed to white pixels or black pixels. Therefore, it is possible to prevent the show-through image from being displayed on the created binary image D3.

  For example, a description will be given of a process of determining uncertain pixels constituting the three unconfirmed regions, region 0, region 1, and region 2 shown in FIG. 5A as white pixels or black pixels. Reads the maximum density value of the undetermined area stored in the area-specific representative density value storage unit 160 and creates a histogram. If the histogram is created as shown in FIG. 6, for example, the second threshold value determination unit 126 scans the histogram from the right (smaller density value side) to the left (larger density value side) of the histogram. The density value corresponding to the first minimum value that appears first is determined as the second threshold value. Therefore, the density value T is determined as the second threshold value in FIG.

  Then, the binary image generation unit 127 compares the magnitudes of the second threshold T and the maximum density value of the undetermined area stored in the area-specific representative density value storage unit 160 to configure the undetermined area. The undetermined pixel is determined as a white pixel or a black pixel. For example, it is assumed that the second threshold T is 150, and the maximum density values of the areas 0, 1 and 2 that are undefined areas are the values shown in FIG. At this time, since the maximum density value of the region 0 is less than the second threshold T as illustrated in FIG. 7, the binary image generation unit 127 determines the undetermined pixels constituting the region 0 as white pixels. On the other hand, the maximum density value of region 1 is 34, and the maximum density value of region 2 is 36, which is equal to or greater than the second threshold value, and therefore, the undetermined pixels constituting region 1 and region 2 are determined to be black pixels. Therefore, an image composed of white pixels and undetermined pixels as shown in FIG. 5A becomes a binary image D3 as shown in FIG. 5B. In this way, a binary image D3 is generated from the multi-valued image D1. As described above, in the present embodiment, 0 is black, 255 is white, 0 is the maximum density value, and 255 is the minimum density value (that is, “0> 255”).

  The image processing method according to the present embodiment has been described above. For example, when a binary image is generated from the multilevel image shown in FIG. 12 by the image processing method according to the present embodiment, a binary image as shown in FIG. 8 is generated. When this binary image is compared with the binary image generated by the conventional image processing method, conventionally, as shown in FIG. However, the binary image generated by the image processing method according to the present embodiment is an image in which the characters “show-through” are not displayed. Further, in order to prevent show-through, the characters “thin characters” that are images of the front surface have been blurred as shown in FIG. 13B, but the characters are generated by the image processing method according to the present embodiment. In the binary image, there is little blurring of the characters of “thin characters” and can be clearly recognized.

  As described above, according to the image processing apparatus 100 and the image processing method according to the embodiment of the present invention, when generating a binary image from a multivalued image, first, the first threshold value determination unit 122 configures the multivalued image. A first threshold value for dividing a pixel to be processed into a white pixel and an undetermined pixel is determined. Then, an undetermined pixel surrounded by white pixels is determined as an undetermined region, and the maximum density value of the undetermined pixels constituting the undetermined region is measured. Further, a histogram for the measured maximum density value is created, and a second threshold value is determined from this histogram. Then, by comparing the maximum density value in the undetermined region with the second threshold value, the undetermined pixels constituting the undetermined region are determined as white pixels or black pixels. As described above, the first threshold value and the second threshold value can be automatically determined by the image processing method according to the present embodiment, and no show-through image appears in the generated binary image. The captured image can also be clearly displayed.

  Here, the representative density value of the undetermined area in step S109 is the maximum density value. However, a histogram for the density value of the undetermined pixel is created for each undetermined area, and the representative density value is determined from this histogram. Good. Hereinafter, a method for determining the representative density value of the undetermined region using the histogram will be described with reference to FIG. FIG. 9 is an explanatory diagram for explaining a method for determining a representative density value of an undefined area using a histogram. FIG. 9A shows an histogram of an undefined area composed of pixels having a small density value as a whole. (B) is an example of a histogram of an undetermined area composed of pixels with a small density value and pixels with a large density value, and (c) is a histogram of an undetermined area composed of pixels with a large density value as a whole. An example is shown.

  The histogram for the density value of the undetermined pixels constituting the undetermined area can be created by the same process as Step S111 of the image processing method according to the embodiment of the present invention. First, the density values of all undetermined pixels constituting the undetermined area are measured. Next, a histogram for the measured density value of the indeterminate pixel is created, and the created histogram is scanned from the larger density value side to the smaller density value side. Then, the density value at the first peak after the start of scanning is determined as the representative density value of this undefined area.

  The density value at the first peak of the histogram can be, for example, the density value at the maximum value that first appears after the start of scanning. At this time, for example, as shown in FIG. 9A, when the density value of the undetermined pixels constituting the undetermined area is small as a whole (for example, the area 0 in FIG. 5A), the first after the start of scanning. The maximum value is a small value. Further, as shown in FIG. 9B, when there are a large density value and a small density value of the undetermined pixels constituting the undetermined area (for example, area 1 in FIG. 5A), after the start of scanning. The first local maximum is a large value. Furthermore, as shown in FIG. 9C, when the density value of the undetermined pixels constituting the undetermined area is large as a whole (for example, the area 2 in FIG. 5A), the first after the start of scanning. The maximum value is a large value. As described above, the representative density value of the determined undetermined area substantially coincides with the case where the representative density value is the maximum density value in the undetermined area.

  Alternatively, the representative density value of the undetermined region may be a density value at the time when the predetermined number of pixels is first exceeded after the start of scanning. For example, when only one pixel having a large density value is included in the undetermined region as shown in FIG. 9A, the presence of this pixel is preferably regarded as an error, but the first maximum value after the start of scanning If this is the representative density value, this large density value is determined as the representative density value. However, it is possible to prevent the selection of a density value to be treated as an error by using the density value of a certain number of pixels as the representative density value.

  In step S113, the density value corresponding to the minimum value in the histogram is determined as the second threshold value. At this time, the limit density value is set so that the second threshold value does not exceed a predetermined value. It may be provided. For example, when the minimum value does not appear, such as when the histogram continues to decrease from the minimum density value side (right side of FIG. 6) to the maximum density value side (left side of FIG. 6), the second threshold is almost maximum. The density value, that is, 0 is set. The binary image created at this time is composed of only white pixels, and there is a possibility that the surface image does not appear. For such a problem, by providing a limit density value so that the second threshold value does not become a predetermined value or more, it is possible to prevent a pixel that is supposed to be a black pixel from being determined as a white pixel. . The limit density value may be a preset density value, for example, and a binary image is created so that a predetermined number (for example, a predetermined ratio with respect to the total number of undefined areas) of undefined areas is black pixels. You may decide for every object.

  On the other hand, the second threshold value is determined by scanning the histogram created in step S111 from the minimum density value side (the right side in FIG. 6) to the maximum density value side (the left side in FIG. 6). The density value may be set when the ratio reaches the number of undetermined areas. Thus, by determining the second threshold value according to the ratio to the total number of unconfirmed regions, it is possible to prevent a minimum value that can be regarded as an error from being set as the second threshold value.

  As described above, according to the image processing method according to the embodiment of the present invention, the first threshold value and the second threshold value used for creating the binary image D3 from the multi-valued image D1 input to the image processing apparatus 100. Can be automatically determined. Therefore, the first threshold value and the second threshold value can be easily determined for each multi-value image D1 input to the image processing apparatus 100, and the threshold value suitable for the image to be processed can be determined each time.

  As mentioned above, although preferred embodiment of this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to the example which concerns. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  For example, in the above embodiment, the discriminant analysis method is used as a method for determining the first threshold value. However, the present invention is not limited to such an example, and for example, a p-tile method, a mode method, or the like can be used.

It is explanatory drawing which shows the structural example using an image processing apparatus. 1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is explanatory drawing which shows an example of a structure of the representative density value storage part classified by area. 4 is a flowchart illustrating an image processing method according to the embodiment. FIG. 4 is an explanatory diagram showing an image processing method according to the embodiment, in which (a) shows a state in which a white pixel or an undetermined pixel is assigned to each pixel of a multi-valued image, and (b) shows a black for an undetermined pixel. A state in which pixels or white pixels are allocated is shown. It is explanatory drawing which shows an example of the histogram regarding the maximum density value of an undetermined area | region. It is explanatory drawing for demonstrating the method of assigning the color to an undetermined pixel in the embodiment. The example of the binary image produced | generated by the image processing method concerning the embodiment from the multi-value image of FIG. 12 is shown. It is explanatory drawing for demonstrating the determination method of the representative density value of an undetermined area using a histogram, (a) is an example of the histogram of the undetermined area which consists of a pixel with a small density value as a whole, (b ) Is an example of a histogram of an undetermined area composed of pixels with a small density value and pixels with a large density value, and (c) shows an example of a histogram of an undetermined area composed of pixels with a large density value as a whole. It is a block diagram which shows the structure of the conventional image processing apparatus. It is a flowchart which shows the conventional image processing method. It is explanatory drawing which shows an example of the multi-value image which has show-through. 12 is an example of a binary image generated by the conventional binarization process from the multi-valued image of FIG. 12, (a) is a binary image having a show-through after the binarization process, and (b) is A binary image in which a part of the image is blurred after the binarization processing is shown.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Image processing apparatus 110 Image input part 120 Control part 121 Frequency distribution measurement part 122 1st threshold value determination part 123 Undecided area | region determination part 124 Representative density value determination part 125 Histogram creation part 126 2nd threshold value determination part 127 Binary image Generation unit 130 Image output unit 140 Multi-value image storage unit 150 Threshold storage unit 160 Representative density value storage unit for each region 170 Binary image storage unit 200 Image capturing device 300 Image display device

Claims (13)

An image processing device that generates a binary image from a multi-valued image,
A first threshold value determination unit for determining a first threshold value for distinguishing white pixels and uncertain pixels in the multi-valued image;
A non-determined area surrounded by the white pixels as an undetermined area, a representative density value determining unit for determining a representative density value of the undetermined area;
A histogram creation unit for creating a histogram related to the representative density value of the undefined region;
A second threshold value determination for determining a second threshold value for determining the uncertain pixel in the uncertain region as a white pixel or a black pixel by a density value corresponding to the minimum value among the minimum values of the histogram. And
Comparing a representative density value of the undetermined region with the second threshold value, and determining a undetermined pixel as a white pixel or a black pixel;
An image processing apparatus comprising: An image processing device that generates a binary image from a multi-valued image,
A first threshold value determination unit for determining a first threshold value for distinguishing white pixels and uncertain pixels in the multi-valued image;
A non-determined area surrounded by the white pixels as an undetermined area, a representative density value determining unit for determining a representative density value of the undetermined area;
A histogram creation unit for creating a histogram related to the representative density value of the undefined region;
The second threshold value for determining the undefined pixel in the undefined area as a white pixel or a black pixel is an undefined area that occupies a predetermined ratio with respect to the total number of undefined areas from the maximum density value side of the histogram. A second threshold value determination unit that is determined by a minimum representative density value in the region;
Comparing a representative density value of the undetermined region with the second threshold value, and determining a undetermined pixel as a white pixel or a black pixel;
An image processing apparatus comprising: The representative density value, wherein the the maximum density value of the undetermined pixels constituting the undetermined area, the image processing apparatus according to claim 1 or 2. The second threshold value determining unit determines the limit density value as the second threshold value when the density value corresponding to the minimum value of the minimum value is larger than the limit density value to be determined for the black pixel. The image processing apparatus according to claim 1 , wherein the image processing apparatus is characterized. Computer, characterized in that to function as an image processing apparatus according to claim 1-4, the computer program. An image processing method for generating a binary image from a multi-valued image,
A first threshold value determining step for determining a first threshold value for distinguishing between white pixels and uncertain pixels in the multi-valued image;
An undetermined area determination step for determining an undetermined pixel surrounded by the white pixels as an undetermined area for the entire area of the multi-valued image;
A representative density value determining step for determining a representative density value in the undefined region;
A histogram creating step for creating a histogram relating to a representative density value of the undefined region;
Second threshold value determination for determining a second threshold value for determining the uncertain pixel in the uncertain region as a white pixel or a black pixel by a density value corresponding to a minimum value among the minimum values of the histogram. Steps,
Comparing the representative density value of the undetermined region with the second threshold, and determining the undetermined pixel as a white pixel or a black pixel;
An image processing method comprising: An image processing method for generating a binary image from a multi-valued image,
A first threshold value determining step for determining a first threshold value for distinguishing between white pixels and uncertain pixels in the multi-valued image;
An undetermined area determination step for determining an undetermined pixel surrounded by the white pixels as an undetermined area for the entire area of the multivalued image
A representative density value determining step for determining a representative density value in the undefined region;
A histogram creating step for creating a histogram relating to a representative density value of the undefined region;
The second threshold value for determining the undefined pixel in the undefined area as a white pixel or a black pixel is an undefined area that occupies a predetermined ratio with respect to the total number of undefined areas from the maximum density value side of the histogram. A second threshold value determining step determined by a minimum representative density value in the region;
A pixel determination step of comparing the representative density value of the undefined area with the second threshold value to determine the undefined pixel as a white pixel or a black pixel;
An image processing method comprising: The image processing method according to claim 6 , wherein the representative density value is a maximum density value of an undetermined pixel constituting the undetermined region. In the second threshold value determining step, when the density value corresponding to the minimum value of the minimum value is larger than the limit density value to be determined for the black pixel, the limit density value is determined as the second threshold value. The image processing method according to claim 6 , wherein the image processing method is characterized. An image processing method for generating a binary image from a multi-valued image,
A first threshold value determining step for determining a first threshold value for distinguishing between white pixels and uncertain pixels in the multi-valued image;
The representative density value determination is performed by repeatedly determining the representative density value in the undetermined area while determining the undetermined area composed of the undetermined pixels surrounded by the white pixels. Steps,
A histogram creating step for creating a histogram relating to a representative density value of the undefined region;
A second threshold value determination for determining a second threshold value for determining the uncertain pixel in the uncertain region as a white pixel or a black pixel by a density value corresponding to the minimum value among the minimum values of the histogram. Steps,
A pixel determination step of comparing the representative density value of the undefined area with the second threshold value to determine the undefined pixel as a white pixel or a black pixel;
An image processing method comprising: An image processing method for generating a binary image from a multi-valued image,
A first threshold value determining step for determining a first threshold value for distinguishing between white pixels and uncertain pixels in the multi-valued image;
The representative density value determination is performed by repeatedly determining the representative density value in the undetermined area while determining the undetermined area composed of the undetermined pixels surrounded by the white pixels. Steps,
A histogram creating step for creating a histogram relating to a representative density value of the undefined region;
The second threshold value for determining the undefined pixel in the undefined area as a white pixel or a black pixel is an undefined area that occupies a predetermined ratio with respect to the total number of undefined areas from the maximum density value side of the histogram. A second threshold value determining step determined by a minimum representative density value in the region;
A pixel determination step of comparing the representative density value of the undefined area with the second threshold value to determine the undefined pixel as a white pixel or a black pixel;
An image processing method comprising: The image processing method according to claim 10 , wherein the representative density value is a maximum density value of an undetermined pixel constituting the undetermined region. In the second threshold value determining step, when the density value corresponding to the minimum value of the minimum value is larger than the limit density value to be determined for the black pixel, the limit density value is determined as the second threshold value. The image processing method according to claim 10 , wherein the image processing method is characterized.

Images