JP2005311991A - Image processing apparatus, image processing method, storage medium, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing technology capable of properly binarizing each object in a multi-value image. <P>SOLUTION: An image processing apparatus provided with a binarizing section 232 including a plurality of binarizing functions each applying binary processing to the multi-value image includes a binarization processing discrimination section 231 that determines whether target pixel data are edge pixels configuring an edge part of an object included in the multi-value image, in-edge pixels being pixels of the object, or background pixels configuring a background region, and a selector section 235 is characterized in to switchingly use a plurality of the binarizing functions depending on the target pixel data belonging to any of the pixels. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing technique for binarizing and outputting a multi-valued image.

  As processing for image data, there is binarization processing for binarizing image data having multi-value density (hereinafter referred to as “multi-value image”) with a predetermined threshold value to reduce the data amount.

  In general, in the binarization process, it is determined whether the density of the target pixel is higher or lower than a preset threshold value. If it is higher, the target pixel is set as a white pixel. For example, by using black pixels, a multi-valued image is converted into a binary image composed of only two types of data, white or black.

  Here, the threshold value when performing the binarization processing is often different for each multi-valued image. Therefore, when actually performing binarization processing, a predetermined threshold value is input and binarization processing is performed. If the binarization result is not appropriate, the threshold value is changed and input. Then, an optimum threshold value is obtained by adjustment, such as performing binarization processing again.

  However, the optimum threshold value is not necessarily one in the same multi-valued image. For example, when a plurality of objects are arranged in regions having different background densities, a predetermined object is extracted. The threshold value that was optimal for this is not necessarily optimal for the extraction of other objects. For this reason, in the same multi-valued image, a region where appropriate binarization has been performed and a region where appropriate binarization has not been performed are mixed.

On the other hand, for example, in Patent Document 1, the threshold value is changed so as to follow the background density of the multi-valued image, so that binarization processing can be performed with an optimum threshold value in each region.
Japanese Patent Laid-Open No. 2001-218051

  However, in the method according to Patent Document 1, the binarization process may not be appropriately performed depending on the multi-value image to be processed. For example, as shown in FIG. 5A, in the case of a multi-valued image in which an object with a high density follows after an object with a high density as shown in FIG. 5A (that is, there are a plurality of objects having a uniform internal density and different densities. In the case of multi-valued images that are mixed, the density of the AA ′ portion is as shown in FIG. For this reason, the tracked threshold value is as shown in FIG.

  In this way, if the threshold value is changed following the multi-valued image, an object having a low density to be processed after an object having a high density will have the same binarization processing result as the background. Depending on the processing target, binarization processing is appropriately performed, such as inconsistent binarization processing results in which the binarization results differ even though the same density continues in the same object. It may not be.

  The present invention has been made in view of the above problems, and an image processing technique capable of appropriately binarizing each object in a multi-valued image when binarizing the multi-valued image with a threshold value. The purpose is to provide.

In order to achieve the above object, an image processing apparatus according to the present invention comprises the following arrangement. That is,
An image processing apparatus comprising a plurality of binarization means for binarizing an input multilevel image,
Edge pixel determination means for determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting an edge part of an object included in the multi-valued image;
In-edge pixel determination means for determining whether or not a predetermined pixel of the multi-valued image is a pixel in the object;
Background pixel determination means for determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting a background area other than the object,
The plurality of binarizing means is determined depending on whether a predetermined pixel of the multi-valued image is a pixel constituting the edge portion, a pixel in the object, or a pixel constituting the background region. It is characterized by switching and using.

  According to the present invention, it is possible to appropriately binarize each object in a multi-valued image when binarizing the multi-valued image with a threshold value.

  Hereinafter, each embodiment of the present invention will be described with reference to the accompanying drawings.

[First Embodiment]
<Configuration of image processing apparatus>
FIG. 1 is a diagram showing a configuration of an image processing apparatus according to the first embodiment of the present invention. In FIG. 1, 103 is a control memory (ROM), 101 is a central processing unit, 102 is a memory (RAM), 106 is an external storage device, 107 is a network I / F, 104 is an input device, 105 is a display device, 108 is It is a bus. A control program for realizing the binarization processing according to the present embodiment is stored in the ROM 103. These control programs and multi-valued images input via the input device 104 are appropriately fetched into the memory 102 through the bus 108 and executed by the central processing device 101 under the control of the central processing device 101.

<Functional configuration of binarization processing>
FIG. 2 is a functional configuration diagram of binarization processing in the image processing apparatus. In FIG. 2, multi-value images are sequentially input in the main scanning and then in the sub-scanning direction, and are processed in units of blocks. Here, processing in units of blocks is a block composed of target pixel data and a plurality of pixel data (peripheral pixel data) arranged in the main scanning direction and the sub-scanning direction so as to surround the target pixel data. Processing for the target pixel data is performed using the target pixel data and the peripheral pixel data, and is generally realized by combining pixel delay and line delay.

  In FIG. 2, reference numeral 210 denotes an edge detection unit, which performs edge detection of the input multi-valued image. Moreover, the correction process with respect to an edge detection result is also performed as needed. As a result, it is possible to distinguish whether the pixel-of-interest data is an edge pixel constituting a line drawing or the like or a non-edge pixel.

  Reference numeral 220 denotes a first threshold value calculation unit, which generates a threshold value (first threshold value) used when binarization processing is performed on pixels determined to be edge pixels.

  A binarization processing unit 230 performs binarization processing on the target pixel data and outputs a binary image. In the binarization process, the processing content is switched based on the edge detection result output from the edge detection unit 210.

  In other words, if the edge detection result output from the edge detection unit 210 is “edge pixel”, the binarization unit 232 uses the first threshold value calculated by the first threshold value calculation unit 220 ( A binarization process is performed using the details described later, and the binarization process result is output.

  On the other hand, when the edge detection result output from the edge detection unit 210 is a “non-edge pixel” and is an in-edge pixel (a pixel in an object, that is, in a region surrounded by edges). The second threshold value (details will be described later) obtained by the second threshold value calculation unit 234 based on the threshold value used when the adjacent pixel data adjacent to the target pixel data is binarized. To perform binarization processing and output the binarization processing result.

  Further, when the edge detection result output from the edge detection unit 210 is a “non-edge pixel” and does not correspond to a pixel within the edge (this pixel is referred to as a “background pixel”), it is set in advance. The predetermined value (background pixel data 233) is output.

  That is, the binarization processing unit 230 determines whether the pixel-of-interest data is an edge pixel, an in-edge pixel, or a background pixel, and switches processing based on the determination result.

<Processing of Edge Detection Unit 210>
FIG. 3 shows an example of a linear filter used in the edge detection unit 210. In FIG. 3A, vertical edge detection is performed, and in FIG. 3B horizontal edge detection is performed. Edge detection is performed by obtaining this value. Here, the filter shown in FIG. 3 is used as the edge detection method, but edge detection using another method such as a Laplacian filter may be used.

  After the edge detection, the edge detection result is corrected as necessary. That is, a case where noise is erroneously detected as an edge or a case where a pixel that should be detected as an edge is not detected as an edge can be considered. Therefore, such erroneous detection is corrected. For example, when making corrections for both false detections, a majority process using the pixel of interest and surrounding pixels is performed, or a thickening process is performed if edge detection is to be performed reliably. Additional processing for reliably detecting edge pixels is performed. In addition to these processes, processes effective for thin line interpolation and isolated point removal may be added.

<Processing of First Threshold Calculation Unit 220>
The first threshold value calculation unit 220 obtains a first threshold value used for the binarization process when it is determined that the target pixel data is an edge pixel. Specifically, the maximum value detection unit 221 and the minimum value detection unit 222 detect the maximum density value and the minimum density value in the block including the target pixel data, respectively, and the calculation unit 223 detects the maximum density value and the minimum density value. An average value of density values is calculated, and this is set as a first threshold value.

  In addition, as a method for obtaining the first threshold, other than this, for example, an average value or a median value obtained from the target pixel data and the adjacent pixel data can be cited, but the data may be biased near the edge. Therefore, it is difficult to say that it is appropriate as a means for calculating the threshold value for the vicinity of the edge. For this reason, in this embodiment, the average value calculated from the maximum density value and the minimum density value is used as described above.

<Processing of the binarization processing unit 230>
Details of processing in the binarization processing unit 230 will be described. In the binarization processing determination unit 231, the following determination is performed based on the edge detection result from the edge detection unit 210.
(1) Whether the target pixel data is an edge pixel.
(2) Whether the target pixel data is a non-edge pixel and is an in-edge pixel in a region surrounded by the edge pixels.
(3) Is the target pixel data a non-edge pixel and a background pixel to be processed as a background?

The criteria for making the above determination are as follows.
(1) If the edge detection result from the edge detection unit 210 for the target pixel data is an edge pixel, it is determined that the target pixel data is an edge pixel.
(2) The edge detection result from the edge detection unit 210 for the target pixel data is a non-edge pixel, and the adjacent pixel data adjacent to the target pixel data is an edge pixel or an in-edge pixel; If the pixel-of-interest data is binarized by the threshold value used when binarizing the adjacent pixel data, if the binarization result is the same as that of the adjacent pixel data, It is determined that the target pixel data is an in-edge pixel.
(3-1) If the edge detection result from the edge detection unit 210 for the target pixel data is a non-edge pixel and the adjacent pixel data of the target pixel data is neither an edge pixel nor an in-edge pixel, the target It is determined that the pixel data is a background pixel.
(3-2) The edge detection result from the edge detection unit 210 for the target pixel data is a non-edge pixel, and the adjacent pixel data of the target pixel data is an edge pixel or an in-edge pixel, and If the pixel-of-interest data is binarized according to the threshold value used when binarizing the adjacent pixel data, if the binarization result is different from the adjacent pixel data, the target pixel It is determined that the pixel data is a background pixel.

Then, the selector unit 235 selects and outputs one of the binarization processing results based on each determination result. That is,
(1) When the pixel-of-interest data is an edge pixel, the binarization unit 232 compares the pixel-of-interest data with the first threshold value obtained by the threshold value calculation unit 220 to obtain 2 It binarizes and outputs the binarization processing result.
(2) When the target pixel data is an in-edge pixel, the binarization unit 232 uses the target pixel data and the adjacent edge pixel (or in-edge pixel) for binarization processing. Based on the threshold value, the second threshold value calculated by the second threshold value calculation unit 234 is compared and binarized, and the binarization processing result is output.
(3) If the target pixel data is a background pixel, it is replaced with a predetermined value (preset background pixel data 233) and output as a binarization result.

  In the above (2), when there are a plurality of adjacent edge pixels and in-edge pixels, the second threshold value calculation unit 234 calculates the average value as the second threshold value. The background pixel data to be replaced with the background pixel may be background pixel data that always becomes a white pixel, or the density of the background pixel is obtained in advance from the multivalued image, and the background pixel data is determined based on the result. Also good.

  In this way, it is determined whether the pixel-of-interest data is an edge pixel, an in-edge pixel, or a background pixel, and the binarization processing is switched based on the determination result, so that In addition, even when an object with a low density exists immediately after an object with a high density, each object can be appropriately binarized.

  In addition, since the threshold value used when binarizing the pixels within the edge is the threshold value used when binarizing the pixel adjacent to the pixel in question, the threshold value within each edge surrounded by the edges is used. Pixels are binarized with substantially the same threshold value, and even if the area (object) surrounded by the edge is filled with the same density, the area is surrounded by the surrounding edges. Binarization processing can be performed with the same threshold value as that of the pixel. That is, it is possible to avoid the occurrence of inconsistent binarization processing results in the object as in the conventional case.

  Hereinafter, the detailed flow of processing in the binarization processing unit 230 will be described with reference to FIG. FIG. 4 is a flowchart showing the flow of processing in the binarization processing unit 230.

  First, in step S401, it is confirmed whether the target pixel data is an edge pixel. If the target pixel data is an edge pixel, the binarization unit 232 performs binarization processing using the first threshold value calculated by the first threshold value calculation unit 220 in step S402. Do.

  On the other hand, if the target pixel data is not an edge pixel, the process proceeds to step S403. In step S403, it is determined whether adjacent pixel data of the target pixel data is an edge pixel or an in-edge pixel. However, since the determination is performed on sequentially input multi-value images, it is not possible to refer to a multi-value image that has not yet been input. That is, the adjacent pixel data refers to the pixels on the previous line for which the binarization processing has already been completed and the previous pixels for which the processing has been completed immediately before the target pixel data, and is determined with reference to the pixels.

  If it is determined in step S403 that the adjacent pixel data is not an edge pixel or an in-edge pixel, the process proceeds to step S404, where the target pixel data is determined as a background pixel, and predetermined background pixel data (233) is binarized. Output as a result. On the other hand, if it is determined that the adjacent pixel data is an edge pixel or an in-edge pixel, the process proceeds to step S405.

  In step S405, the pixel-of-interest data is binarized using the threshold value (first or second threshold value) when the adjacent pixel data (= the previous line pixel and the previous pixel) is binarized. In this case, it is determined whether the result is the same as the binarization processing result of the adjacent pixel data (that is, whether the target pixel data is an in-edge pixel). If the same result is obtained, it is determined that the pixel-of-interest data is an in-edge pixel, and the process proceeds to step S407. On the other hand, if the same result is not obtained, the process proceeds to step S406.

  In step S406, it is determined whether the diagonally forward pixel (the pixel on the front line side and the front pixel side of the target pixel data) is an edge pixel or an in-edge pixel. Accordingly, even when the diagonally previous pixel is an edge pixel or an in-edge pixel, the process proceeds to step S407 assuming that the target pixel data may be an in-edge pixel. On the other hand, if the diagonally previous pixel is neither an edge pixel nor an in-edge pixel, the pixel-of-interest data is assumed to be a background pixel, and the process proceeds to step S404.

  In step S407, the threshold values of the adjacent pixel data (= the previous line pixel and the previous pixel) adjacent to the target pixel data are averaged, and the value is set to 2 as the threshold value (second threshold value) of the target pixel data. Perform value processing.

  Thereafter, in step S408, it is determined whether or not the binarization processing result in step S407 is different from the background pixel data (233) preset as the background pixel. If it is determined in step S408 that they are different from each other, the process proceeds to step S410, and binarization processing is performed using the second threshold value with the pixel of interest as an in-edge pixel.

  On the other hand, if it is determined in step S408 that they are the same, the binarization processing result is compared with the binarization processing result of the adjacent pixel data in step S409. Based on this comparison, it is determined whether the determination that the pixel-of-interest data is a background pixel is a correct determination. That is, if it is determined in step S408 that they are the same, it is determined that the target pixel data is likely to be background pixel data, and the process proceeds to step S409. However, in step S409, the target pixel data is binarized. Is the same as the binarization processing result of the adjacent pixel data, it is determined again as an in-edge pixel, and the process proceeds to step S410 to binarize using the second threshold as the in-edge pixel. Process.

  On the other hand, if it is determined in step S408 that they are the same, and even in step S409, if the result of the binarization processing of the adjacent pixel data is different, the process proceeds to step S404, and the target pixel data is determined as the background pixel. The predetermined background pixel data is output as a binarization result.

  As is clear from the above description, according to the present embodiment, the edge of each object in the multi-valued image is detected, and the target pixel data is an edge pixel, an in-edge pixel, or a background pixel. By determining whether or not there is a configuration in which the binarization processing is switched based on the determination result, the binarization processing can be appropriately performed for each object.

  The second threshold value when binarizing the pixels in the edge is the threshold value (first or second threshold value) used when binarizing the adjacent pixel data adjacent to the pixel. Since the threshold value calculated based on the (value) is used, each pixel within the edge (each pixel within the object) surrounded by the edge is binarized with substantially the same threshold value. Even when the area surrounded by the edge is filled with the same density, it is possible to perform binarization processing in the area with the same threshold value as the surrounding edge pixels. That is, it is possible to avoid inconsistent processing results within the same object.

[Second Embodiment]
In the first embodiment, binary processing is performed so that a consistent binarization processing result is obtained in the same object (that is, when the target pixel data is determined to be an edge pixel or an in-edge pixel). It was decided to perform the conversion process. Specifically, the edge detection result from the edge detection unit 210 for the target pixel data is a non-edge pixel, and the adjacent pixel data of the target pixel data is an edge pixel or an in-edge pixel, and If the pixel-of-interest data is binarized by the threshold value used when binarizing the adjacent-pixel data, the pixel-of-interest is obtained when a binarization result different from the adjacent-pixel data is obtained. The data was determined to be background pixels. If it is determined that the pixel-of-interest data is a background pixel, the pixel-of-interest data is converted from the background pixel data without performing binarization processing (comparison with the first or second threshold value). I decided to replace it.

  However, the multi-value image to be processed does not necessarily include only objects that are filled with the same density.For example, as shown in the table, the periphery is composed of ruled lines, and the background pixels are included inside. It is possible that an object such as a character mixed with edge pixels is included.

  For this reason, when it is determined that the pixel is a background pixel according to the above criteria, if it is easily replaced with the background pixel data, a comparison with the threshold value is executed when the internal edge pixel is erroneously recognized as the background pixel. Without being replaced by the background pixel data. Therefore, it is necessary to carefully determine whether or not the target pixel data is a background pixel in the region (object) surrounded by the edges.

  Therefore, in this embodiment, in determining the background pixels in the region (object) surrounded by the edges, it is a requirement that the background pixels are connected in addition to the above-described determination criteria. Specific processing contents will be described below.

  FIG. 7 is a diagram showing a flow of processing in the binarization processing unit according to the present embodiment. Here, if the target pixel data is a background image, a background counter that is counted up, otherwise stopped or cleared is newly arranged, and whether the background pixel is connected or not is determined. The determination is based on the counter value of the counter.

  As in FIG. 4, in step S401, it is confirmed whether or not the target pixel data is an edge pixel. If the pixel-of-interest data is an edge pixel, the binarization unit 232 performs binarization processing using the first threshold value calculated by the threshold value calculation unit 220 in step S702, Clear the background counter.

  On the other hand, if the target pixel data is not an edge pixel, the process proceeds to step S403. In step S403, it is determined whether adjacent pixel data adjacent to the target pixel data is an edge pixel or an in-edge pixel.

  If it is determined in step S403 that the adjacent pixel data is not an edge pixel or an in-edge pixel, the process proceeds to step S704, where the target pixel data is determined as a background pixel, and predetermined background pixel data is output as a binarization result. At the same time, the background counter is counted up. On the other hand, if it is determined that the adjacent pixel data is an edge pixel or an in-edge pixel, the process proceeds to step S405. The processing from step S405 to step S408 is the same as that in FIG.

  If it is determined in step S408 that the binarization processing result of the target pixel data is different from the binarization processing data of the background pixel data, the process proceeds to step S710, and the second pixel is set as the in-edge pixel. The binarization process is performed using the threshold value and the background counter is cleared.

  On the other hand, if it is determined in step S408 that the values are the same, the binarization processing result is compared with the binarization processing result of the adjacent pixel data in step S409. If it is different from the binarization processing result of the adjacent pixel data in step S409, the process proceeds to step S711, and it is determined whether the background counter is equal to or greater than a certain value. If not, the process proceeds to step S710, where the pixel-of-interest data is a pixel within the edge, and is binarized with the second threshold value, and the background counter is cleared.

  On the other hand, if the background counter is greater than or equal to a certain value in step S711 (that is, if the background pixels are connected), the process proceeds to step S704, where the target pixel data is the background pixel and the background pixel is determined. Replace with data and advance background counter.

<Example>
FIG. 6A shows a processing result when the above binarization processing is applied to the multi-valued image shown in FIG. In FIG. 6, “A” is a portion that is binarized as an edge pixel, “B” is a pixel within the edge but counted up because it is the same as a background pixel, and “C” is a background count that is equal to or greater than a certain value. As a result, a portion to be a background pixel, “D” is a portion to be binarized as an in-edge pixel, “E” is not an edge pixel, and a background whose adjacent pixel data is neither an edge pixel nor an in-edge pixel Part. Further, the threshold value generated for each pixel in the binarization processing unit 230 is as shown in FIG. 6B, and as a result, a good binarization processing result as shown in FIG. 6C is obtained. be able to.

  As described above, the binarization process is performed when it is determined that the pixel-of-interest data is a pixel in the region (object) surrounded by the edge (when the process proceeds to “YES” in step S403). Even if the result is the same as the background pixel, unlike the adjacent pixel data, the edge pixel is mistakenly recognized as the background pixel by not determining it as the background pixel unless the background counter is greater than a certain value. In addition, it is possible to avoid the replacement with the background pixel data without performing the comparison with the threshold value. As a result, appropriate binarization processing can be realized even if the object is composed of ruled lines around the inside and contains objects such as background pixels and edge pixels such as characters. Is possible.

[Third Embodiment]
In the first and second embodiments, the processing is performed by software. However, the present invention is not limited to this, and can be realized by hardware.

  FIG. 8 is a diagram illustrating an example of a hardware configuration in a case where binarization processing is performed by hardware on multi-valued images that are continuously input.

  In FIG. 8, reference numeral 801 denotes an edge detection circuit having a block width (window width) of 5 pixels. The vertical edge detection and horizontal edge detection shown in FIGS. 3A and 3B are performed, and the edge detection detected as a result is performed. The logical sum of the results is taken and a 1-bit edge flag is output for each pixel.

  Reference numeral 802 denotes a majority circuit having a window width of 5 pixels adopted as an edge detection / correction unit, and compares pixels determined to be edge pixels out of 25 pixels centered on the target pixel with pixels not determined to be edge pixels. The larger one is output as the final edge flag of the pixel of interest.

  Reference numeral 803 denotes a first threshold value calculation circuit. In FIG. 8, a maximum value and a minimum value are obtained from pixels with a window width of 5 pixels centered on the target pixel, and the average is output as a threshold value. . Reference numeral 804 denotes a binarization processing circuit, the operation of which is the same as the flowchart shown in FIG.

[Other embodiments]
Note that the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, and a printer), and a device (for example, a copying machine and a facsimile device) including a single device. You may apply to.

  Another object of the present invention is to supply a storage medium recording a program code of software for realizing the functions of the above-described embodiments to a system or apparatus, and the computer (or CPU or MPU) of the system or apparatus stores the storage medium. Needless to say, this can also be achieved by reading and executing the program code stored in.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

  As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, nonvolatile memory card, ROM, or the like is used. be able to.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (operating system) operating on the computer based on the instruction of the program code. It goes without saying that a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

1 is a diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is a functional block diagram of the binarization process in an image processing apparatus. It is a figure which shows an example of an edge detection filter. It is a flowchart which shows the flow of a process in a binarization process part. It is a figure which shows an example of the conventional binarization process. It is a figure which shows an example of the binarization process by the image processing apparatus concerning the 2nd Embodiment of this invention. It is a flowchart which shows the flow of a process in a binarization process part. It is a figure which shows an example of the binarization processing circuit by hardware.

Explanation of symbols

210 Edge Detection Unit 220 First Threshold Calculation Unit 230 Binarization Processing Unit 231 Binarization Processing Determination Unit 232 Binarization Unit 233 Background Pixel Data 234 Second Threshold Calculation Unit 235 Selector Unit 801 Edge Detection Circuit 802 majority decision circuit 803 first threshold value calculation circuit 804 binarization processing circuit

Claims (12)

An image processing apparatus comprising a plurality of binarization means for binarizing an input multilevel image,
Edge pixel determination means for determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting an edge part of an object included in the multi-valued image;
In-edge pixel determination means for determining whether or not the predetermined pixel of the multi-valued image is a pixel in a region surrounded by the edge;
Background pixel determination means for determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting a background area other than the object,
Depending on whether the predetermined pixel of the multi-valued image is a pixel constituting the edge part, a pixel in an area surrounded by the edge, or a pixel constituting the background area An image processing apparatus characterized by switching binarization means. The edge pixel determination means includes
The determination as to whether or not the predetermined pixel is an edge region based on a density gradient between the predetermined pixel of the multi-valued image and a plurality of pixels surrounding the predetermined pixel. Item 8. The image processing apparatus according to Item 1. The in-edge pixel determination means includes
In the edge pixel determination means, it is determined that the predetermined pixel of the multi-valued image is not a pixel constituting the edge portion,
When a pixel adjacent to a predetermined pixel of the multi-valued image is a pixel that constitutes the edge portion or is in a region surrounded by the edge, and the adjacent pixel is binarized If the predetermined pixel is binarized according to the threshold value used in the above, if the binarization result is the same as that of the adjacent pixel, the predetermined pixel is within the region surrounded by the edge. The image processing apparatus according to claim 1, wherein the image processing apparatus determines that the pixel is located in the area. The background pixel determining means includes
In the edge pixel determination means, it is determined that the predetermined pixel of the multi-valued image is not a pixel constituting the edge portion,
When the predetermined pixel of the multi-valued image is determined not to be a pixel within the region surrounded by the edge, the predetermined pixel is a pixel constituting the background region. The image processing apparatus according to claim 1, wherein the image processing apparatus is determined to be present. The background pixel determining means includes
In the edge pixel determination means, it is determined that the predetermined pixel of the multi-valued image is not a pixel constituting the edge portion,
When a pixel adjacent to a predetermined pixel of the multi-valued image constitutes the edge portion or is in a region surrounded by the edge, and binarization processing is performed on the adjacent pixel If the predetermined pixel is binarized according to the threshold value used, if the binarization result is different from that of the adjacent pixel, the predetermined pixel is a pixel constituting the background region. The image processing apparatus according to claim 1, wherein the determination is performed. The background pixel determining means determines that the predetermined pixel is a pixel constituting the background area when a predetermined number or more of the pixels determined to be the pixel constituting the background area are connected. The image processing apparatus according to claim 5, wherein The binarization means includes
When it is determined that the predetermined pixel of the multi-valued image is a pixel constituting the edge portion, a threshold value calculated based on the density of a plurality of pixels surrounding the predetermined pixel is used. The image processing apparatus according to claim 1, wherein binarization processing is performed. The binarization means includes
When the predetermined pixel of the multi-valued image is determined to be a pixel in the region surrounded by the edge, it was used when the pixel adjacent to the predetermined pixel was binarized The image processing apparatus according to claim 1, wherein binarization processing is performed using a threshold value calculated based on the threshold value. The binarization means includes
The predetermined pixel of the multi-valued image is replaced with a predetermined density when it is determined that the predetermined pixel of the multi-valued image is a pixel constituting the background region. Image processing device. An image processing method comprising a plurality of binarization steps for binarizing an input multilevel image,
An edge pixel determination step of determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting an edge part of an object included in the multi-valued image;
An in-edge pixel determination step for determining whether or not a predetermined pixel of the multi-valued image is a pixel in a region surrounded by the edge;
A background pixel determination step for determining whether or not the predetermined pixel of the multi-valued image is a pixel constituting a background region other than the object,
Depending on whether the predetermined pixel of the multi-valued image is a pixel constituting the edge part, a pixel in an area surrounded by the edge, or a pixel constituting the background area An image processing method, wherein the binarization process is switched and used. A storage medium storing a control program for realizing the image processing method according to claim 10 by a computer. A control program for realizing the image processing method according to claim 10 by a computer.

Images