JP2017147485A - Image processing device and region determination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To discriminate a line pair region having a line pair pattern representing a half tone more surely from the prior art.SOLUTION: An image processing device 5 includes resolution converting means 521 for reducing an original image D2 which is an image read from a document to generate a reduced image D2s, binarizing means 522 for binarizing the original image and the reduced image, respectively, counting means for counting variation points having different pixel values between adjacent pixels in a pixel matrix for each of the binarized original image and reduced image, and discrimination means 524 for discriminating the pixel matrix in the original image as a line pair region when the count value M1 of the variation point of the original image is not less than a first threshold value th1 or when a resolution error which is the difference between the count value M1 of the variation point of the original image and the count value M2 of the variation point of the reduced image is not less than a second threshold value th2.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an image processing apparatus and a region discrimination method.

  Image forming apparatuses such as copiers and MFPs (Multi-functional Peripherals) divide input images into, for example, character areas and halftone areas in order to improve the quality of output images. Then, image processing suitable for the area is performed. For this reason, when the input image is an image read from a document, area determination is performed to determine what area each part of the input image is based on the image data of the input image.

  As a prior art for discriminating a halftone area printed using a halftone screen in an image of a document, there is a technique described in Patent Document 1. Japanese Patent Application Laid-Open No. 2004-133830 discloses determining whether a line pair pattern is included in an image of a document based on a difference in the number of line screen patterns (linear screen patterns) in two directions orthogonal to each other. ing. The line pair pattern is a pattern in which line screen patterns are regularly arranged.

  In Patent Document 2, in order to determine whether or not the area is suitable for resolution conversion, the resolution of the image data input in units of tiles is converted to a lower resolution, and before and after the resolution conversion. It describes that the difference between image data is calculated.

  On the other hand, there is a technique described in Patent Document 3 as a prior art for forming an image by electrophotography. Patent Document 2 describes that the number of black lines is counted, and whether or not the black lines are in a black background area is determined depending on whether the count value is larger than a predetermined value.

JP 2009-232444 A JP 2008-042686 A JP 2014-099767 A

  With the improvement of printing technology, the number of lines on the halftone screen has been increased, and the line pair pattern appearing on the printed matter has become high definition. For this reason, it has been difficult for the conventional technique disclosed in Patent Document 1 to correctly determine the line pair region.

  For example, theoretically, 300 (Nyquist frequency) is the limit as the number of line pairs per inch that can be resolved by a scanner capable of reading an image with a resolution of 600 dpi. When a line pair region in which 300 or more line patterns per inch are arranged is read, the line pitch in the read image (line pattern array cycle) becomes larger than the line pitch of the document image. That is, a phenomenon called “false resolution” occurs in which the line pair pattern is read more coarsely than it actually is. If the fineness of reading is reduced from 600 dpi to, for example, 300 dpi in order to increase the reading speed, the Nyquist frequency becomes 150, and the reading is performed more coarsely.

  When false resolution occurs, the number of line pair patterns in the read image is smaller than the actual number (the number in the document). For this reason, an area that should not be determined as a line pair area (for example, an area that includes small-sized characters having parallel line segments) and an area that should be determined as a line pair area are determined depending on the number of line pairs. The problem of being indistinguishable has arisen.

  Since the techniques of Patent Documents 2 and 3 are not for discriminating the line pair region, this problem cannot be solved.

  The present invention has been made in view of the above problems, and an object of the present invention is to more reliably discriminate a line pair region having a line pair pattern representing a halftone than in the past.

  An image processing apparatus according to an embodiment of the present invention is an image processing apparatus that processes an image read from a document, and includes a resolution conversion unit that reduces the original image that is the image to generate a reduced image, and the original A binarization unit that binarizes each of the image and the reduced image, and each of the binarized original image and the binarized reduced image between adjacent pixels in a pixel matrix of a predetermined size. Counting means for counting change points having different pixel values, and a count value of the change point of the binarized original image is equal to or greater than a first threshold value or the binarized original image When the resolution error that is the difference between the count value of the change point and the count value of the change point of the binarized reduced image is greater than or equal to a second threshold value, the pixel matrix in the original image is Multiple la Has a determination means for determining a line pair region where down-like pattern arranged, the.

  According to the present invention, it is possible to more reliably discriminate a line pair region having a line pair pattern representing a halftone than in the prior art.

1 is a diagram illustrating an outline of a hardware configuration of an image forming apparatus including an image processing apparatus according to an embodiment of the present invention. It is a figure which shows the example of a functional structure of an image processing apparatus. It is a figure which shows the example of the read image read from the original. It is a figure which shows the several line pair pattern from which a line period differs, and the pattern which read them by switching reading resolution. It is a graph which shows the relationship between the line period in a manuscript image, and the number of line pairs in a reading image about several reading resolution. It is a figure which shows the number of line pairs in the read image corresponding to each of the several value of the line period in a document image by a table format about several reading resolution. It is a figure which shows the example of the table for setting the reduction rate of resolution conversion, a 1st threshold value, and a 2nd threshold value. It is a figure which shows the example of the direction of the count of the change point in a binarized image, and the example of the filter used for a count. It is a figure which shows the flow of a line pair area | region discrimination | determination process.

  FIG. 1 shows an outline of a hardware configuration of an image forming apparatus 1 including an image processing apparatus 5 according to an embodiment of the present invention, FIG. 2 shows an example of a functional configuration of the image processing apparatus 5, and FIG. Each shows an example of a read image 8 read from the document 7.

  1 to 3, the image forming apparatus 1 is an MFP in which functions such as a copier, a printer, and a facsimile machine are integrated. The image processing apparatus 1 includes an image reading unit 2, an image processing device 5, and a printer unit 6.

  The image reading unit 2 includes an automatic document feeder (ADF) 3 and a scanner 4. The automatic document feeder 3 conveys one or a plurality of sheet-like documents 7 set on a paper feed tray to a paper discharge tray, and reads a document image from the document 7 into pixels while reading. The scanner 4 reads a document image from a document 7 placed on a platen glass. The automatic document feeder 3 and the scanner 4 output image data D1 representing the gradation value of each pixel of the read image (read image) 8 to the image processing device 5.

  The image reading unit 2 reads a document image at a reading resolution of 600 dpi in the normal mode. When the high-speed mode is selected instead of the normal mode, reading is performed by switching the reading resolution to a value of 500 dpi or less, which is smaller than 600 dpi.

  Based on the image data D1, the image processing apparatus 5 generates print data D4 for printing an image corresponding to the original image by the printer unit 6. At that time, the image processing apparatus 5 performs various image processes including a line pair area discrimination process described later.

  The printer unit 6 is an image forming unit that prints an image on a sheet based on the print data D4 input from the image processing device 5. The printer unit 6 includes a printer engine that forms a full-color or monochrome print image by, for example, electrophotography, and a paper feed unit that supplies paper to the printer engine.

  As shown in FIG. 2, the image processing apparatus 5 includes an image data adjustment unit 51, a line pair region determination unit 52, another region determination unit 53, a gradation conversion unit 54, a print data generation unit 55, and the like. .

  Each component of the image processing device 5 is realized by a hardware configuration such as a CPU (Central Processing Unit), a memory, and other circuit elements, and by a predetermined program being executed by the CPU. The whole or part of the image processing apparatus 5 may be integrated and used as a microcomputer or an application specific integrated circuit (ASIC).

  The image data adjustment unit 51 performs background removal processing for subtracting the background level (background signal level) from the image data D1, and color system conversion for converting the RGB image data D1 into YMCK image data D2. Apply processing.

  The line pair area determination unit 52 determines a line pair area in the read image 8. A line pair area is formed by regularly arranging line patterns (line-shaped patterns) 31 as in an area 823 shown in FIG. A line pair (lp) 30 composed of a portion (base portion) appears repeatedly, and is defined as a region where the repetition period (line period H) is equal to or less than a set value.

  The line pair region determination unit 52 includes a resolution conversion unit 521, a binarization unit 522, a count unit 523, and a determination unit 524. The functions of these units will be described later.

  The other region discriminating unit 53 discriminates a character region representing a character in the read image 8 and a halftone dot region in which isolated points are regularly arranged.

  The gradation conversion unit 54 performs edge enhancement processing, smoothing processing, and the like on the image data D2 based on the determination results of the line pair region determination unit 52 and the other region determination unit 53, and outputs the processed image data D3. The data is sent to the print data generation unit 55. The edge emphasis process is a process that emphasizes the edge of a character in order to improve visibility, and the smoothing process is performed on pixels in a halftone area so that a change in lightness (density) between adjacent pixels becomes smooth. This is a process for setting a lightness value or a density value. This smoothing process prevents moiré in the printed image. There are a line pair area and a halftone area as halftone areas to which smoothing processing is applied.

  The print data generation unit 55 generates, for example, binary print data D4 by performing screen processing or error diffusion processing on the multi-value image data D3.

  In FIG. 3A, the read image 8 has a halftone area 82. The halftone area 82 includes a relatively light area 821 and a relatively dark area 823.

  The light area 821 is a halftone dot area in which a large number of isolated points 21 are regularly arranged vertically and horizontally as shown in FIG. That is, the portion corresponding to the light area 821 in the original image is printed by using a halftone screen technique. In FIG. 3B, the screen angle α of the halftone screen is approximately 55 degrees.

  The dark region 823 is the line pair region described above as shown in FIG. The extending direction dx of each line pattern 31 is substantially equal to the direction dv of the isolated points 21 in the light area 821, and the pitch Px in which the line patterns 31 are arranged is substantially equal to the pitch Pv of the isolated points 21. In other words, the portion corresponding to the dark area 823 in the original image is also printed using a technique that represents a halftone with a halftone screen. A dark area 823 is a halftone area to be subjected to the smoothing process.

  When a halftone screen having a large isolated point size is used to represent a dark gradation, the isolated points are connected to form a line pattern 31 depending on printing conditions such as the paper quality of the document 7, and the halftone pattern is shown in FIG. A line pair pattern like the region 823 of 3 (C) may be obtained.

  FIG. 4 shows a plurality of line pair patterns having different line periods H and patterns read by switching the reading resolution. FIG. 5 shows the relationship between the line period H in the original image and the number M of line pairs in the read image 8 for a plurality of reading resolutions. FIG. 6 shows the number M of line pairs in the read image 8 corresponding to each of a plurality of values of the line period H in the document image for a plurality of reading resolutions in a table format.

  In FIG. 4, a chart on which nine line pair patterns having a line period H value of 2 to 10 lp / mm are printed is shown in the uppermost row as a pattern of the document 7. The result (read image) obtained by switching the line pair pattern while switching the reading resolution in the range of 600 to 200 dpi is described in the 5 × 9 squares surrounded by the thick line.

  The line period H is the number (number) of line pairs per 1 mm, and is an indicator of the fineness of the document 7. In FIG. 4, the values of the line period H are 2, 3, 4, 5, 6, 7, 8, 9, and 10. There are five reading resolutions: 600 dpi, 500 dpi, 400 dpi, 300 dpi, and 200 dpi.

  The number in each square is the number M of line pairs per predetermined number of pixels (40 pixels in this example). In addition, the numbers enclosed in parentheses described in the squares of the 600 dpi line are theoretical values of the number M of line pairs. For example, when the line period H is 2, the number M of line pairs per inch of the document 7 is 25.4 × 2 = 50.8. Since the size of one pixel of 600 dpi is 25.4 mm ÷ 600≈0.042 mm, the theoretical value of the number M of line pairs per 40 pixels is 50.8 × (0.042 mm × 40 ÷ 25.4 mm) ≈3 .4.

  The following can be understood with reference to FIG.

  If the reading resolution is 600 dpi or more, the number M of line pairs is almost equal to the theoretical value (the difference between the two is 1 or less) regardless of the line period H being 2 to 10. And the line pair number M is so large that the line period H is large. This means that the line pair region can be determined by a conventional method of comparing the number M of line pairs with a predetermined threshold value. For example, when it is desired to discriminate an area including a line pair pattern having a line period H of 8 or more as a line pair area, the threshold value is set to “12”, for example, and the number M of line pairs is 12 or more. A line pair area is determined, and when the number M of line pairs is less than 12, it may be determined that the area is not a line pair area.

  However, when the line pair pattern is read at a reading resolution of 500 dpi or less, false resolution may occur depending on the value of the line period H of the line pair pattern.

  That is, when reading at 500 dpi, false resolution occurs when the line period H is 8 or more, and when reading at 400 dpi, the line period H is 7 or more. When reading at 300 dpi, false resolution occurs when the line period H is 6 or more, and when reading at 200 dpi, the line period H is 5 or more. In particular, when reading at 200 dpi, it is not possible to resolve line pair patterns with a line period H of 8 or more as grayscale patterns, and the number M of line pairs becomes 0 (zero).

  As described above, when reading is performed at a reading resolution of 500 dpi or less, which may cause false resolution, the line pair region is determined by a conventional method depending on which range of the line period H is set as the line pair region. There are cases where it can be determined and cases where it cannot be determined.

  For example, when reading at 500 dpi, by setting the threshold value to “9”, for example, an area including a line pair pattern with a line period H of 6 or more can be determined as a line pair area. However, it is impossible to determine only a region including a line pair pattern having a line period H of 7 or more as a line pair region. This is because the number M of line pairs in an area including a line pair pattern having a line period H of 6 (desired to be other than the line pair area) and an area including an line pair pattern having a line period H of 10 (desired as a line pair area). This is because the number of pairs M is the same 10, and the number M of line pairs of both is not large and the threshold value cannot be set.

  Incidentally, as described above, in the high-speed mode, reading is performed with a reading resolution of 500 dpi or less. The image processing apparatus 5 can determine the line pair region even in the high speed mode. The mechanism of this function will be described below.

  FIG. 7 shows an example of a table T2 for setting the resolution conversion reduction ratio R, the first threshold th1, and the second threshold th2. FIG. 8 shows an example of count directions d1 to d12 of change points in a binarized image and examples of filters Fd3 and Fd12 used for counting.

  Referring also to FIG. 2, the resolution conversion unit 521 of the line pair region determination unit 52 reduces the original image based on the image data D2 input as the original image to generate reduced image data D2s. At that time, based on the contents of the table T2 shown in FIG. 3, the resolution (reading resolution) of the original image according to the lower limit value of the line period H set as the condition of the area to be determined as the line pair area. The reduction ratio R is determined according to the above.

  For example, when the lower limit of the line period H is 8, that is, when an area having a line pair pattern with a line period H of 8 or more is defined as a line pair area, and the resolution of the original image is 400 dpi, the reduction ratio The value of R is determined to be 75%. That is, the resolution of the reduced image is set to 300 dpi.

  When the resolution of the original image is 600 dpi, regardless of the setting of the lower limit value of the line period H, the value of the reduction ratio R is set to 100% (same size) so that the resolution conversion is not substantially performed. .

  The reduced image generated in this way corresponds to a read image when the original image is read with the resolution of the reduced image as the read resolution. That is, by reducing the original image, two read images having different reading resolutions can be obtained for the same original image.

  The binarization unit 522 binarizes the image data D2 of the original image and the image data D2s of the reduced image. That is, each pixel value is set to “1” or “0”, for example. Any combination of other numerical values such as “255” and “0” may be used as long as the combination represents a black pixel and a white pixel.

  The count unit 523 performs the inter-adjacent pixels in the pixel matrix 90 of a predetermined size shown in FIG. 8 for each of the binarized original image (image data D21) and the binarized reduced image (image data D21s). The change points with different pixel values are counted. Specifically, the number of change points from 1 to 0 or from 0 to 1 is counted. The count value for the original image is M1, and the count value for the reduced image is M2. The count values M1 and M2 indicate the number of line pairs.

  More specifically, the counting unit 523 has a plurality of directions d1 to d12 including a plurality of mutually orthogonal directions shown in FIG. 8A for at least one of the binarized original image and the binarized reduced image. Count the change points in each. In FIG. 8A, a plurality of directions d1 to d12 are directions that spread radially for each angle obtained by dividing 360 degrees in increments of 30 degrees around the target pixel TG in the pixel matrix 90. Two directions are orthogonal to each of the plurality of directions d1 to d12. For example, directions d3 and d9 are orthogonal to direction d12.

  In counting, a filter provided in advance for each of the plurality of directions d1 to d12 is used. FIG. 8B shows the filter Fd3 in the direction d3, and FIG. 8C shows the filter Fd12 in the direction d12. The filters Fd3 and Fd12 are composed of a set of pixels that continue from the target pixel TG along the corresponding directions d3 and d12. The number of pixels of the illustrated filters Fd3 and Fd12 is 10, but in reality, the number of pixels is larger than this, for example, 40.

  8B and 8C, when the change point from the white pixel to the black pixel is counted as the change point, the count value M1 in the direction d3 in FIG. 8B is “4”. The count value M1 in the direction d12 in (C) is “0”.

  The count unit 523 has the largest difference in the change point count value M1 (or M2) from the orthogonal direction among the plurality of directions d1 to d12, and the change point count value M1 (or M2) than the orthogonal direction. ) Is the direction for counting. The count direction for determination is a direction corresponding to the count values M1 and M2 evaluated by the determination unit 524.

  In addition, the count unit 523 has a count value M1 (or M2) in a certain direction of “0” and a count value M1 (or M2) in one of two directions orthogonal to the direction is 1 or more. In this case, the larger count value M1 (or M2) of the two orthogonal directions is set as the determination count direction.

  That is, since the line pair pattern is composed of continuous straight lines, the count value M1 (or M2) in the direction parallel to the straight line is 0 or a value close thereto, and the count value M1 in the direction orthogonal to this direction (or M2) is maximized. Based on this, the direction orthogonal to the direction in which the count value M1 (or M2) is the smallest is taken as the determination count direction.

  After any one of the plurality of directions d1 to d12 is set as the determination count direction, the count value M1, of the change point in the determination count direction for both the binarized original image and the binarized reduced image M2 is output to the determination unit 524. For example, when 12 change points in the directions d1 to d12 are counted only for the original image and the discrimination count direction is specified, only the change points in the discrimination count direction are counted for the reduced image. The value M2 is obtained. Since the direction of the line pair does not change even when the image is reduced, counting in 11 directions other than the determination count direction for the reduced image can be omitted.

  In this way, by counting the change points in the plurality of directions d1 to d12 and specifying the counting direction for determination, the screen angle α shown in FIG. Even if a document image is read from the document 7, the line pair area can be determined.

  The determination unit 524 binarizes the count value M1 of the change point of the binarized original image that is equal to or greater than the first threshold th1 or the binarization value of the change point of the original image M1. When the resolution error dM, which is the difference from the count value M2 of the changed point of the reduced image, is equal to or greater than the second threshold th, the pixel matrix 90 in the original image is determined as the line pair region.

  The first threshold th1 is a threshold for discriminating the line pair region by the conventional method described above, and is predetermined as shown in the table T2 of FIG.

  For example, when it is desired to discriminate an area having a line period H of 8 or more as a line pair area, and the resolution (reading resolution) of the original image is 400 dpi, the first threshold th1 is “11”. .

  4 and 5, in the case of 400 dpi, the largest line pair number M within the range of the line period H of 2 to 7 is “10”. Accordingly, since the first threshold th1 is larger than “10”, when the count value M1 of the original image (that is, the number of line pairs per 40 pixels) is equal to or greater than the first threshold th1, (Pixel matrix 90) is a line pair region.

  The second threshold value th2 is a threshold value used when the count value M1 of the original image is less than the first threshold value th1, and is predetermined as shown in the table T2. For example, as in the above example, when it is desired to discriminate an area having a line period H of 8 or more as a line pair area and the resolution (reading resolution) of the original image is 400 dpi, the second threshold th2 Is “4”.

  4 and 5, the largest value of the resolution error dM between the number of line pairs in the case of 400 dpi and the number of line pairs in the case of 300 dpi in the range of the line period H of 2 to 7 is the line period H "1" when is 6 or 7. On the other hand, the resolution error dM in the range where the line period H is 8 or more is 5 or more. Specifically, when the line period H is 8, 13-7 = 6, when the line period H is 9, 9-4 = 5, and when the line period H is 10, 8-3 = 5. . Accordingly, since the second threshold th2 is larger than the maximum value “1” of the resolution error dM in the range of the line period H of 2 to 7, the original image, the reduced image, and the resolution error dM are the second threshold. When the value is equal to or greater than th2, the discrimination target area (pixel matrix 90) is a line pair area.

  FIG. 9 shows the flow of the line pair area discrimination process. The line pair area determination process is executed by the line pair area determination unit 52.

  Resolution conversion for reducing the original image is performed to generate a reduced image (# 11). Each of the original image and the reduced image is binarized (# 12). The change points are counted for each of the original image and the reduced image (# 13).

  If the count value M1 obtained for the original image is greater than or equal to the first threshold th1 (YES in # 14), the discrimination result for the attention area is set to “line pair area” (# 18), and the discrimination result is Output (# 19).

  If the count value M1 is less than the first threshold th1 (NO in # 14), a resolution error dM between the original image and the reduced image is calculated (# 15).

  If the calculated resolution error dM is greater than or equal to the second threshold th2 (YES in # 16), the discrimination result for the attention area is set to “line pair area” (# 18), and the discrimination result is output ( # 19).

  If the calculated resolution error dM is less than the second threshold th2 (NO in # 16), the discrimination result for the attention area is set to “other than the line pair area” (# 17), and the discrimination result is output. (# 19).

  According to the above embodiment, since the original image is reduced and the resolution error dM is intentionally generated to determine the magnitude of the resolution error dM, a line pair having a fine line pair pattern 30 representing a halftone is obtained. The region 823 can be determined more reliably than in the past. Further, the line pair region 823 can be determined no matter which direction the line pair is arranged. Since the smoothing process can be performed on the line pair area 823, which is a halftone area to be subjected to the smoothing process, the printing quality can be improved.

  In the embodiment described above, the lower limit value of the line period H, which is a condition for distinguishing between the line pair region and the other region, is set to 8 or 7. However, it is not limited to this. For example, it can be determined according to the font size of characters included in the document 7. When the original 7 includes small-size characters partially having a fine line pair pattern equivalent to the line pair pattern derived from the halftone screen, the lower limit value of the line period H is made relatively large. When the document 7 is limited to those containing only large characters, the lower limit value of the line period H can be made relatively small.

  With respect to the directions d1 to d12 in which the change points are counted, 360 degrees can be divided more finely than the 12 divisions in FIG. For example, the number of directions may be set to 36 by dividing by 10 degrees. In that case, 36 filters are prepared.

  The configuration of the entire image processing apparatus 5 or each unit, the content, order, or timing of processing, the values of the first threshold th1 and the second threshold th2, and the like can be changed as appropriate in accordance with the spirit of the present invention. it can.

5 Image processing device 7 Document 90 Pixel matrix 521 Resolution converter (resolution converter)
522 Binarization unit (binarization means)
523 Count unit (counting means)
524 discriminating unit (discriminating means)
823 area (line pair area)
D1 Image data (read image)
D2 Image data (original image)
D2s image data (reduced image)
D21 Image data (binarized original image)
D21s Image data (binarized reduced image)
M1, M2 Count value dM Resolution error d1-d12 Direction H Line period th1 First threshold th2 Second threshold R Reduction rate

Claims (5)

An image processing apparatus for processing an image read from a document,
Resolution conversion means for reducing the original image, which is the image, to generate a reduced image;
Binarization means for binarizing each of the original image and the reduced image;
For each of the binarized original image and the binarized reduced image, a counting unit that counts change points having different pixel values between adjacent pixels in a pixel matrix of a predetermined size;
The count value of the change point of the binarized original image is greater than or equal to a first threshold value, or the count value of the change point of the binarized original image and the binarized reduction When the resolution error, which is the difference from the count value of the change point of the image, is greater than or equal to a second threshold value, the pixel matrix in the original image is determined as a line pair region in which a plurality of line-shaped patterns are arranged. Discriminating means to perform,
An image processing apparatus comprising: The resolution converting means reduces the original image at a reduction ratio according to the resolution of the original image;
The image processing apparatus according to claim 1. The resolution conversion unit reduces the original image at a reduction rate according to a lower limit value of a line period set as a condition of the area to be determined as the line pair area.
The image processing apparatus according to claim 1. The counting means counts the change points in each of a plurality of directions including a plurality of directions orthogonal to each other for at least one of the binarized original image and the binarized reduced image. The direction that has the largest difference in the count value of the change point from the orthogonal direction and the direction that the count value of the change point is larger than the orthogonal direction is defined as the count direction for determination, and is binarized. For both the original image and the binarized reduced image, the count value of the change point in the determination count direction is output,
The determination means compares a difference between count values of the change points in the determination count direction with the second threshold value as the resolution error;
The image processing apparatus according to claim 1. An area determination method for determining an area in an image,
Reduce the original image, which is the image read from the document, to generate a reduced image,
Each of the original image and the reduced image is binarized,
For each of the binarized original image and the binarized reduced image, change points where pixel values differ between adjacent pixels in a pixel matrix of a predetermined size are counted,
The count value of the change point of the binarized original image is greater than or equal to a first threshold value, or the count value of the change point of the binarized original image and the binarized reduction When the resolution error, which is the difference from the count value of the change point of the image, is greater than or equal to a second threshold value, the pixel matrix in the original image is determined as a line pair region in which a plurality of line-shaped patterns are arranged. In other cases, the pixel matrix is determined as an area other than the line pair area.
An area discrimination method characterized by the above.

Images