JP2011128878A - Image processor, image processing method, computer program for the same, and recording medium with program recorded thereon - Google Patents

Abstract

An image processing apparatus capable of obtaining an inspection area including an inspection object at high speed is provided.
An inspection object determination unit 33 sets a measurement point in image data according to a thinning amount, and determines whether or not it is a candidate point to be inspected according to an RGB luminance value of each measurement point. . The inspection line setting unit 34 sets a line in which a measurement point determined as a candidate point by the inspection target determination unit 33 exists as an inspection line. The inspection area enlargement unit 35 enlarges the inspection area according to the inspection line set by the inspection line setting unit 34. Then, the inspection area determination unit 36 determines whether or not the inspection area enlarged by the inspection area enlargement unit 35 is an inspection area including an inspection object according to the feature amount. Therefore, it is possible to obtain an inspection area including an inspection object at high speed.
[Selection] Figure 3

Description

  The present invention relates to an image processing technique used for position detection and the like, and more particularly, to an image processing apparatus, an image processing method, a computer program thereof, and a recording medium on which a program is recorded.

  In recent years, image processing techniques have been used in various fields such as mounting and inspection of electronic components. For example, video captured by a camera or video is converted into data, a desired pattern image is searched from the image data, and used for positioning and the like.

  One of such techniques is a method using pattern matching. In general, pattern matching is often performed by scanning the entire target processing area of image data. In pattern matching processing, for example, normalized correlation processing, since the amount of calculation in one matching processing is large, the processing time tends to increase as the number of matching increases. Now that CCD (Charge Coupled Device) cameras have increased in pixel count, an increase in processing time becomes a more serious problem.

  Patent Document 1 aims to appropriately set an area not to be processed in the pattern matching process. The processing region setting method in pattern matching sets processing conditions based on the degree of overlap between the pattern region and the mask region when the pattern region straddles the target region and the mask region at the stage of moving the pattern region. When the moved pattern area is at a position overlapping the mask area, the area ratio of the overlap at that position is calculated, and when this value is larger than a predetermined ratio, it is excluded from processing. Alternatively, how much the pattern area protrudes from the target area is represented by coordinates, and the processing target is up to a predetermined value. A flag is set with the region to be processed determined in this way as a template region.

JP 2003-109002 A

  In the above-mentioned Patent Document 1, only the necessary region is extracted from the image data and the matching process is performed to reduce the number of times of matching, thereby reducing the amount of calculation. However, since the determination when the pattern area extends over the two areas of the mask area and the scan area is processed for each matching, the amount of calculation increases.

  In addition, if the position and orientation of the inspection target vary in the image, the pattern area may overlap the mask area by a set amount or more, and the above determination method may determine that the area is not matched. There was a problem that it could not be extracted.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an image processing apparatus, an image processing method, a computer program thereof, and an image processing apparatus capable of obtaining an inspection area including an inspection object at high speed. It is to provide a recording medium on which a program is recorded.

  According to one aspect of the present invention, there is provided an image processing apparatus for detecting an inspection region including an inspection object from image data, wherein measurement points are set in the image data according to a thinning amount, and RGB values of the respective measurement points are set. An inspection target determining means for determining whether or not a candidate point is an inspection target, and an inspection line setting means for setting a line in which a measurement point determined as a candidate point by the inspection target determining means exists as an inspection line Determining whether the inspection area is enlarged according to the inspection line set by the inspection line setting means, and whether the inspection area enlarged by the inspection area enlargement means is an inspection area including an inspection object. Inspection area discrimination means.

  Preferably, the inspection object determination unit determines whether or not the inspection object candidate point is in accordance with the RGB luminance values of the measurement points.

  Preferably, the inspection object determination unit determines whether or not the inspection object candidate point is in accordance with the hue value of the measurement point.

  Preferably, the inspection line setting means uses the measurement point outside the inspection target when the measurement point changes from the non-inspection measurement point to the inspection target candidate point, and sets the measurement point outside the inspection target from the inspection target candidate point. The point that is not subject to the inspection when the point changes to the falling point is set as the falling point, and the point from the rising point to the falling point is set as the inspection line.

  More preferably, the inspection line setting means sets the same attribute as the inspection line on the one pitch when the start point or the end point of the inspection line on the one pitch exists within the range of the target inspection line. When there is no start point or end point of the inspection line on one pitch, an attribute different from that on the inspection line on one pitch is given to the target inspection line to be a different inspection target.

  Preferably, if the target inspection line is the uppermost line, the inspection area enlarging means sets the range of the target inspection line as the horizontal direction of the inspection area, and sets the inspection area of the target inspection line from 1 pixel to 1 If the target inspection line is not the top line, the target inspection line is compared with the inspection line one pitch up, and the inspection is performed over a wide range. The area is expanded in the horizontal direction, the inspection area of the target inspection line is expanded from one pixel up by one pitch-2 pixels, and if the target inspection line is the bottom line, the target inspection line The inspection area of the target inspection line is enlarged from one pixel below by one pitch-2 pixels below, and the smallest rectangular area including all the enlarged inspection areas is finally obtained. The inspection area.

Preferably, different values are set for the thinning-out amount in the horizontal direction and the vertical direction.
Preferably, the inspection region determination unit determines whether the inspection region enlarged by the inspection region enlargement unit is an inspection region including the inspection target, using the area value of the inspection target as a feature amount.

  Preferably, the inspection area determination unit determines whether the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object, using the aspect ratio of the inspection object as a feature amount.

  Preferably, the inspection area determination unit determines whether or not the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object, using the diagonal length of the inspection object as a feature amount.

  Preferably, when there are a plurality of colors in the inspection object, after obtaining the inspection area according to the predetermined color, the inspection area enlarging means sets the inspection area according to the size or shape of the inspection object. Expand further.

  According to another aspect of the present invention, there is provided an image processing method for causing a computer to detect an inspection region including an inspection object from image data, and causing a computer to set measurement points in image data according to a thinning amount, A step of determining whether or not a candidate point to be inspected according to an RGB value of the measurement point, a step of setting a line in which the measurement point determined to be a candidate point exists as an inspection line, and a set inspection The method includes a step of enlarging the inspection region according to the line and a step of determining whether or not the enlarged inspection region is an inspection region including an inspection object.

  According to still another aspect of the present invention, there is provided a computer program for causing a computer to detect an inspection area including an inspection object from image data, and setting a measurement point in the image data according to a thinning amount in the computer A step of determining whether or not a candidate point to be inspected according to the RGB value of each measurement point, a step of setting a line in which a measurement point determined to be a candidate point exists as an inspection line, and setting A step of enlarging the inspection region in accordance with the inspection line that has been made, and a step of determining whether or not the enlarged inspection region is an inspection region including an inspection object.

  According to still another aspect of the present invention, there is provided a recording medium on which a computer program for causing a computer to detect an inspection area including an inspection object from image data is recorded. To set measurement points, and to determine whether or not a candidate point to be inspected according to the RGB value of each measurement point, and to set a line in which the measurement point determined to be a candidate point exists as an inspection line A step of enlarging the inspection region according to the set inspection line, and a step of determining whether or not the enlarged inspection region is an inspection region including an inspection object.

  According to an aspect of the present invention, the inspection object determination unit sets a measurement point in the image data according to the thinning amount, and determines whether or not the inspection object candidate point is according to the RGB value of each measurement point. Since the determination is made, the inspection area can be obtained at high speed.

  In addition, since the inspection target determination unit determines whether or not the inspection target candidate point is based on the RGB luminance values of the measurement points, it is possible to determine the inspection region according to the change in brightness. .

  In addition, since the inspection target determination unit determines whether or not the inspection target candidate point is based on the hue value of the measurement point, it is possible to determine the inspection region according to the change in brightness.

  In addition, when the inspection line setting means changes from a measurement point that is not the inspection target to a candidate point that is the inspection target, the measurement point that is not the inspection target is set as the rising point, and the measurement point that is not the inspection target is changed to the measurement point that is not the inspection target. Since the point that is not subject to inspection at the time of the change is set as the falling point and the point from the rising point to the falling point is set as the inspection line, the inspection area can be easily detected.

  Further, when the inspection line setting means has the start point or the end point of the inspection line on one pitch within the range of the target inspection line, the inspection line setting means gives the same inspection line as the inspection line on the one pitch to the target inspection line. When the start point or end point of the inspection line on one pitch does not exist, an attribute different from that on the inspection line on one pitch is given to the target inspection line to make it a different inspection target. Even if included, it is possible to detect an inspection region including each inspection object.

  Further, since the inspection area enlarging means sets the minimum rectangular area including all the enlarged inspection areas as the final inspection area, the inspection area can be accurately specified.

  In addition, since different values are set for the thinning amount in the horizontal direction and the vertical direction, it is possible to easily detect inspection objects having various shapes.

  In addition, since the inspection area determination means determines whether the inspection area enlarged by the inspection area enlargement means is an inspection area including the inspection object by using the area value of the inspection object as a feature amount, the inspection area is included. It is possible to easily determine whether or not it is an inspection area.

  In addition, since the inspection area determination means determines whether the inspection area enlarged by the inspection area enlargement means is an inspection area including the inspection object by using the aspect ratio of the inspection object as a feature amount, the inspection area is included. It is possible to easily determine whether or not it is an inspection area.

  Further, since the inspection area determination means determines whether the inspection area enlarged by the inspection area enlargement means is an inspection area including the inspection object by using the diagonal length of the inspection object as a feature amount, the inspection area is included. It is possible to easily determine whether or not the inspection area.

  In addition, when there are a plurality of colors in the inspection object, after obtaining the inspection area according to the predetermined color, the inspection area enlarging means further adds the inspection area according to the size or shape of the inspection object. Since the image is enlarged, even when there are a plurality of colors in the inspection target, it is possible to easily detect the inspection region including the inspection target.

It is a figure which shows the structural example of the test | inspection system using the image processing apparatus in embodiment of this invention. It is a block diagram which shows an example of the hardware constitutions of the image processing apparatus 1 in embodiment of this invention. It is the block diagram which showed the image processing apparatus 1 in embodiment of this invention by the functional structure. It is a flowchart for demonstrating the process sequence of the image processing apparatus in embodiment of this invention. It is a figure for demonstrating inspection object and pitch amount Lx, Ly. It is a figure for demonstrating the inspection line set by the inspection line setting part. It is a figure which shows the inspection line which set the attribute. It is a figure which shows an example of the image data in which the test | inspection area | region was set. It is a figure for demonstrating the method of calculating | requiring the test | inspection area | region in case a some color exists in the test target 42.

  FIG. 1 is a diagram illustrating a configuration example of an inspection system using an image processing apparatus according to an embodiment of the present invention. This inspection system includes an image processing apparatus 1, a camera 2, an optical system 3 fixed to the camera 2, and a mounting stage 5 on which a product 4 to be inspected is mounted.

  The camera 2 is fixed directly above the product 4 mounted on the mounting stage 5 so that the optical system 3 faces the product 4. The optical system 3 is equipped with a lens and an illumination device (not shown).

  The image processing apparatus 1 is connected to a camera 2 and takes an image captured by the camera 2 and processes the image data.

  FIG. 2 is a block diagram illustrating an example of a hardware configuration of the image processing apparatus 1 according to the embodiment of the present invention. The image processing apparatus 1 is realized by a general computer, and includes a computer main body 11, a display device 12, an FD drive 13 in which an FD (Flexible Disk) 14 is mounted, a keyboard 15, a mouse 16, and a CD-. A CD-ROM device 17 on which a ROM (Compact Disc-Read Only Memory) 18 is mounted and a network communication device 19 are included.

  The image processing program is supplied by a recording medium such as the FD 14 or the CD-ROM 18. The image processing program is executed by the computer main body 11 to perform image processing. The image processing program may be supplied to the computer main body 11 from another computer via a communication line.

  The recording medium is not limited to the FD 14 and the CD-ROM 18, but may be a magnetic tape, an MO (Magnetic Optical disk), an MD (Mini Disc), a DVD (Digital Versatile Disc), an IC (Integrated Circuit) card, or the like. May be.

  The computer main body 11 includes a CPU 20, a ROM (Read Only Memory) 21, a RAM (Random Access Memory) 22, and a hard disk 23. The CPU 20 performs processing while inputting / outputting data to / from the display device 12, FD drive 13, keyboard 15, mouse 16, CD-ROM device 17, network communication device 19, ROM 21, RAM 22 or hard disk 23. The image processing program recorded on the FD 14 or the CD-ROM 18 is stored in the hard disk 23 by the CPU 20 via the FD drive 13 or the CD-ROM device 17. The CPU 20 performs image processing by appropriately loading an image processing program from the hard disk 23 into the RAM 22 and executing it.

  FIG. 3 is a block diagram showing a functional configuration of the image processing apparatus 1 according to the embodiment of the present invention. The image processing apparatus 1 includes a threshold setting unit 31, a search interval setting unit 32, an inspection object determination unit 33, an inspection line setting unit 34, an inspection region enlargement unit 35, and an inspection region determination unit 36.

  The threshold value setting unit 31 includes a range of RGB luminance values used when determining whether or not a part is an inspection target, and a range of feature amounts used when determining whether or not the part is an inspection target. Set as threshold. In the present embodiment, the area value to be inspected is used as the feature amount.

  The search interval setting unit 32 sets a thinning amount when searching for image data. This thinning-out amount is set according to the size or shape of the inspection object, and is set so that the measurement point hits a part of the inspection object. Here, the measurement point is a pixel for determining whether or not it is a part of the inspection object in the image data, and is set for each thinning amount in the X-axis direction and the Y-axis direction.

  The inspection object determination unit 33 sets measurement points in the image data and determines whether or not each measurement point is a candidate point to be inspected. The inspection line setting unit 34 sets an inspection line based on the points determined by the inspection target determination unit 33 as candidate points to be inspected.

  The inspection area enlargement unit 35 enlarges the inspection area so as to include the inspection line set by the inspection line setting unit 34. The inspection region discriminating unit 36 determines whether the inspection region is the inspection range depending on whether the area value of the inspection region enlarged by the inspection region expanding unit 35 is within the range of the area value set by the threshold setting unit 31. It is determined whether or not there is.

  FIG. 4 is a flowchart for explaining the processing procedure of the image processing apparatus according to the embodiment of the present invention. First, the threshold value setting unit 31 sets a determination threshold value (RGB luminance value range) and an area value (feature value) range as threshold values (S1). Then, the search interval setting unit 32 sets thinning amounts (pitch amounts) Lx and Ly when searching for image data (S2).

  FIG. 5 is a diagram for explaining the inspection target and the pitch amounts Lx and Ly. FIG. 5A shows the inspection object 42 existing in the image data 41. It is assumed that the size and shape of the inspection object 42 are known in advance.

  FIG. 5B shows the pitch amounts Lx and Ly set by the search interval setting unit 32. As shown in FIG. 5B, the measurement points 43 are represented by black dots, and are set at intervals of Lx in the X-axis direction and at intervals of Ly in the Y-axis direction. The pitch amounts Lx and Ly are set according to the size or shape of the inspection object 42, and are set so that the measurement point 43 always hits a part of the inspection object.

  Next, the inspection object determination unit 33 determines whether each of the measurement points 43 is a candidate point for the inspection object 42 depending on whether or not the RGB luminance values of the measurement points 43 are within the threshold range. Determine (S3).

  If the RGB luminance value of the measurement point 43 is within the threshold value range (S3, Yes), the measurement point 43 is set as a candidate point for the inspection object 42 (S4). If the RGB luminance value of the measurement point 43 is not within the threshold range (S3, No), the measurement point 43 is excluded from the target candidates (S5).

  Next, the inspection line setting unit 34 sets the inspection line from the rising point to the falling point for each row based on the points determined as the candidate points of the inspection target 42 by the inspection target determination unit 33, and points at both ends thereof. The X coordinate and its range Zi are stored (S6). At this time, the inspection line is numbered.

  FIG. 6 is a diagram for explaining the inspection line set by the inspection line setting unit 34. In FIG. 6, a point 44 a is a rising point, and a point outside the inspection target when changing from a point outside the inspection target to a candidate point is set as a rising point 44 a. Further, the point 44b is a falling point, and when the point changes from a candidate point to a point that is not the inspection target, the point that is not the inspection target is set as the falling point 44b. Therefore, the hatched line 45 from the point 44a to the point 44b becomes the inspection line.

  Next, the inspection line setting unit 34 has the same attribute for all of the numbered inspection lines depending on whether or not the inspection line start point or end point is one pitch higher in the inspection line range Zi. Is determined (S7).

  If there is a start point or end point of an inspection line that is one pitch higher in the inspection line range Zi (Yes in S7), the same attribute (previous attribute) as the inspection line one pitch higher than that inspection line L1 is assigned (S8). If neither the start point nor the end point of the inspection line that is one pitch higher is present (S7, No), the new attribute L2 is assigned (S9). It is determined that inspection lines having the same attribute L1 are included in the same inspection object 42, and inspection lines having the new attribute L2 are included in different inspection objects 42.

  FIG. 7 is a diagram illustrating an inspection line in which attributes are set. As shown in FIG. 7, by the processing of steps S7 to S9, it is determined that the two left inspection lines are the same inspection target inspection lines, the attribute L1 is attached, and the right one inspection line is different inspection target. And the attribute L2 is attached.

  Next, the inspection area enlargement unit 35 determines whether or not the current inspection line is the top row in the inspection line group labeled with the same attribute by the inspection line setting unit 34 (S10). If the current inspection line is the top line (S10, Yes), the current inspection line range Zi is set to the horizontal direction of the inspection range, and inspection is performed for one pixel (1 pitch-2 pixels) from the current inspection line. The area is expanded upward (S11).

  If the current inspection line is not the top line (S10, No), the X coordinates of both ends of the current inspection line are compared with the X coordinates of both ends of the inspection line on one pitch, and the two inspection lines are compared. The inspection area is expanded laterally by the wider range Zj, and the inspection area is expanded upward by one pixel (1 pitch-2 pixels) from the current inspection line (S12).

  Then, the inspection area enlargement unit 35 determines whether or not the current inspection line is the lowest line (S13). If the current inspection line is not the lowest line (S13, No), the process returns to step S8 and the subsequent processing is repeated.

  If the current inspection line is the bottom line (S13, Yes), the current inspection line range Zi is set to the horizontal direction of the inspection range, and one pixel below (1 pitch-2 pixels) from the current inspection line. Only the inspection area is expanded downward (S14).

  Next, it is determined whether or not inspection ranges have been set for all inspection lines having attributes (S15). If there is an inspection line for which no inspection range is set (S15, No), the process returns to step S7 and the subsequent processing is repeated. If the inspection range is set for all the inspection lines (S15, Yes), the smallest rectangular area including all the inspection areas having the same attribute is set as the final inspection area (S16). This process is performed for each attribute.

  FIG. 8 is a diagram illustrating an example of image data in which an inspection area is set. An area A1 indicates an inspection area set in the uppermost inspection line with the same attribute L1. An area A2 indicates an inspection area set in the lower inspection line having the same attribute L1. Further, since this inspection line is also the lowermost inspection line, an inspection region B is set below the inspection line.

  A region C1 indicates a minimum rectangular region including the inspection regions A1, A2, and B, and this region C1 is a final inspection region. Note that different inspection areas C2 are set for inspection lines having different attributes L2.

  Next, the inspection area determination unit 36 determines whether or not the area value (feature amount) of the inspection area for each attribute is within the set range (S17). If the area value of the inspection area is not within the threshold value range set by the threshold value setting unit 31 (S17, No), the inspection area is set as an area outside the inspection area (S19).

  If the area value of the inspection region is within the threshold range (S17, Yes), the inspection region is set as the inspection range (S18). If there is an inspection area that has not been determined (S20, No), the process returns to step S17 and the subsequent processing is repeated. If all the inspection areas have been determined (S20, Yes), the process ends.

  With the above processing, even if the position and orientation of the inspection object 42 vary in the image data 41, the inspection region (inspection range) C1 can be automatically determined at high speed.

  In the above description, the RGB luminance value is used as the determination criterion. However, if the hue value is used as the determination criterion, the inspection range can be determined without being affected by the change in the brightness of the image data 41. Become. In this case, calculation for obtaining the hue value from the RGB value is required. However, by obtaining the hue value corresponding to the RGB value in advance and storing it in the LUT (Look Up Table), high speed can be obtained without performing the calculation. The hue value can be obtained.

  Further, the thinning amounts Lx and Ly can use the same value or different values depending on the size and shape of the inspection object 42.

  In addition, even if there is something in the image data 41 that is similar in color but different in size and shape, it can be determined whether or not it is the inspection object 42. Although the determination is made using the area value, the aspect ratio and diagonal length of the image to be inspected, or a combination thereof may be used.

  Further, there may be a case where a plurality of colors exist in the inspection object 42. FIG. 9 is a diagram for explaining a method for obtaining an inspection area when a plurality of colors exist in the inspection object 42. As shown in FIG. 9A, consider a case where a two-color image exists on the inspection object 42.

  First, one color having a low color frequency or no other color in the image data 41 is selected. In FIG. 9A, it is assumed that the color of the image shown in black is selected. Then, when the inspection area is searched using the selected color as the RGB feature quantity, an inspection area Ci as shown in FIG. 9B is obtained. And as shown in FIG.9 (c), what is necessary is just to expand the test | inspection area | region Ci further according to the magnitude | size and shape of the test object 42. FIG.

  As described above, according to the image processing apparatus of the present embodiment, the thinning-out amounts Lx and Ly when searching for image data are set, and data of only some measurement points 43 of the image data 41 is used. Since the inspection area Ci is obtained, the inspection area Ci can be obtained at high speed.

  Further, since the thinning amount of the measurement point 43 can be set to different values (Lx, Ly) in the horizontal direction and the vertical direction according to the size and shape of the inspection object 42, the inspection object in the image data 41 Even if the position and orientation of 42 vary, the inspection object 42 can be easily detected.

  Further, even when another region similar in color to the inspection target 42 exists, the inspection range including only the inspection target 42 is specified by comparing the characteristic amount of the inspection region with the characteristic amount of the inspection target 42. It became possible.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  DESCRIPTION OF SYMBOLS 1 Image processing apparatus, 2 cameras, 3 optical systems, 4 products, 5 mounting stage, 11 computer main body, 12 display apparatus, 13 FD drive, 14 FD, 15 keyboard, 16 mouse, 17 CD-ROM apparatus, 18 CD-ROM , 19 Network communication device, 20 CPU, 21 ROM, 22 RAM, 23 Hard disk, 31 Threshold setting unit, 32 Search interval setting unit, 33 Inspection object determination unit, 34 Inspection line setting unit, 35 Inspection region enlargement unit, 36 Inspection region Discriminating unit, 41 image data, 42 inspection object, 43 measurement points, 44a rising point, 44b falling point, 45 inspection line, C1, C2 inspection range.

Claims (14)

An image processing apparatus for detecting an inspection area including an inspection object from image data,
Inspection point determination means for setting a measurement point in the image data according to the thinning amount, and determining whether or not the inspection point candidate point according to the RGB value of each measurement point;
An inspection line setting unit that sets a line in which a measurement point determined as a candidate point by the inspection target determination unit exists as an inspection line;
Inspection area enlarging means for enlarging the inspection area according to the inspection line set by the inspection line setting means;
An image processing apparatus comprising: an inspection area determination unit that determines whether or not the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object.   The image processing apparatus according to claim 1, wherein the inspection target determination unit determines whether or not the inspection target candidate point is based on an RGB luminance value of the measurement point.   The image processing apparatus according to claim 1, wherein the inspection target determination unit determines whether or not the inspection target candidate point is based on a hue value of the measurement point.   The inspection line setting means uses a measurement point outside the inspection target when the measurement point changes from a non-inspection measurement point to an inspection target candidate point, and changes from the inspection target candidate point to a measurement point outside the inspection target. The image processing apparatus according to claim 1, wherein a point that is not subject to inspection at the time of the inspection is a falling point, and an area from the rising point to the falling point is an inspection line. The inspection line setting means, when the start point or the end point of the inspection line on one pitch exists within the range of the target inspection line, gives the same attribute as the inspection line on one pitch to the target inspection line,
The image processing apparatus according to claim 4, wherein when there is no start point or end point of the inspection line on one pitch, an attribute different from that on the inspection line on one pitch is given to the target inspection line to make a different inspection target. If the target inspection line is the uppermost line, the inspection area enlarging means sets the range of the target inspection line as the horizontal direction of the inspection area, and sets the inspection area of the target inspection line from one pixel to one pitch- Magnify up by 2 pixels,
If the target inspection line is not the top line, the target inspection line range is compared with the inspection line range one pitch above, and the inspection area is expanded in the horizontal direction by a wide range to inspect the target. Enlarging the inspection area of the line from 1 pixel up by 1 pitch-2 pixels,
If the target inspection line is the bottom line, the range of the target inspection line is set to the horizontal direction of the inspection area, and the inspection area of the target inspection line is expanded from one pixel below by one pitch-2 pixels. And
The image processing apparatus according to claim 1, wherein a minimum rectangular area including all of the enlarged inspection area is set as a final inspection area.   The image processing apparatus according to claim 1, wherein different values are set for the thinning-out amount in the horizontal direction and the vertical direction.   The inspection area determination unit determines whether the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object, using an area value of the inspection object as a feature amount. Image processing apparatus.   The inspection area determination unit determines whether or not the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object, using the aspect ratio of the inspection object as a feature amount. Image processing apparatus.   The inspection area determination unit determines whether or not the inspection area enlarged by the inspection area enlargement unit is an inspection area including the inspection object, using a diagonal length of the inspection object as a feature amount. The image processing apparatus described.   When there are a plurality of colors in the inspection object, after obtaining the inspection area according to a predetermined color, the inspection area enlarging means determines the inspection area according to the size or shape of the inspection object. The image processing apparatus according to claim 1, further expanding. An image processing method for causing a computer to detect an inspection area including an inspection object from image data,
Causing the computer to set measurement points in the image data according to the thinning amount, and determining whether or not it is a candidate point to be inspected according to the RGB value of each measurement point;
Setting a line in which a measurement point determined as the candidate point is present as an inspection line;
Expanding the inspection area according to the set inspection line;
And determining whether or not the enlarged inspection area is an inspection area including the inspection object. A computer program for causing a computer to detect an inspection area including an inspection object from image data,
Causing the computer to set measurement points in the image data according to the thinning amount, and determining whether or not it is a candidate point to be inspected according to the RGB value of each measurement point;
Setting a line in which a measurement point determined as the candidate point is present as an inspection line;
Expanding the inspection area according to the set inspection line;
And determining whether or not the enlarged inspection area is an inspection area including the inspection object. A recording medium recording a computer program for causing a computer to detect an inspection area including an inspection object from image data,
Causing the computer to set measurement points in the image data according to the thinning amount, and determining whether or not it is a candidate point to be inspected according to the RGB value of each measurement point;
Setting a line in which a measurement point determined as the candidate point is present as an inspection line;
Expanding the inspection area according to the set inspection line;
Discriminating whether or not the enlarged inspection region is an inspection region including the inspection object.

Images