JP2001043387A - Device and method for processing image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the position detection of an object in an inspection image from being dispersed by an operator by judging a detection procedure by trying the detection of the object according to the preset detection procedure. SOLUTION: A designation area 20a composed of line segments is designated and divided into two with a boundary detecting position 20b to considerably change luminance by visual confirmation as a boundary. When the maximum and minimum values of the luminance found for each divided area are not overlapped between areas, the position detection of a band-shaped object 10 in deep color is tried by the first position detection procedure. A spot, where the adjacent pixels of a binarized image are changed to white or black, is detected and a boundary 10a is detected. Afterwards, the position detection is performed by this procedure. When the maximum and minimum values are overlapped between areas, the position detection based on the second position detection procedure is tried. A spot, where an extreme value exists in the luminance differentiated value of pixels inside the designated area 20a, is detected and the boundary 10a is detected. Afterwards, the position detection is performed by this procedure.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for performing an inspection process / measurement process on an image obtained by capturing an image of an object to be inspected and determining whether or not the object 10 is good / bad and measuring. The present invention relates to the image processing method. More specifically, an image processing apparatus having a mechanism for instructing a region including an edge or a detection mark and a position of an edge or a mark to be detected with respect to an inspection target projected on a display unit such as a monitor, and an image thereof. Regarding the processing method.

[0002]

2. Description of the Related Art Conventionally, this type of image processing apparatus has a display unit and an arithmetic processing unit. The display unit is, for example, a monitor, and is provided with a display screen that displays an inspection image 20 in which the object 10 is imaged. The arithmetic processing unit performs position detection for detecting the display position of the target object 10 in the inspection image 20.

[0003] There are several procedures for detecting the position of the object 10 by the above-mentioned method, and the following two are representative of them. First, the first position detection procedure is as follows.
Based on a comparison between the luminance value on the display screen and a binarization level 100 set in advance based on the luminance distribution, the boundary 10 of the object 10 in the inspection image 20 along the predetermined direction is determined.
This is a method for detecting a. Specifically, the luminance value on the display screen is compared with a preset binarization level 100, and pixels having a luminance higher than the binarization level 100 are set to white, and a pixel having a luminance lower than the binarization level 100 is provided. By making the pixel that is black, two pixels consisting of white or black
Create a binarized image. Then, a portion where adjacent pixels change to white or black in the binarized image is referred to as an inspection image 2
The boundary 10a of the object 10 at 0 is set.

For example, as shown in FIG. 29 (a), when the object 10 is a dark band, when detecting a boundary 10a of the object 10 in a designated area 20a designated in advance,
When a position at which the luminance greatly changes is determined by visual recognition, the designated area 20a is divided into two along a predetermined direction at the determined position, and the luminance distribution in each of the two divided areas is determined. As shown in b), two luminance distributions each having a peak are obtained. In other words, the luminance of the pixel includes a low luminance side distribution of pixels having relatively low luminance based on the dark color of the object 10 and the object 1
Since it is not 0, it has a high-luminance side distribution of pixels that are light-colored and have relatively large luminance.
The average value of the medians of these two distributions is calculated as binarization level 100
By setting a position where adjacent pixels change to white or black when set as the boundary 10a of the target object 10 in the inspection image 20, each pixel is set to white or black, and a binary image is created. The point where adjacent pixels in the converted image change to white or black is identified as the object 10 in the inspection image 20.
Is the boundary 10a.

In the second position detection procedure, the inspection image 2 is determined along a predetermined direction based on the differential value of the luminance of the pixel.
This is a procedure for detecting the boundary 10a of the object 10 at 0. Specifically, based on the luminance value on the display screen, a grayscale image is created with a plurality of levels of colors from white to black. Then, at a place where adjacent pixels in this grayscale image change to white or black, the grayscale change locally increases, in other words, the absolute value of the differential value of the luminance of the pixel locally increases, The location where the absolute value of the differential value locally increases is defined as the boundary 10a of the object 10 in the inspection image 20.

For example, as shown in FIG. 30A, when the object 10 is a dark band, when detecting the boundary 10a of the object 10 in the designated area 20a designated in advance,
When a position at which the luminance greatly changes is visually recognized, and the obtained value is included in the graph, and a differential value of the luminance of the pixel is graphed along a predetermined direction, as shown in FIG. Become like Since it is possible to consider that the portion having the minimum value has changed from light color to dark color, the portion where the luminance differential value has the minimum value is identified as the inspection image 20.
Is the boundary 10a of the target object 10 in FIG.

[0007] The first and second position detection procedures described above include:
Each has advantages and disadvantages. That is, the luminance value on the display screen and the preset binarization level 100
Is easily compared with the operation of determining whether or not the differential value of the luminance value on the display screen is an extreme value, so that the luminance value on the display screen and the preset value of 2 are compared. In the first position detection procedure in which the calculation for comparing with the binarization level 100 is performed, the second position detection procedure in which the calculation for determining whether or not the differential value of the luminance of the pixel is an extreme value is performed. Also, the boundary 10a of the object 10 in the inspection image 20 can be detected in a short time.

However, in the first position detection procedure,
As shown in FIG. 29C, in the case where the luminance distributions of the two regions divided by the boundary 10a visually determined as the positions where the luminance greatly changes overlap, any region between the two distributions Since there is no luminance level which does not belong to the display screen, the luminance level is compared with the luminance value on the display screen.
The price level 100 cannot be set, and as a result,
The boundary 10a of the object 10 in the inspection image 20 cannot be detected.

On the other hand, in the second position detection procedure, as shown in FIG. 29 (c), the luminance distribution of each area divided by the boundary 10a visually determined as a position where the luminance changes greatly overlaps. Even in such a case, as described above, when the differential value of the luminance of the pixel is graphed, it has an extreme value. Therefore, by setting the location having the extreme value as the boundary 10a of the object 10, , The boundary 10a of the object 10 in the inspection image 20 can be detected.

[0010]

In the conventional image processing apparatus described above, the advantages of both the position detection procedures are utilized by selectively using the first and second position detection procedures on a case-by-case basis. Can be. That is, by using the first position detection procedure, it is possible to detect the boundary 10a of the target object 10 in the inspection image 20 in a short time, and by using the second position detection procedure, Even if it cannot be detected by the position detection procedure, the boundary 10a of the target object 10 in the inspection image 20 can be detected.

However, this is the first or second type.
The determination as to which of the position detection procedures should detect the boundary 10a of the object 10 in the inspection image 20 may vary depending on the experience of the operator who sets the position detection procedure. However, the first position detection procedure may be used, and conversely, the second position detection procedure may be used. There was a possibility that even the position detection could vary.

The present invention has been made in view of the above points, and an object of the present invention is to provide an image processing apparatus and an image processing method thereof in which the position detection of an object in an inspection image does not vary. It is in.

[0013]

According to a first aspect of the present invention, there is provided an image processing apparatus for detecting an object in a designated area on a display screen displaying an inspection image of the object. An image processing apparatus that attempts to detect the target object by a preset detection procedure, and detects the target object by the detection procedure when the target object can be detected by the detection procedure. Judgment, when the object can not be detected by the detection procedure,
The detection of the object is determined by a different detection procedure.

According to a second aspect of the present invention, there is provided an image processing apparatus for detecting a position of an object in a designated area on a display screen displaying an inspection image of the object.
If the target object can be detected based on a binarization level and the binarization of the luminance value with the binarization level as a boundary, the object is detected based on the binarization of the luminance value. When the detection is determined and the target cannot be detected based on the binarization of the luminance value, the detection of the target is performed based on the luminance differential value as a different detection procedure. .

According to a third aspect of the present invention, there is provided the image processing apparatus according to the second aspect.
In the image processing apparatus described in the above, when it is not possible to detect the target based on the luminance differential value, it is configured to determine to detect the target based on template matching in a predetermined scanning range as a different detection procedure. ing.

According to a fourth aspect of the present invention, there is provided an image processing apparatus according to the second aspect.
Alternatively, in the image processing apparatus according to claim 3, the binarization level is set based on a luminance distribution over the entire specified area.

According to a fifth aspect of the present invention, there is provided the image processing apparatus according to the second aspect.
5. The image processing apparatus according to claim 4, wherein, when a plurality of boundaries of the object have a predetermined order along a predetermined direction, the object is determined based on the predetermined order. Is detected to detect the object.

An image processing apparatus according to a sixth aspect of the present invention provides the image processing apparatus according to the second aspect.
5. The image processing apparatus according to claim 4, wherein when a plurality of boundaries of the object are present along a predetermined direction with a predetermined positional relationship between each other, the predetermined positional relationship is provided. The boundary of the target object is detected based on the target, and the target object is detected.

An image processing apparatus according to a seventh aspect is the second aspect.
7. The image processing apparatus according to claim 1, wherein the plurality of inspection images include a plurality of the inspection images, and the object is detected when the object cannot be detected in any of the plurality of inspection images. It is configured such that it is determined that the object is not possible and that the detection of the object is performed by a different detection procedure.

[0020] The image processing apparatus according to the eighth aspect is the second aspect.
8. The image processing apparatus according to claim 1, wherein when the number of the inspection images is plural, the binarization level is determined based on a luminance distribution of the designated area in the plurality of inspection images. It is configured to set the activation level.

An image processing apparatus according to a ninth aspect is the third aspect of the invention.
The image processing apparatus according to claim 8, wherein, when the plurality of inspection images are provided, when performing the template matching, the scanning range is set based on a variation between the plurality of inspection images. It has a variable configuration.

According to a tenth aspect of the present invention, in the image processing apparatus according to any one of the third to ninth aspects, when the inspection image is misaligned, the inspection is performed with a misalignment. When the position of the image is corrected, the position of the designated area or the scanning range is corrected.

An image processing method according to an eleventh aspect of the present invention is an image processing method for an image processing apparatus for detecting an object in a designated area on a display screen displaying an inspection image of the object. Attempt to detect the object by the detection procedure, and when the object can be detected by the detection procedure, determine that the object is to be detected by the detection procedure, and determine the object by the detection procedure. When it is impossible to detect the object, it is determined to detect the object by a different detection procedure.

According to a twelfth aspect of the present invention, there is provided an image processing method for an image processing apparatus for detecting a position of an object in a designated area on a display screen displaying an inspection image of the object. When the binarization level is set and the object can be detected based on the binarization of the luminance value with the binarization level as a boundary, the object is detected based on the binarization of the luminance value. If it is determined that the target object cannot be detected based on the binarization of the luminance value, a different detection procedure is used to determine the target object based on the luminance differential value.

According to a thirteenth aspect of the present invention, in the image processing method of the twelfth aspect, when the object cannot be detected based on the luminance differential value, a different scanning procedure is performed for a predetermined scanning range. The detection of the object is determined based on template matching.

According to a fourteenth aspect of the present invention, in the image processing method according to the twelfth or thirteenth aspect, the binarization level is set based on a luminance distribution over the entire specified area. I am trying to do it.

[0027]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to FIGS. This image processing device
This image processing device includes a TV camera 1, an A / D converter 2,
An image memory 3, an arithmetic processing unit 4, a D / A converter 5, a display unit 6, and an arithmetic memory 7 are provided.

The TV camera 1 captures an image of an object and outputs it to the A / D converter 2 as analog data. The A / D converter 2 converts the reference image into digital data, converts the inspection image obtained from the inspection target into digital data, and outputs the digital data to the image memory 3. The image memory 3 has an A / D
Both images of the digital data input from the converter 2 are stored.

The arithmetic processing section 4 reads out the image stored in the image memory 3 and executes an arithmetic processing including a position detection for detecting the object 10 in the specified area 20a specified in the inspection image 20. The D / A converter 5 includes an image memory 3
The image of the digital data input from is converted into analog data and output to the display unit 6. The display unit 6 displays an image of the analog data input from the D / A converter 5. The calculation memory 7 stores a procedure of a calculation process including position detection.

Next, the procedure of the above-described position detection will be described in detail with reference to FIGS. For example, FIG.
As shown in the figure, when the object 10 is in the form of a dark band,
First, a designated area 20a composed of line segments in a predetermined direction is designated from a light-colored area to a dark-colored area. Subsequently, a location where the luminance greatly changes by visual recognition is determined, and the location is set as a boundary detection position 20b. I do. Subsequently, the designated region 20a is divided into two along the boundary detection position 20b along a direction perpendicular to the predetermined direction, and the maximum value and the minimum value of the luminance of the pixel are respectively determined for each of the divided regions. Ask. When the calculated maximum value and minimum value do not overlap between the two areas as shown in FIG. 3B, the first position detection procedure set before the position detection work is performed. By
An attempt is made to detect the position of the object 10.

In the first position detection procedure, first, the average value of the median of the distribution for each region is set to the binarization level 100. Then, the binarization level 100 and the designated area 20
By comparing the absolute value of the luminance of the pixel in a with the pixel having a luminance higher than the binarization level 100 is set to white, and the pixel having a luminance lower than the binarization level 100 is set to black. Thus, a binarized image composed of binary values of white or black is created. Subsequently, a portion where adjacent pixels change to white or black in the binary image can be estimated as the boundary 10a.
The boundary 10a is detected by detecting a portion where the adjacent pixel changes to white or black in the binarized image. When the detected boundary 10a matches the boundary detection position 20b, the detection is sufficient. If so, it is determined that position detection can be performed by the first position detection procedure, and thereafter, position detection is performed by the first position detection procedure. Conversely, when the detected boundary 10a does not coincide with the boundary detection position 20b, it is determined that the detection is insufficient, and it is determined that the position detection by the first position detection procedure cannot be executed.

Further, between the above-mentioned maximum value and minimum value,
As shown in FIG. 3C, when the two areas overlap each other, it is determined that the position detection by the above-described first position detection procedure cannot be performed, and the second position detection is performed before the position detection operation. The position detection is performed by the position detection procedure of No. 2.

In this second position detection procedure, first, a luminance differential value is obtained by executing a differential processing of the luminance of the pixels in the designated area 20a. Then, it is determined whether or not an extreme value exists in the luminance differential value. Here, as shown in FIG. 4D, when an extreme value exists, the location where the extreme value exists is defined as the boundary 1 of the object 10 in the inspection image 20.
Since it can be estimated as 0a, the boundary 10a is detected by detecting the location where the extreme value exists, and when the detected boundary 10a matches the boundary detection position 20b, it is determined that the detection is sufficient. It is determined that position detection can be performed by the second position detection procedure, and thereafter, position detection is performed by the second position detection procedure. Conversely, when the detected boundary 10a does not coincide with the boundary detection position 20b, it is determined that the detection is insufficient, and it is determined that the position detection by the second position detection procedure cannot be performed, and the position detection procedure is automatically performed. Is set, the operator appropriately sets the position detection procedure.

In such a position detecting device, a binarization level 100 is set, and the binarization level 100 is
If the object 10 can be detected based on the binarization of the luminance value as a, it is determined that the object 10 is detected based on the binarization of the luminance value as a different detection procedure, and the luminance value 2 is determined.
If the object 10 cannot be detected based on the binarization, the judgment to detect the object 10 is made based on the luminance differential value. The procedure for detecting the target in the image 20 does not differ, and thus, the position detection of the target in the inspection image 20 does not vary.

Further, the fact that the position detection is possible but the position detection becomes impossible due to the operator not selecting an appropriate position detection procedure does not occur.

Next, a second embodiment of the present invention will be described below with reference to FIGS. The same parts as those in the first embodiment are denoted by the same reference numerals, and only the parts different from the first embodiment will be described. This embodiment is basically the same as the first embodiment, but when it is determined that the position detection cannot be performed by the second position detection procedure, the object 10 is determined based on the luminance differential value. When the detection is not possible, a determination to detect an object is made based on template matching of a scanning range 20c described later as a different detection procedure. This apparatus is provided with a reference image memory 8 for storing a reference image described later.

Next, the procedure of the above-described position detection will be described in detail with reference to FIGS. For example, FIG.
As shown in (a), when the object 10 has a dark band shape, a rectangular designated area 20a is designated from a light colored area to a dark colored area, and subsequently, the luminance is increased by visual recognition. The change position is determined, and the change position is set as the boundary detection position 20b. Subsequently, the designated region 20a is divided into two along the X-axis direction which is one of the predetermined directions, that is, in the direction perpendicular to the horizontal direction in FIG. The maximum value and the minimum value of the luminance of the pixel are obtained for each of the divided areas. It is determined whether or not the calculated maximum value and minimum value overlap each other between the two areas.

Subsequently, the designated area 20a is divided into two along the Y-axis direction which is another predetermined direction, that is, the direction perpendicular to the vertical direction in FIG. The maximum value and the minimum value of the luminance of the pixel are obtained for each of the divided areas, and it is determined whether the calculated maximum value and the minimum value overlap each other between the two areas.

Here, the area between the maximum value and the minimum value does not overlap between the two areas divided along the direction perpendicular to the X-axis direction, as shown in FIG. , FIG.
As shown in (1), when the two areas divided along the direction perpendicular to the Y-axis direction overlap each other, position detection is attempted by the first position detection procedure. Further, the maximum value and the minimum value overlap between the two regions divided along the X-axis direction, and between the two regions divided along the direction perpendicular to the Y-axis direction. Even if they do not overlap, position detection is attempted by the first position detection procedure.

On the other hand, the range between the maximum value and the minimum value is between the two regions divided along the direction perpendicular to the X-axis direction and between the two regions divided along the direction perpendicular to the Y-axis direction. However, if they overlap, it is determined that position detection by the first position detection procedure is impossible, and position detection is attempted by the second position detection procedure. In addition, when the maximum value and the minimum value do not overlap between both regions divided along the direction perpendicular to the X-axis direction or between both regions divided along the direction perpendicular to the Y-axis direction. Also, it is determined that the position detection by the first position detection procedure is impossible, and the position detection is attempted by the second position detection procedure.

If the detected boundary 10a does not coincide with the boundary detection position 20b after the first position detection procedure is tried, it is determined that the position detection by the first position detection procedure is impossible, and Position detection is attempted in accordance with the position detection procedure (2).

In the second position detection procedure, first, a luminance differential value is obtained by executing a differential processing of the luminance of each pixel along the X-axis direction and the Y-axis direction. Then, it is determined whether or not an extreme value exists in the luminance differential value.

Here, as shown in FIG.
An extreme value exists along the axial direction, and the location of the extreme value coincides with the boundary detection position 20b.
As shown in (b), when no extreme value exists along the Y-axis direction, it is determined that the position can be detected by the second position detecting procedure, and thereafter, the position is detected by the second position detecting procedure. Will be tried. Also, there is no extreme value along the X-axis direction, and there is an extreme value along the Y-axis direction,
Also when the extreme value coincides with the boundary detection position 20b, it is determined that the position can be detected by the second position detection procedure, and thereafter, the position detection is tried by the second position detection procedure.

On the other hand, when it is determined that the position can not be detected by the second position detection procedure, for example, as shown in FIG. The extreme value is present, the extreme value location coincides with the boundary detection position 20b, and the extreme value exists along the Y-axis direction, and the extreme value location is the boundary detection position. If the position coincides with 20b, position detection is attempted by a third position detection procedure set before the position detection operation.

In the third position detection procedure, first, the designated area 20a is registered as a reference image and stored in the reference image memory 8. Subsequently, the entire area of the inspection image 20 is set as a scanning range 20c, template matching for scanning the reference image including the designated area 20a is executed in the scanning range 20c, and matching with the scanned reference image in the scanning range 20c is performed. Since a point having a high degree can be estimated as the position of the object 10, the position is detected by detecting the point having a high degree of coincidence, and when the detected position matches the boundary detection position 20b, Assuming that the detection is sufficient, it is determined that position detection can be executed by the third position detection procedure, and thereafter, position detection is executed by the third position detection procedure. Conversely, when the detected position does not coincide with the boundary detection position 20b, it is determined that the detection is insufficient, and it is determined that the position detection by the third position detection procedure cannot be performed, and the position detection procedure is automatically executed. Instead of the setting, the operator appropriately sets the position detection procedure.

In this image processing apparatus, in addition to the effects of the first embodiment, when the object cannot be detected based on the luminance differential value, the template matching of the predetermined scanning range 20c is performed as a different detection procedure. The procedure for detecting the boundary of the object in the inspection image does not differ from worker to operator, and thus the position detection of the object in the inspection image varies. The effect of disappearing can be further achieved.

In the template matching, although the position can be detected almost surely, the first and second positions can be detected.
Since the calculation is complicated and takes a long time compared to the calculation in the position detection procedure, the position detection procedure including the template matching is performed in the third step after the first and second position detection procedures. By using the position detection procedure,
Time can be reduced.

Next, a third embodiment of the present invention will be described below with reference to FIGS. The same parts as those in the first embodiment are denoted by the same reference numerals, and only the parts different from the first embodiment will be described. In the first embodiment, the designated area 20a is divided into two parts by the boundary detection position 20b, and the maximum value and the minimum value of the luminance of the pixel are obtained for each of the divided areas. When the value and the minimum value do not overlap between the two regions, the average value of the median of the distribution for each region is set to the binarization level 100, whereas in the present embodiment, the designated region 20a Is set such that the binarization level 100 is set based on the luminance distribution over the entire area.

More specifically, as shown in FIG. 12, a luminance distribution over the entire designated area 20a is obtained, and when the luminance distributions do not overlap, a partition is made between the low luminance side distribution and the high luminance side distribution. The achievable luminance is set to the binarization level 100.

In this image processing apparatus, in addition to the effect of the first embodiment, when there are a plurality of boundaries 10a of the object 10, the brightness is increased as in the first embodiment. The position to be changed is visually determined, and the designated area 20a is divided into two along a predetermined direction at the determined position, and the average value of the median of the distribution of each of the divided areas is converted to a binarization level 100. In such a case, the luminance distribution is increased by including the boundary 10a in any of the divided regions, and as a result,
It becomes difficult to set the binarization level to 100,
In the first position detection procedure, it is determined that position detection cannot be performed, whereas the binarization level 100 is
Since the setting is made based on the luminance distribution over the entire area of a, it is not impossible to set the binarization level 100 by having a plurality of boundaries 10a of the object 10, In spite of the fact that position detection can be executed in the first position detection procedure, it is no longer determined that execution is impossible.

Next, a fourth embodiment of the present invention will be described below with reference to FIGS. It is to be noted that the same reference numerals are given to substantially the same portions as in the second embodiment, and only portions different from the second embodiment are described. In the second embodiment, the designated region 20a is divided into two along a predetermined direction at the boundary detection position 20b, and the maximum value and the minimum value of the luminance of the pixel are obtained for each of the divided regions. When the calculated maximum value and minimum value do not overlap between the two regions, the average value of the median of the distribution for each region is set to the binarization level 100. Is configured to be set based on the luminance distribution over the entire specified area 20a, as in the third embodiment.

In this image processing apparatus, in addition to the effects of the second embodiment, the binarization level 100 is set based on the luminance distribution over the entire designated area 20a, as in the third embodiment. The boundary 1 of the object 10
0a means that the binarization level is 10
It is not impossible to set 0, and it is not determined that execution is impossible even though position detection is possible in the first position detection procedure.

Next, a fifth embodiment of the present invention will be described below with reference to FIG. The same parts as those in the third embodiment are denoted by the same reference numerals, and only different points from the third embodiment will be described. The present embodiment is basically the same as the third embodiment, except that a boundary 1
When a plurality of 0a exist in a predetermined order, the boundary 10a of the object 10 can be detected based on the predetermined order.

For example, as shown in FIG. 18, the boundary 10a of the object 10 is
If one exists, the second boundary 1 along the predetermined direction
0a, the second boundary 10a is identified along the predetermined direction from the two detected boundaries 10a, and then the second boundary 10a is detected by the first or second position detection procedure. It is determined whether or not the boundary 10a matches the boundary detection position 20b.

In this image processing apparatus, in addition to the effects of the third embodiment, even when a plurality of boundaries 10a exist in a predetermined order along a predetermined direction, the boundaries are determined based on the predetermined order. It is possible to specify one boundary 10a for determining whether position detection is possible or impossible from a plurality of boundaries 10a, and to reliably determine whether position detection is possible or impossible. Can be

Next, a sixth embodiment of the present invention will be described below with reference to FIG. The same parts as those in the fourth embodiment are denoted by the same reference numerals, and only different points from the fourth embodiment will be described. This embodiment is basically the same as the fourth embodiment, but along the X direction which is one of the predetermined directions,
When a plurality of boundaries 10a of the object 10 exist in a predetermined order, the object 10 is determined based on the predetermined order.
Is configured to be able to detect the boundary 10a.

For example, as shown in FIG. 19, when three boundaries 10a of the object 10 exist in the designated area 20a along the predetermined direction, the boundary 10a must be the second boundary 10a along the predetermined direction. 3 is detected when
After specifying the second boundary 10a along the predetermined direction from the two boundaries 10a, the boundary 10a and the boundary detection position 20b detected by the first or second position detection procedure are used.
Is determined as to whether or not.

In this image processing apparatus, in addition to the effect of the third embodiment, even when a plurality of boundaries 10a exist in a predetermined order along a predetermined direction, the boundary 10a is determined based on the predetermined order. It is possible to specify one boundary 10a for determining whether position detection is possible or impossible from a plurality of boundaries 10a, and to reliably determine whether position detection is possible or impossible. Can be

Next, a seventh embodiment of the present invention will be described with reference to FIG.
This will be described below based on (a) and (b). The same parts as those in the fifth embodiment are denoted by the same reference numerals, and only different points from the fifth embodiment will be described. This embodiment is basically the same as the fifth embodiment. However, when a plurality of boundaries 10a are present along a predetermined direction with a predetermined positional relationship between each other, the predetermined Is configured so that the boundary 10a can be detected based on the positional relationship.

For example, as shown in FIG. 3A, the boundary 10a of the object 10 is
When there are four, a predetermined positional relationship between the boundaries 10a in a predetermined area 20c including the boundary detection positions 20b,
More specifically, when the predetermined positional relationship that the target boundary 10a exists between the boundaries 10a that match the mark 20d shown in FIG. Is different from the state shown in FIG. 4B, after specifying a boundary 10a having a predetermined positional relationship with another boundary 10a from among the detected four boundaries 10a, It is determined whether the boundary 10a detected by the first or second position detection procedure matches the boundary detection position 20b.

In this image processing apparatus, in addition to the effects of the third embodiment, even when a plurality of boundaries 10a exist in a predetermined direction and have a predetermined positional relationship between each other, the same effect can be obtained. Based on a predetermined positional relationship, a plurality of boundaries 10
From a, it is possible to identify one boundary 10a at which the determination as to whether the position detection is possible or not is possible, and it is possible to reliably determine whether the position detection is possible or not.

Even if all four boundaries 10a are not displayed on the display screen, it is determined whether position detection is possible or impossible from a plurality of boundaries 10a based on a predetermined positional relationship. One boundary 10a where the determination is made can be specified, and the determination as to whether the position can be detected or not can be made surely.

Next, an eighth embodiment of the present invention will be described with reference to FIG.
This will be described below based on (a) and (b). The same portions as those in the sixth embodiment are denoted by the same reference numerals, and only the portions different from the sixth embodiment will be described. This embodiment is basically the same as the sixth embodiment, but in this embodiment,
Further, along the X direction which is one of the predetermined directions, the boundary 1
When a plurality of 0a have a predetermined positional relationship between each other, the configuration is such that the boundary 10a can be detected based on the predetermined positional relationship.

For example, as shown in FIG. 9A, when there are four boundaries 10a of the object 10 in the designated area 20a along the X direction which is one of the predetermined directions, the boundary detection position Boundary 10 within predetermined area 20c including 20b
When a predetermined positional relationship between the four boundaries 10a is known, even if the state shown in FIG. 2B is different from the state shown in FIG. A boundary 10a having a predetermined positional relationship with the boundary 10a
Is determined, it is determined whether or not the boundary 10a detected by the first or second position detection procedure matches the boundary detection position 20b.

In this image processing apparatus, in addition to the effect of the sixth embodiment, even when a plurality of boundaries 10a exist in a predetermined direction with a predetermined positional relationship along a predetermined direction, this Based on a predetermined positional relationship, a plurality of boundaries 10
From a, it is possible to identify one boundary 10a at which the determination as to whether the position detection is possible or not is possible, and it is possible to reliably determine whether the position detection is possible or not.

Also, as in the seventh embodiment, even when all four boundaries 10a are not displayed on the display screen, it is possible to reliably determine whether position detection is possible or impossible. it can.

Next, a ninth embodiment of the present invention will be described below with reference to FIGS. The same parts as those in the eighth embodiment are denoted by the same reference numerals, and only the parts different from the eighth embodiment will be described. This embodiment is basically the same as the eighth embodiment, but when there are a plurality of inspection images 20, the boundary 10a of the object 10 cannot be detected in any of the plurality of inspection images 20. In such a case, it is determined that the position detection cannot be executed.

More specifically, as shown in FIGS.
Regardless of whether the position detection is attempted in any of the first to third position detection procedures, one of the plurality of inspection images 20
When the boundary 10a of the target object 10 cannot be detected, it is determined that the position detection cannot be executed.

In this image processing apparatus, in addition to the effect of the eighth embodiment, when it is impossible to detect the boundary 10a in any of the plurality of inspection images 20, it is determined that the position detection cannot be executed. Therefore, it is determined that the position detection can be executed only when the boundary 10a can be detected in all of the plurality of inspection images 20.
It is possible to determine whether the position detection can be executed or not in consideration of the variation between zeros, and it is possible to execute the position detection in a stable state.

This embodiment basically has the configuration of the eighth embodiment. Further, when the boundary 10a cannot be detected in any of the plurality of inspection images 20, the position detection cannot be executed. However, it basically has the configuration of the seventh embodiment, and furthermore, when the boundary 10a cannot be detected in any of the plurality of inspection images 20, the position detection cannot be executed. It is possible to determine whether position detection can be performed or not in consideration of the variation between the plurality of inspection images 20 even when the configuration is determined as follows.
There is an effect that position detection can be performed in a stable state.

Next, a tenth embodiment of the present invention will be described below with reference to FIG. The same parts as those in the ninth embodiment are denoted by the same reference numerals, and only the differences from the ninth embodiment will be described. In the eighth embodiment, the binarization level 100
Is set based on the luminance distribution over the entire specified area 20a specified in advance in one inspection image 20, whereas in the present embodiment, a plurality of inspection images 20
26, the luminance distribution over the entire designated area 20a designated in advance is obtained as shown in FIG. 26, and the configuration is set based on this luminance distribution.

In this image processing apparatus, in addition to the effects of the ninth embodiment, the binarization level 100 is set based on the luminance of the plurality of inspection images 20 to provide a plurality of images. The binarization level 100 can be set in consideration of the variation between the inspection images 20, and as a result,
Position detection can be performed in a stable state.

The present embodiment basically has the configuration of the ninth embodiment, and furthermore, based on the luminance distribution over the entire designated area 20 a designated in advance in the plurality of inspection images 20. Although the configuration is such that the binarization level 100 is set, the configuration basically has the configuration of the seventh embodiment, and furthermore, the luminance distribution over the entire designated area 20 a designated in advance in the plurality of inspection images 20. Even if the binarization level 100 is set based on the above, it is possible to determine whether the position detection can be executed or not in consideration of the variation between the plurality of inspection images 20, As a result, there is an effect that position detection can be performed in a stable state.

Next, an eleventh embodiment of the present invention will be described below with reference to FIGS. 27 and 28. The same parts as those in the tenth embodiment are denoted by the same reference numerals, and only the parts different from the tenth embodiment will be described. The present embodiment is basically the same as the tenth embodiment, except that the scanning range 20c in which the reference image including the designated area 20a is scanned is based on the variation in the display position among the plurality of inspection images 20. The configuration is variable.

More specifically, the variation of the display position along the X direction is σx, and the variation of the display position along the Y direction is σx.
y, the display position of the average is (X, Y), and the designated area 20a is determined from the boundary detection position 20b set by the operator.
As shown in FIG. 28, the distance to the end of (d1x, d
2x) and (d1y, d2y), the scanning range 2
0c, the scanning start point (sx, sy) and the scanning end point (e
x, ey) is obtained by the equations (1) to (4).

Sx = X−M × σx−d1x (1) sy = Y−M × σy−d1y (2) ex = X−M × σx + d2x (3) ey = Y−M × σy + d2y (4) Such image processing In the apparatus, in addition to the effects of the tenth embodiment, when the variation in the display position among the plurality of inspection images 20 is small, the designated area 20a can be reduced, and the scanning range can be reduced. Thus, the number of arithmetic processes required for position detection by template matching is reduced, and position detection can be performed at high speed.

Next, a twelfth embodiment of the present invention will be described below. The same parts as those in the eleventh embodiment are denoted by the same reference numerals, and only the parts different from the eleventh embodiment will be described.
This embodiment is basically the same as the eleventh embodiment, except that when the inspection image 20 is misaligned, a predetermined position correction is performed on the misaligned inspection image 20. The designated area 20a and the scanning range 20c are configured to have a predetermined position correction. More specifically, a predetermined position correction is performed on the scanning range 20c in a state where the position is corrected based on the variation of the display position between the plurality of inspection images 20.

In this image processing apparatus, in addition to the effects of the tenth embodiment, the inspection image 20
When the predetermined position is corrected, the specified position is also corrected in the designated range 20a and the scanning range 20c, so that the position can be accurately detected.

In the first to twelfth embodiments, when it is determined that the position detection by the first position detection procedure is impossible, the position detection by the second position detection procedure is tried. However, even in a configuration in which the third position detection procedure is tried without performing the second position detection procedure, the procedure for detecting the boundary 10a of the target object 10 in the inspection image 20 differs depending on the operator. In addition, the effect that the position detection of the object 10 in the inspection image 20 does not vary can be achieved.

The second embodiment, the fourth embodiment, and the sixth embodiment
In the embodiment and the eighth to twelfth embodiments, when it is determined that the position detection by the first position detection procedure is impossible, the position detection by the second position detection procedure is tried, and the second position detection is performed. When it is determined that the position detection by the detection procedure is impossible, the position detection by the third position detection procedure is tried. However, without performing the first position detection procedure, the second position detection is performed first. Even if a configuration is attempted in which the position detection is attempted by the third position detection procedure when the position detection by the second position detection procedure is determined to be impossible, the position detection is attempted by the third position detection procedure. The procedure for detecting the boundary 10a of the target object 10 in the inspection image 20 does not differ, and the effect that the position detection of the target object 10 in the inspection image 20 does not vary can be obtained.

[0081]

According to the first aspect of the present invention, the image processing apparatus tries to detect an object by a preset detection procedure, and if the object can be detected by the detection procedure, the object is detected by the detection procedure. In the case where the detection of the object is not detected by the detection procedure and the detection of the object is detected by a different detection procedure, unlike the conventional example, The procedure for detecting the target in the inspection image does not differ depending on the operator, and the position detection of the target in the inspection image does not vary.

According to the image processing apparatus of the present invention, a binarization level is set and a luminance value of 2 is set with the binarization level as a boundary.
If the object can be detected based on the binarization, it is determined to detect the object based on the binarization of the luminance value, and if the object cannot be detected based on the binarization of the luminance value Differs from the conventional example in that the procedure for detecting an object in an inspection image is different from that of the conventional example because the detection of the object is performed based on the luminance differential value as a different detection procedure. As a result, the detection of the position of the object in the inspection image does not vary.

According to a third aspect of the present invention, when an object cannot be detected based on the luminance differential value, a determination to detect the object based on template matching in a predetermined scanning range is performed as a different detection procedure. 3. The image processing apparatus according to claim 2, wherein the procedure for detecting the boundary of the object in the inspection image does not differ depending on the operator, and the position detection of the object in the inspection image does not vary. Can be further achieved.

The image processing apparatus according to the fourth aspect is the second aspect of the invention.
Or, in addition to the effect of the image processing apparatus according to claim 3, for example, when there are a plurality of boundaries of the object, a position where the luminance largely changes is visually determined and the determined position is determined. A specified area at the boundary
If the average value of the median of the distributions of the two divided areas is set as the binarization level,
In any one of the two regions, the distribution of luminance is increased by including the boundary. As a result, it is difficult to set the binarization level, and it is determined that position detection cannot be performed. On the other hand, since the binarization level is set based on the luminance distribution over the entire designated area, the binarization level can be set by having a plurality of object boundaries. It does not disappear, and it is no longer determined that the position detection is executable but the execution is not possible.

The image processing apparatus according to the fifth aspect is the second aspect of the invention.
In addition to the effects of the image processing apparatus according to any one of claims 4 to 5, when a plurality of boundaries exist in a predetermined order along a predetermined direction, a plurality of boundaries are determined based on the predetermined order. One of the boundaries from which the determination as to whether the position can be detected or not can be specified, and the determination as to whether or not the position can be detected can be ensured.

The image processing apparatus according to the sixth aspect is the second aspect of the invention.
In addition to the effects of the image processing apparatus according to any one of claims 4 to 4, when a plurality of boundaries exist along a predetermined direction with a predetermined positional relationship between each other, the predetermined positional relationship Based on the plurality of boundaries, it is possible to identify one boundary from which a determination as to whether position detection is possible or impossible is made, and to reliably determine whether position detection is possible or impossible. can do.

The image processing apparatus according to the seventh aspect is the second aspect of the invention.
In addition to the effects of the image processing device according to any one of claims 6 to 6, when a boundary cannot be detected in any of the plurality of inspection images, the object is regarded as undetectable, and the object is detected by a different detection procedure. Because it is determined to detect the object, considering the variation between the plurality of inspection images,
It is possible to determine whether the execution of the position detection is possible or impossible, and it is possible to execute the position detection in a stable state.

The image processing apparatus according to the eighth aspect is the second aspect.
In addition to the effects of the image processing apparatus according to any one of claims 6 to 6, the binarization level is set based on the luminance of the plurality of inspection images, so that the variation between the plurality of inspection images is obtained. , The binarization level can be set, and the position can be detected in a stable state.

The image processing apparatus according to the ninth aspect provides the image processing apparatus according to the third aspect.
In addition to the effects of the image processing apparatus according to any one of claims 8 to 12, when the variation between a plurality of inspection images is small, the scanning range can be reduced, and the position required for position detection by template matching can be reduced. The number of arithmetic processes is reduced, and position detection can be performed at high speed.

According to a tenth aspect of the present invention, in addition to the effects of the second to ninth aspects of the present invention, when the position correction is performed on the inspection image that has been displaced, Since the position is also corrected in the designated range or the scanning range, the position can be accurately detected.

According to the image processing method of the eleventh aspect,
Attempt to detect an object by a preset detection procedure,
If the object can be detected by the detection procedure, it is determined to detect the object by the detection procedure.If the object cannot be detected by the detection procedure, the object is detected by a different detection procedure. Therefore, unlike the conventional example, the procedure for detecting the object in the inspection image does not differ from the conventional example, and the position detection of the object in the inspection image differs. It doesn't stick.

According to the image processing method of the twelfth aspect,
When the binarization level is set and the target can be detected based on the binarization of the luminance value with the binarization level as a boundary, a determination to detect the target based on the binarization of the luminance value If the target object cannot be detected based on the binarization of the luminance value, a different detection procedure is used to determine the detection of the target object based on the luminance differential value. Unlike the above, the procedure for detecting the object in the inspection image does not differ depending on the operator, and the position detection of the object in the inspection image does not vary.

According to the image processing method of the thirteenth aspect,
If the object cannot be detected based on the luminance derivative,
Since the detection of the target is determined based on the template matching of the predetermined scanning range as a different detection procedure, the procedure for detecting the boundary of the target in the inspection image does not differ depending on the operator, and thus,
The effect of the image processing method according to the twelfth aspect that the position detection of the object in the inspection image does not vary can be further exhibited.

According to the image processing method of the fourteenth aspect,
In addition to the effects of the image processing method according to any one of claims 2 and 3, for example, when there are a plurality of boundaries of an object, a position where the luminance greatly changes is visually determined and the position is determined. If the designated region is divided into two along the predetermined direction at the boundary of the position and the average value of the median of the distribution of each of the divided regions is set as the binarization level, it is divided into two. In any of the regions, the distribution of luminance is increased by including the boundary, and as a result, 2
It is difficult to set the binarization level, and it is determined that position detection cannot be performed. On the other hand, the binarization level is set based on the luminance distribution over the entire specified area. However, the presence of a plurality of object boundaries does not prevent the binarization level from being set, and it is determined that the position detection can be performed but the position detection cannot be performed. It will not be.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating a procedure according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of the above.

FIG. 3 is an explanatory diagram showing a position detection procedure according to the above.

FIG. 4 is a configuration diagram of a second embodiment of the present invention.

FIG. 5 is a flowchart showing a procedure of the above.

FIG. 6 is a flowchart showing a procedure of the above.

FIG. 7 is a flowchart showing a procedure of the above.

FIG. 8 is a flowchart showing a procedure of the above.

FIG. 9 is an explanatory diagram showing a position detection procedure along the X-axis direction according to the above.

FIG. 10 is an explanatory diagram showing a position detection procedure along the Y-axis direction according to the above.

FIG. 11 is an explanatory diagram of a reference image used for template matching according to the above.

FIG. 12 is an explanatory diagram showing a binarization level according to the third embodiment of the present invention.

FIG. 13 is a flowchart showing a procedure of the above.

FIG. 14 is a flowchart illustrating a procedure according to a fourth embodiment of the present invention.

FIG. 15 is a flowchart showing a procedure of the above.

FIG. 16 is a flowchart showing a procedure of the above.

FIG. 17 is a flowchart showing a procedure of the above.

FIG. 18 is an explanatory diagram showing a position detection procedure according to a fifth embodiment of the present invention.

FIG. 19 is an explanatory diagram showing a position detection procedure according to a sixth embodiment of the present invention.

FIG. 20 is an explanatory diagram showing a position detection procedure according to the seventh embodiment of the present invention.

FIG. 21 is an explanatory diagram showing a position detection procedure according to the eighth embodiment of the present invention.

FIG. 22 is a flowchart illustrating a procedure according to a ninth embodiment of the present invention.

FIG. 23 is a flowchart showing a procedure of the above.

FIG. 24 is a flowchart showing a procedure of the above.

FIG. 25 is a flowchart showing a procedure of the above.

FIG. 26 is an explanatory diagram showing a binarization level according to the tenth embodiment of the present invention.

FIG. 27 is an explanatory diagram of a reference image used for template matching according to the eleventh embodiment of the present invention.

FIG. 28 is an explanatory diagram of a designated area that is a reference image of the above.

FIG. 29 is an explanatory diagram showing a first position detection procedure of a conventional example.

FIG. 30 is an explanatory diagram showing a second position detection procedure of the above.

[Explanation of symbols]

 4 Arithmetic processing unit 6a Display screen 10 Object 10a Boundary 20 Inspection image 20a Designated area 20c Scanning range 100 Binarization level

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Yoshihito Hashimoto, Inventor F-term (reference) 2F065 AA03 AA12 AA61 DD00 FF01 JJ03 JJ26 QQ03 QQ08 QQ13 QQ23 QQ24 QQ25 QQ32 QQ38 QQ39 RR07 SS13 UU05 5B057 AA02 BA02 BA24 CE12 DA03 DA07 DB02 DB09 DC16 DC23 DC33 DC38 5L096 AA06 BA03 CA02 CA24 DA02 EA43 FA06 FA32 FA37 FA69 FA76 GA02 GA19 JA03 JA11 JA16 9A001 DD15 HH23 HH25 LL05

Claims (14)

[Claims] An image processing apparatus for detecting an object in a designated area on a display screen displaying an inspection image of the object, wherein the image processing apparatus attempts to detect the object by a preset detection procedure. When the target can be detected by the detection procedure, a determination is made to detect the target by the detection procedure, and when the target cannot be detected by the detection procedure, a different detection procedure is used. An image processing apparatus, wherein a determination is made to detect the object. 2. An image processing apparatus for detecting a position of an object in a designated area on a display screen for displaying an inspection image obtained by imaging the object, comprising: setting a binarization level; If the object can be detected based on the binarization of the luminance value as a boundary, it is determined that the object is detected based on the binarization of the luminance value.
The image processing apparatus according to claim 1, wherein when the object cannot be detected based on the binarization, the detection of the object is determined based on a luminance differential value as a different detection procedure. 3. The method according to claim 1, wherein when the object cannot be detected based on the luminance differential value, the detection of the object is determined based on template matching in a predetermined scanning range as a different detection procedure. Item 3. The image processing device according to Item 2. 4. The binarization level is set based on a luminance distribution over the entire specified area.
An image processing apparatus according to claim 3. 5. When a plurality of boundaries of the object are present along a predetermined direction in a predetermined order, the boundaries of the objects are detected based on the predetermined order, and the boundary of the object is detected. The image processing apparatus according to claim 2, wherein the image processing device detects the image. 6. When a plurality of boundaries of the object are present along a predetermined direction with a predetermined positional relationship therebetween, the boundary of the object is detected based on the predetermined positional relationship. The image processing apparatus according to claim 2, wherein the object is detected. 7. When there are a plurality of inspection images, and when the object cannot be detected in any of the plurality of inspection images, the object is regarded as undetectable, and the inspection object is determined by a different detection procedure. The image processing apparatus according to claim 2, wherein a determination is made to detect an object. 8. When there are a plurality of inspection images,
8. The position detecting device according to claim 2, wherein the binarization level is set based on a luminance distribution of the designated area in the plurality of inspection images. 9. The method according to claim 3, wherein, when the template matching is performed in a case where there are a plurality of inspection images, the scan range is variable based on a variation between the plurality of inspection images. The image processing device according to any one of the above. 10. The method according to claim 10, wherein when the inspection image is misaligned, the position of the designated area or the scanning range is corrected when the position of the misaligned inspection image is corrected. The image processing apparatus according to claim 2. 11. An image processing method of an image processing apparatus for detecting an object in a designated area on a display screen displaying an inspection image of the object, wherein the object is detected by a predetermined detection procedure. Attempt to detect, if the object can be detected by the detection procedure, determine the detection of the object by the detection procedure, if the object is not detectable by the detection procedure, An image processing method for an image processing apparatus, wherein a determination for detecting the target object is made according to different detection procedures. 12. An image processing method for an image processing apparatus for detecting a position of an object in a designated area on a display screen displaying an inspection image of the object, wherein a binarization level is set, and When the object can be detected based on the binarization of the luminance value having the boundary of the binarization level, it is determined that the object is detected based on the binarization of the luminance value. When the target object cannot be detected based on the binarization of the image processing apparatus, the detection of the target object is determined based on a luminance differential value as a different detection procedure. Processing method. 13. The method according to claim 1, wherein when the object cannot be detected based on the luminance differential value, the detection of the object is determined based on template matching in a predetermined scanning range as a different detection procedure. Item 13. The image processing method of the image processing device according to Item 12. 14. The image processing method according to claim 12, wherein the binarization level is set based on a luminance distribution over the entire specified area.