JP2016217872A - Inspection device, inspection method, program, and storage media - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an inspection device capable of inspecting the defect of a pattern more easily.SOLUTION: An inspection device detects the defect of a pattern by binarizing the photographic image of the pattern of a test object to extract the pattern and comparing a feature quantity calculated relative to the pattern with the feature quantity obtained in advance relative to a standard pattern. The inspection device calculates an area and a barycentric coordinate for each separated pattern section from each other included in the pattern as a feature quantity to compare these with a value obtained in advance regarding the standard pattern.SELECTED DRAWING: Figure 6

Description

  The present invention relates to an inspection apparatus for inspecting a pattern such as a wiring.

  As a method for inspecting a defect in a pattern such as a wiring formed on a sheet or the like, a method of performing image processing on a photographed image obtained by photographing a pattern is known (for example, see Patent Document 1).

  In this method, an image of a pattern having no defect is registered in advance as a reference pattern, and the defect is often detected by comparing the pattern on the captured image with the reference pattern by difference processing or the like. In order to improve inspection accuracy, known image processing such as expansion and contraction may be performed in advance on the pattern on the captured image.

Japanese Patent Publication No. 5-013443

  In the method for comparing with the reference pattern as described above, since the image data of the reference pattern needs to be registered in the inspection apparatus in advance, there is a disadvantage that the amount of memory used becomes large and the registration work becomes complicated. there were.

  Furthermore, in recent years, high-definition patterns of such wiring and the like have been advanced in various fields, and it has become necessary to compare higher resolution images. Along with this, the time required for the comparison processing due to the difference from the reference pattern and the like is also increasing, and a method capable of more easily inspecting pattern defects has been demanded.

  An object of this invention is to provide the inspection apparatus etc. which can test | inspect the defect of a pattern more simply.

  According to a first aspect of the present invention for solving the above-described problem, an extraction unit that extracts a pattern from a captured image of a pattern to be inspected, a calculation unit that calculates a feature amount of the extracted pattern, and the calculated feature amount And a detecting means for detecting a defect of the pattern to be inspected by comparing a feature amount obtained in advance with respect to a reference pattern.

  In the present invention, a captured image of a pattern to be inspected is acquired, a pattern is extracted by binarizing the acquired captured image to obtain a feature amount, and the pattern defect is compared with the feature amount of the reference pattern. To detect. Since the pattern to be inspected and the reference pattern are compared not by an image but by a feature amount, the information to be registered in advance is only a numerical value, which is easier to register than conventional image data and the memory usage amount The amount of time required for the comparison process can be shortened.

The feature amount of the pattern is preferably a value calculated for each pattern portion included in the pattern. The feature amount preferably includes a barycentric coordinate of the pattern portion. It is also desirable to include the area of the pattern portion.
By using the value for each pattern part separated from each other as the feature amount, the accuracy of defect detection is improved. Further, by using the barycentric coordinates and area of each pattern portion, it is possible to suitably detect a defect.

It is preferable that the detection unit determines the type of defect by comparing the area of the pattern portion as the feature amount.
Based on the comparison of the area of the pattern part, such as determining that a disconnection has occurred if the area of the pattern part is small compared to the reference pattern, or conversely if the area is large, it is determined that a short circuit has occurred. It is possible to determine the type of defect.

  According to a second aspect of the invention, the inspection device extracts the pattern from the captured image of the pattern to be inspected, calculates the feature amount of the extracted pattern, the calculated feature amount, and a reference pattern And a step of detecting a defect of a pattern to be inspected by comparing with a feature amount obtained in advance for the inspection method.

  According to a third aspect of the present invention, the computer uses an extraction unit that extracts the pattern from a captured image of the pattern to be inspected, a calculation unit that calculates a feature amount of the extracted pattern, and the calculated feature amount. This is a program for functioning as an inspection apparatus having detection means for detecting a defect in a pattern to be inspected by comparing a feature amount obtained in advance with respect to a pattern.

  According to a fourth aspect of the present invention, the computer uses an extraction unit that extracts the pattern from a captured image of a pattern to be inspected, a calculation unit that calculates a feature amount of the extracted pattern, and the calculated feature amount. This is a recording medium on which a program for functioning as an inspection apparatus having detection means for detecting a defect of a pattern to be inspected by comparing a feature amount obtained in advance with respect to a pattern is recorded.

  According to the present invention, it is possible to provide an inspection apparatus or the like that can inspect pattern defects more easily.

Diagram showing inspection system 1 The figure which shows the hardware constitutions of the inspection apparatus 3. Example of pattern 13 Example of feature data 40 Example of captured image 100 of pattern 13 Flow chart showing the inspection method Example of feature data 50 Example of feature data 50 Diagram explaining the inspection method Another example of feature

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

(1. Inspection system 1)
FIG. 1 is a diagram showing an inspection system 1 according to an embodiment of the present invention. As shown in FIG. 1, the inspection system 1 includes an inspection device 3, an imaging device 5, and the like. The inspection system 1 shoots a web-like sheet 10 (inspection target) while being conveyed as indicated by an arrow by a conveyance device (not shown) such as a conveyance roller, and a pattern formed on the sheet 10 from the photographed image. The inspection is performed.

  The inspection device 3 controls the photographing device 5 and the like to photograph the sheet 10 and inspects the pattern by detecting a pattern defect from the photographed image using a technique such as image processing.

  FIG. 2 is a diagram illustrating a hardware configuration of the inspection apparatus 3. As shown in FIG. 2, the inspection apparatus 3 can be realized by, for example, a computer configured by connecting a control unit 31, a storage unit 32, an input unit 33, a display unit 34, a communication unit 35, and the like through a bus 36. However, the present invention is not limited to this, and various configurations can be taken as appropriate.

  The control unit 31 includes a CPU, a ROM, a RAM, and the like. The CPU calls and executes a program related to processing of the inspection apparatus 3 stored in a storage medium 32, a recording medium such as a ROM, in a work memory area on the RAM. The ROM is a non-volatile memory and permanently holds programs such as a computer boot program and BIOS, and data. The RAM is a volatile memory, and temporarily stores programs and data loaded from the storage unit 32, the ROM, and the like, and includes a work area used by the control unit 31 for performing various processes.

  The storage unit 32 is, for example, a hard disk drive, and stores a program executed by the control unit 31, data necessary for program execution, an OS, and the like. These programs and data are read by the control unit 31 as necessary, transferred to the RAM, and executed. In the present embodiment, the feature amount data of the reference pattern to be compared with the pattern is stored in the storage unit 32 in advance for use in pattern inspection. The reference pattern is a pattern having no defect. The feature amount will be described later.

The input unit 33 inputs data and includes an input device such as a keyboard, a pointing device such as a mouse, and a numeric keypad.
The display unit 34 includes a display device such as a liquid crystal panel and a logic circuit (such as a video adapter) for realizing a display function in cooperation with the display device.
The communication unit 35 is a communication interface that mediates communication via a network or the like, and performs communication with other devices.
The bus 36 is a path that mediates transmission / reception of control signals, data signals, and the like between the respective units.

  Returning to the description of FIG. The photographing device 5 performs photographing of the sheet 10 under the control of the inspection device 3. The imaging device 5 is not particularly limited, and for example, a line camera or an area camera is used.

  The inspection system 1 is not limited to this, and any inspection system 1 may be used as long as the sheet 10 can be imaged and the pattern can be inspected from the captured image. For example, the inspection system 1 is also provided with illumination for imaging (not shown). As an imaging method, the sheet 10 may be imaged by the transmitted light method or the sheet 10 may be imaged by the reflected light method. . In any case, the pattern formed on the sheet 10 only needs to appear in the captured image. Further, the sheet 10 is not limited to the web shape, and may be a sheet shape. The conveying means and conveying method of the sheet 10 are not particularly limited.

(2. Pattern 13 of sheet 10)
FIG. 3 is a diagram showing a part of the pattern 13 of the sheet 10 and shows the pattern 13 of the inspection range in the present embodiment. The pattern 13 is, for example, a wiring pattern made of metal or the like, but is not limited thereto, and the material and the forming method are not particularly limited. The material of the sheet 10 itself is not particularly limited.

  In this example, the pattern 13 is formed in a stripe shape in which linear pattern portions 13-1 to 13-6 that are separated from each other and are arranged in parallel at a predetermined pitch. However, the shape of the pattern 13 is not limited to this.

(3. Reference pattern features)
In the present embodiment, the defect-free pattern 13 as shown in FIG. 3 is used as a reference pattern (reference pattern), and the feature amount is registered in the inspection apparatus 3 in advance and recorded in the storage unit 32. This feature amount can be obtained, for example, by performing processing described in S1 and S2 of FIG.

  FIG. 4A is an example of the feature amount data 40 of the reference pattern. In this embodiment, the area and barycentric coordinates of each pattern portion 13-1 to 13-6 are registered. The barycentric coordinates represent the positions of the pattern portions 13-1 to 13-6, and coordinates (X coordinate, Y coordinate) along two orthogonal axes on the plane are registered. In this embodiment, the X coordinate corresponds to the horizontal direction of the captured image, and the Y coordinate corresponds to the vertical direction of the captured image. FIG. 4B shows the center of gravity 15 of each pattern portion 13-1 to 13-6 in a circle.

(4. Pattern 13 inspection method)
In the present embodiment, while the sheet 10 is being conveyed, the imaging device 5 is controlled by the inspection device 3 and the pattern 13 in the inspection range is imaged. The inspection device 3 acquires a captured image and inspects the pattern 13 based on the captured image.

  FIGS. 5A and 5B are examples of the captured image 100 obtained by capturing the pattern 13. FIG. 5A is an example in which the pattern 13 has no defect, as in the reference pattern, and FIG. 5B is an example in which the pattern 13 has a defect. In FIG. 5B, a broken portion 131 and a short-circuit portion 132 are generated in the pattern 13 as defects.

  FIG. 6 is a flowchart showing a method for inspecting the pattern 13 by the inspection apparatus 3. Each step in the figure is executed by the control unit 31 of the inspection apparatus 3.

  In the present embodiment, first, the inspection apparatus 3 performs a binarization process with a predetermined threshold on the captured image 100, and extracts the pattern 13 from the captured image 100 (S1).

  In the present embodiment, by performing binarization processing, the pattern portions 13-1 to 13-6 separated from each other are extracted as dark regions with respect to the other portions (the ground portion of the sheet 10). And However, the pattern portions 13-1 to 13-6 may be extracted as bright regions. In the present embodiment, numbers (“13-1”, “13-2”,...) Are assigned to the extracted pattern portions in ascending order of the X-coordinate of the center of gravity (in the order of the Y-coordinate being the same if the X-coordinate is the same). It shall be labeled.

  The inspection device 3 calculates the area and barycentric coordinates of the pattern portions 13-1 to 13-6 as the feature amount of the pattern 13 (S2).

  FIG. 7A is an example of the feature amount data 50 obtained by calculating the feature amount when the pattern 13 has no defect (see FIG. 5A). FIG. 7B shows the center of gravity 15 of each pattern portion 13-1 to 13-6 as a circle.

  On the other hand, FIG. 8A is an example of the feature amount data 50 obtained by calculating the feature amount when the pattern 13 has a defect (see FIG. 5B). FIG. 8B shows the center of gravity 15 of each pattern portion 13-1 to 13-6 as a circle.

  The inspection device 3 compares the feature amount (feature amount data 50) of the pattern 13 obtained in S2 with the feature amount of the reference pattern registered in advance (feature amount data 40, see FIG. 4A). A defect is detected by determining the presence or absence of a defect depending on whether or not both correspond (S3). The result is recorded in the storage unit 32 or displayed on the display unit 34, for example.

  In S3, for example, as shown in FIG. 9, the area of each pattern portion 13-1 to 13-6 is compared with the X coordinate and Y coordinate of the center of gravity 15 for both the feature amount data 40 and 50, and even one of them is compared. If the values are different, it can be determined that there is no correspondence, that is, there is a defect. Note that even if there is no defect, the values do not necessarily match exactly (the difference between the values is not 0), and it is considered that there is an error. Can be considered to be in agreement if they are within the error range.

  However, the determination method is not limited to the above. For example, for both the feature data 40 and 50, the area of each pattern portion 13-1 to 13-6 and the X coordinate and Y coordinate of the center of gravity 15 are compared and the difference is accumulated, and the accumulated value is predetermined. If the value is exceeded, it can be determined that both do not correspond (there is a defect).

  As can be seen by comparing FIG. 4A and FIG. 7A, when there is no defect in the pattern 13, the feature value data 50 obtained in S2 shows good agreement with the feature value data 40 of the reference pattern. . In such a case, it is determined that there is no defect.

  On the other hand, as can be seen by comparing FIG. 4A and FIG. 8A, when the pattern 13 has a defect such as a broken portion 131 or a shorted portion 132, the feature amount data 50 obtained in S2 and the reference pattern When comparing the feature amount data 40, it can be seen that some values are different and do not correspond. For example, in the example of FIG. 8A, the disconnection portion 131 generates pattern portions 13-3 and 13-4 having a smaller area than the reference pattern feature amount data 40, and the short-circuit portion 132 causes the reference pattern. A pattern portion 13-6 having a larger area than the feature amount data 40 is generated. Further, the coordinates of the center of gravity 15 of these pattern portions 13-3, 13-4, and 13-6 are also changed. In such a case, it is determined that there is a defect.

  In S3, in addition to detecting a defect, for example, the type of defect such as disconnection or short circuit can be determined from the comparison of the areas as described above, and recorded in the storage unit 32 together with the detection result of the defect or displayed on the display unit 34 can be displayed.

  Further, in S3, it is possible to determine the presence or absence of a defect by comparing only the area or only the barycentric coordinates and detect the defect. As for the barycentric coordinates, only the X coordinate (or Y coordinate) can be compared, and the defect can be detected by the difference in pitch between the pattern portions. In addition, first, only the barycentric coordinates are compared to determine the presence / absence of a defect, and then, for the pattern 13 determined to have a defect, the type of defect is determined by comparing the area of the pattern portion as described above. You can also.

  As described above, according to the present embodiment, the captured image 100 of the pattern 13 to be inspected is acquired, and the acquired captured image 100 is binarized to extract the pattern 13 to obtain the feature amount. The defect of the pattern 13 is detected by comparing with the stored feature amount of the reference pattern. Since the pattern 13 to be inspected and the reference pattern are compared not by the image but by the feature amount, the information to be registered in advance is only a numerical value, which is easier to register than the conventional image data and the memory usage amount And the time required for the comparison process can be shortened.

  Moreover, the accuracy of defect detection improves by calculating | requiring the value for every pattern part 13-1 to 13-6 isolate | separated as a feature-value. Further, by using the barycentric coordinates and areas of the pattern portions 13-1 to 13-6, it is possible to suitably detect defects.

  In addition, if the area of the pattern portion is small compared to the reference pattern, it is determined that a disconnection has occurred, and conversely, if the area is large, it is determined that a short circuit has occurred. It is also possible to determine the type of defect based on the above.

  However, the present invention is not limited to this. For example, the value to be obtained for each pattern portion is not limited to the area and the barycentric coordinates, and it is sufficient that the feature of the pattern portion can be expressed. As an example, the number of vertices 133 indicated by circles in FIG. 10A is calculated as a value representing the shape of the pattern portion, or the length of the pattern portion in the horizontal direction and the length in the vertical direction of the image as the slenderness ratio. You can also find the ratio. Alternatively, it is possible to calculate the size of the rectangular area 134 circumscribing the pattern portion as indicated by the dotted line in FIG.

  In addition, it is also possible to calculate the number of pattern portions as the feature amount of the pattern 13 as indicated by the encircled numbers in FIG. 10C and compare it with the reference pattern. In addition, it is also possible to compare S3 after excluding the pattern portion extracted in S1 under specific conditions, such as excluding small and round pattern portions as dust. Of course, by performing known image processing on the photographed image 100, a portion unnecessary for inspection such as dust can be erased in advance.

  The preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these are naturally within the technical scope of the present invention. Understood.

DESCRIPTION OF SYMBOLS 1; Inspection system 3; Inspection apparatus 5; Imaging | photography apparatus 10; Sheet 13; Pattern 13-1 to 13-6; Pattern part 15: Gravity center 40, 50; Feature-value data 131; ; Rectangular area

Claims (8)

Extracting means for extracting the pattern from a captured image of the pattern to be inspected;
Calculating means for calculating the feature amount of the extracted pattern;
Detecting means for detecting a defect in a pattern to be inspected by comparing the calculated feature amount with a feature amount obtained in advance for a reference pattern;
An inspection apparatus comprising:   The inspection apparatus according to claim 1, wherein the feature amount of the pattern is a value calculated for each pattern portion included in the pattern.   The inspection apparatus according to claim 2, wherein the feature amount includes a barycentric coordinate of the pattern portion.   The inspection apparatus according to claim 2, wherein the feature amount includes an area of the pattern portion.   5. The inspection apparatus according to claim 4, wherein the detection unit determines the type of defect by comparing the area of the pattern portion as the feature amount. Inspection equipment
Extracting the pattern from a captured image of the pattern to be inspected;
Calculating a feature amount of the extracted pattern;
Detecting the defect of the pattern to be inspected by comparing the calculated feature amount with a feature amount obtained in advance for a reference pattern;
The inspection method characterized by performing. Computer
Extracting means for extracting the pattern from a captured image of the pattern to be inspected;
Calculating means for calculating the feature amount of the extracted pattern;
Detecting means for detecting a defect in a pattern to be inspected by comparing the calculated feature amount with a feature amount obtained in advance for a reference pattern;
A program for functioning as an inspection apparatus having Computer
Extracting means for extracting the pattern from a captured image of the pattern to be inspected;
Calculating means for calculating the feature amount of the extracted pattern;
Detecting means for detecting a defect in a pattern to be inspected by comparing the calculated feature amount with a feature amount obtained in advance for a reference pattern;
The recording medium which recorded the program for functioning as an inspection apparatus which has.

Images