JPH0718811B2 - Defect inspection method - Google Patents

Description

Detailed Description of the Invention [Industrial applications]

The present invention relates to a defect inspection method for inspecting a defective portion on the surface of an inspection object such as a chip, a crack, a stain, and a foreign object by image processing.

[Prior art]

Conventionally, a spatial region including an inspection object is imaged by an image input means such as a television camera, the density of each pixel of an original image is binarized by appropriately using a threshold value, and a defective portion is identified based on this binary image. A method for detecting is known (see, for example, Japanese Patent Laid-Open No. Sho 62-62
-88946 gazette). That is, in the binary image, if there is a portion where the value is inverted inside the area that can be regarded as the inspection object, it is regarded as a defective portion.

[Problems to be Solved by the Invention]

In the above conventional method, since the defective portion is determined based on the binary image, the density of the defective portion and the density around the defective portion are included in the same region with respect to the threshold value (for example, If the density is also above the threshold value), there is a problem that the defective portion cannot be identified. The present invention is intended to solve the above problems, and detects an edge flag point that can be regarded as a point on the edge of a defective portion, and changes the density in the vicinity of the edge flag point or each pixel on the contour line of the defective portion. An object of the present invention is to provide a defect inspection method in which the accuracy of detecting a defective portion is improved by detecting the density change in the vicinity of.

[Means for Solving the Problems]

In the present invention, in order to achieve the above object, at least one search line is set inside the contour line of the inspection object for the edge image, and an edge flag point that can be regarded as an edge of a defective portion is detected on the search line. The defect portion is identified based on the change in density of the original image near the edge flag point. Further, when at least one inspection area is set inside the contour line of the inspection object for the edge image, and an edge flag point that can be regarded as the edge of the defect portion is detected by searching while scanning the entire inspection area, The contour line of the defective portion may be traced using the edge flag point as a starting point, and the defective portion may be identified based on the density change of the original image near each pixel on the contour line.

[Action]

According to the first method, the edge flag point that can be regarded as a point on the edge of the defective portion is detected, and the density change in the original image in the vicinity of the edge flag point is detected. Therefore, the contrast with the surroundings is relatively small. Even a defective portion can be detected, and the detection accuracy of the defective portion is improved. Also in the second method, since the density change in the vicinity of the contour line of the defective portion is detected, it is possible to detect the defective portion having a relatively small contrast.

First Embodiment In the present invention, as shown in FIG. 1, an inspection target is imaged by an image input device 1 such as a television camera, and the density of each pixel is converted into a digital signal in an analog-digital conversion unit 2. After that, by performing the following preprocessing in the preprocessing unit 3, a differential absolute value image, a differential direction value image, and an edge image are obtained in addition to the original image obtained by the analog-digital conversion unit 2. That is, the original image P ₀ obtained by imaging the spatial region including the inspection object is a grayscale image, and as shown in FIG. 2A, the inspection object O, the defect X ₁ , the foreign material X _2, etc. The image includes. The process of extracting an edge such as a contour line of the inspection object O from the grayscale image is based on the idea that “the edge corresponds to a portion where the density change is large”. Therefore, the edge is extracted by differentiating the density. The differentiation process is performed by dividing the original image P ₀ into 3 × 3 pixel local parallel windows W, as shown in FIG. That is, a pixel E of interest and eight pixels (8 neighborhoods) A to D and F to I around the pixel E form a local parallel window W, and the density of the pixels A to I in the local parallel window W is The vertical density change ΔV and the horizontal density change ΔH are obtained by the following formula, and ΔV = (A + B + C) − (G + H + I) ΔH = (A + D + G) − (C + F + I) Further, the differential absolute value | e | _E value and the differential The direction value ∠e _E is calculated by the following equation. | e | _E = (ΔV ² + ΔH ² ) ^1/2 However, A to I indicate the densities of the corresponding pixels. As is clear from the above equation, the differential absolute value | e | _E represents the maximum change rate of the density in the neighborhood area of the pixel of interest in the original image, and the differential direction value ∠e _E is the maximum density in the neighborhood area. It represents the direction orthogonal to the direction of change. By performing the above calculation for all the pixels of the original image P _O , the contour of the inspection object O, the defect X ₁ , the foreign substance X _{2 and the} like where the density change is large and the direction of the change is large. Can be extracted. An image in which the density of each pixel is expressed by a differential absolute value | e | _E is called a differential absolute value image, and an image in which the density is expressed by a differential direction value ∠e _E is called a differential direction value image. Next, thinning processing is performed. The thinning process is performed by paying attention to the fact that the larger the differential absolute value, the larger the density change. That is, the differential absolute value of each pixel is compared with the differential absolute value of the surrounding pixels, and those having a larger value than the surrounding pixels are connected to extract an edge having a width of one pixel. As shown in FIG. 4, considering the differential absolute value image P _{1 in} which the position of each pixel is represented by XY coordinates and the differential absolute value is taken on the Z axis, the thinning process determines the ridgeline on this curved surface. Equivalent to that. By the processing so far, regardless of the magnitude of the differential absolute value,
All edges are extracted. Since the ridge line obtained at this stage includes unnecessary small peaks due to noise or the like, the threshold value SL is appropriately set as shown in FIG.
A noise component is removed by adopting only a value equal to or higher than the threshold value SL. The image obtained by this processing is likely to be a discontinuous line when the contrast of the original image is insufficient or when there is a lot of noise. Therefore, edge extension processing is performed. In the edge extension process, starting from the end point of the discontinuous line, the pixel of interest is compared with the surrounding pixels, and the line is extended in the direction in which the evaluation function f (e _j ) expressed by Continue until you hit the end of the line. Here, e ₀ is differential data of the central pixel (corresponding to E of the local parallel window W), e _j is differential data of adjacent pixels (corresponding to 8 neighborhoods of the local parallel window W), and j = 1,2, ..., 8. As a result of the above processing, as shown in FIG. 2B, an edge image P ₄ that traces a portion of the original image where the density change is large is obtained. In the edge image P _4, it can be regarded as what each line represents a test object O, defect X _1, outline l ₀ to l ₂ of foreign objects such as X _2. By the above pre-processing, the original image, differential absolute value image,
Four types of images, a differential direction code image and an edge image, are obtained, and each image is stored in the original image memory 4, the differential absolute value image memory 5, the differential direction value image memory 6, and the edge image memory 7. In the following description, it is assumed that the pixel position of each image is represented by XY coordinates, and the pixel densities in each image are f ₁ (x, y), f ₂ (x, y), and f ₃ (x , y), f
₄ (x, y). The original image is a grayscale image, and the density is usually represented by 8 bits, so the density a (= f ₁ (x, y)) at each pixel
Is 0 ≦ a ≦ 255. Further, the density of the differential absolute value image (that is, the differential absolute value) b (= f ₂ (x, y)) is represented by 6 bits, for example, 0 ≦ b ≦ 63, and the density of the differential direction value image (that is, , Differential direction value) c (= f ₃ (x,
y)) is expressed in 16 directions, for example, and 0 ≦ c ≦ 15. For the edge image, only the presence or absence of lines,
Pixels forming lines are represented as "1" and other pixels are represented as "0". That is, the range of f ₄ (x, y) is {0,1}. In the following description, the term density represents white density, and the larger the density value, the brighter. Next, in the search processing unit 6, as shown in FIG. 7, the search line S is set inside the contour line l ₀ for the edge image P ₄ . As shown in FIGS. 8A to 8D, the search line S can be arbitrarily set as long as it has a closed loop shape. That is, the search line S in which the start point and the end point match is set. After setting the search line S, the search is started, and when a point where f ₄ (x, y) = 1 is found on the search line S, this point is set as an edge flag point. Next, in the stick mask setting unit 9, as shown in FIG. 9, a plurality of pixels before and after the edge flag point F on the search line S are centered on the pixel of the edge flag point F found first from the start point. (Pixels indicated by diagonal lines in FIG. 9) are extracted and the stick mask M is set. Stick mask M
Is a region having a width of 1 and a length of 2n + 1, where n is a natural number and is appropriately set. Now, the coordinates of the edge flag point F are (x _F , y _F ), and the coordinates of each pixel in the stick mask M are R _n (x _{F + n} , y _{F + n} ), ..., R ₁ (x _{F + 1} , Y _{F + 1} ), F
(X _F , y _F ), S ₁ (x _F-1 , y _F-1 ), ..., S _n (x _Fn ,
y _Fn ), the density calculation unit 10 calculates the sum of the differences in the pixel densities on the original image P ₀ at symmetrical positions in the stick mask M with the edge flag point F sandwiched between them. Find D. D = Σ | f ₁ (x _{F + i} , y _{F + i} ) −f ₁ (x _Fi , y _Fi ) | However, i = 1,2, ..., n, or i = m, m + 1, ... …, N
(M ≠ 1). The determination unit 11 compares the total sum D obtained as described above with a predetermined threshold value SL1. If D> SL1, it means that there is a defective portion such as a chip or a foreign substance on the search line S. to decide. That is, if the edge flag point F is a point on the outline of the defective portion, the stick mask M is
Since the defect portion and the portion which is not the defect portion are straddled, the contrast in the vicinity of the edge flag point F can be detected by obtaining the density difference of the pixels existing symmetrically with respect to the edge flag point F. That is why. Further, since the density difference is detected for a plurality of pixels, the contrast is emphasized and the detection accuracy can be improved. D ≦ SL
When it is 1, the search on the search line S is continued and the presence or absence of the edge flag point is detected. Thus, the search line S
When the edge flag point is not detected at the stage when the end point is reached, it is determined that there is no defective portion. The flow of the above processing is shown in FIG.

Second Embodiment In the above-described embodiment, the search line S is formed in a closed loop shape, but in the present embodiment, as shown in FIG.
The start point and the end point of the search line S are set to different positions. Here, the search line S can have any shape such as a straight line, a curved line, and a polygonal line as long as it is inside the contour line of the inspection object, and the direction can also be set arbitrarily. The processing flow is as shown in FIG. 10. After setting the search line S, the edge flag point F is detected in the same manner as in the first embodiment, and the edge flag point F is used in the first embodiment.
The stick mask M is set by a method similar to the above, and the total sum D of the density differences of pixels existing at symmetrical positions on the stick mask M is obtained. If this total D satisfies D> SL1 with respect to a predetermined threshold value SL1, then this edge flag point F is made a defect candidate point. However, in the present embodiment, even if the edge flag point F satisfies D> SL1, it is not immediately determined to be a defective portion, but after the total number K of such defect candidate points on the search line S is obtained, The total number K of defect candidate points is compared with a predetermined threshold value SL2, and when K> SL2, it is determined that a defective portion exists. That is, the condition for determining whether or not a defective portion exists is stricter than that in the first embodiment, and the detection accuracy can be further improved.

Third Embodiment In this embodiment, as shown in FIG. 13, a plurality of search lines S _{1 to} S _n having the same shape are set. In this case, as shown in FIG. 14, if there is defective portion X _1, since the defect portion X ₁ is found in succession for a plurality of search line, defect X ₁
The search line that has found is a defect candidate line, a plurality of adjacent search lines are defect candidate lines, and the total number L of defect candidate lines is compared with a predetermined threshold value SL3 so that L> SL3. Sometimes, it is determined that the defective portion X ₁ exists. Here, the defect X ₁ on one search line
Whether or not is present is determined in Example 1 or Example 2.
Either method may be used. The processing flow of this embodiment is as shown in FIG. 12, and is indicated by the broken line by detecting the defective portion X ₁ by the method of the first embodiment except for the portion indicated by the broken line in FIG. If a portion is included, it will be detected by the method of Example 2.

Fourth Embodiment In each of the above embodiments, the search line is set inside the contour line of the inspection object, but in the present embodiment, as shown in FIG. 16, the inside of the contour line l ₀ of the inspection object is set. Standard concentration setting area DM
₁ and inspection area DM ₂ are set. The flow of processing is 15th
As shown in the figure, it will be described below. In the reference density setting area DM ₁ , the reference density SD, which is the average value of the densities of all the pixels in the reference density setting area DM ₁ , is obtained from the original image. Next, for the inspection area DM ₂ , as shown by the arrow in FIG. 17, raster scanning is performed to obtain edge flag points where f ₄ (x, y) = 1. The coordinates of the edge flag point F ₀ found first by raster scanning are set to (x ₀ , y ₀ ), and this edge flag point F ₀ is used as the starting point. As shown in FIG. 18, the edge flag point F ₀ (x ₀ , y ₀ ) each edge flag point F _{n that} continues in sequence
The following processing is performed for. In the following description, the coordinates of each edge flag point F _k are respectively (x _k , y _k ). However, n = 1, 2, ... Now, it is assumed that processing is performed on the edge flag point F ₁ . For the edge flag point F ₁ (x _i , y _i ) of interest, the differential direction value b ₁ is obtained from the differential direction value image, and the direction orthogonal to the direction indicated by the differential direction value b ₁ (normal direction) Set the stick mask M to. Stick mask M
As shown in FIG. 19, the shape is set to be symmetrical about the edge flag point F ₁ , but the pixels in the stick mask M do not have to be in contact with each other. Here, pixels in the stick mask M are n pixels away from the edge flag point F ₁ (x ₁ , y ₁ ) and m pixels away. The coordinates of each point of the stick mask M are (x _{i + m} , y _{i + m} ),
As (x _{i + n} , y _{i + n} ), (x _in , y _in ), (x _im , y _im ), the density value of each point is obtained on the original image, and f ₁ (x _{i + m} , y _{i + m) = d 1 f} 1 (x i + n, y i + n) = d 2 f 1 (x in, y in) = d 3 f 1 (x im, and y _im) = d _4. Next, the defective portion X ₁ existing in the inspection area DM ₂ is
A defect that is brighter than the non-defective portion (in the present embodiment,
In order to determine whether the defect is a white defect portion) or a dark defect (hereinafter, referred to as a black defect portion in this embodiment) as compared with a non-defect portion, the density in the stick mask M is increased. The values d _{1 to} d ₄ are compared with the reference concentration SD. put it here,
d _1, is satisfied _{d 2, d 3, d 4} > condition that SD (all d ₁ to d ₄ is greater than SD), in subsequent processing, white threshold
WSL is used, and if the condition is not satisfied, black threshold BSL is used. The white threshold value WSL and the black threshold value BSL may be appropriately set according to the contrast of the defective portion X ₁ . That is, when the condition of using the white threshold value WSL is satisfied, the density difference of the symmetric points in the stick mask M is | d ₁ −d ₄ |> WSL, | d ₂ −d ₃ |> WSL. It is checked whether or not each of the conditions is satisfied, and if either one of them is satisfied, the density difference satisfying the condition is set as the white addition value. If the condition of using the black threshold value BSL is satisfied, whether or not the conditions of | d ₁ −d ₄ |> BSL and | d ₂ −d ₃ |> BSL are similarly satisfied. If any of the conditions is satisfied, the density difference satisfying the condition is set as the black addition value. After the white added value and the black added value are obtained as described above, the presence or absence of the edge flag point that is adjacent to the edge flag point F _{K in the} clockwise direction is checked. If there are adjacent edge flag points,
The above process is performed, and the presence or absence of adjacent edge flag points is checked to perform tracking. While the tracking is being performed, the total sum of the white added value and the black added value is obtained. This tracking ends when one of the following stop conditions is met. When the number of pixels to be traced exceeds the preset maximum number of traces Tm When an adjacent edge flag point goes out of the inspection area DM _{2 When} there is no adjacent edge flag point When the tracking ends, white addition is performed The total sum TW of the values and the total sum TB of the black addition values are respectively compared with the preset threshold value W ₁ for the white defect portion and the threshold value B ₁ for the black defect portion, and TW> W ₁ is satisfied. When there is a white defect in the inspection area DM ₂ , and when TB> B ₁ is satisfied, it is determined that a black defect exists in the inspection area DM ₂ . Further, when none of the conditions is satisfied, it is determined that there is no defective portion in the inspection area DM ₂ . The threshold values WSL, BSL, W ₁ and B ₁ and the maximum number of traces Tm may be appropriately set depending on the size of the defect portion and the contrast.

Fifth Embodiment In this embodiment, as shown in FIG. 21, only the inspection area DM is set within the contour line of the inspection symmetry object, and raster scanning is performed in this inspection area DM to determine the starting point of tracking. The edge flag point is detected. The processing procedure is as shown in FIG. 20, and similarly to the fourth embodiment, the stick mask M is set and the densities d _{1 to} d ₄ of each pixel are obtained in the stick mask M. Further, the density difference at the symmetrical point on the stick mask M is obtained and compared with a predetermined threshold value SL ₁ . That is, the edge flag point is traced while checking whether or not at least one of the condition of | d ₁ −d ₄ |> SL ₁ | d ₂ −d ₃ |> SL ₁ is satisfied. Here, the number of edge flag points satisfying the condition is counted during the tracking. Further, the differential absolute value is obtained for the traced edge flag points based on the differential absolute value image, and the differential absolute values are sequentially added. The stop condition for tracking the edge flag point is the same as that in the fourth embodiment. At the time when the tracking is completed, the sum K of the edge flag points satisfying the above conditions and the sum L of the differential absolute values of the edge flag points are obtained. Therefore, the preset threshold values SL ₂ and SL ₃ are obtained. In comparison, if at least one of the conditions of K> SL ₂ L> SL ₃ is satisfied, it is determined that there is a defective portion in the inspection region DM ₂ . When no defective portion is found, the above-mentioned processing is performed on the portion of the inspection area DM which is not raster-scanned, and when the edge flag point is not found, or the conditions for the threshold values SL ₂ and SL ₃ are both satisfied. If not, it is determined that there is no defective portion in the inspection area DM.

Sixth Embodiment In this embodiment, similarly to the fourth embodiment, the reference density setting area DM ₁ and the inspection area DM ₂ are set, and the average value of the densities in the reference density setting area DM ₁ is set as the reference density SD. The following processing is as shown in FIG. 22, and a defect portion having a density value brighter than the reference density SD (hereinafter, referred to as a white defect portion in this embodiment),
And a defect portion having a dark density value (in the present embodiment,
In order to detect a black defect portion), a white threshold value WSL for detecting a white defect portion and a black threshold value BSL for detecting a black defect portion are detected.
And. That is, as shown in FIG. 23, the light and dark offset values LT and DK are set with respect to the reference density SD, and the white threshold value WSL and the black threshold value BSL are set according to the following equations. WSL = SD + LT BSL = SD-DK In this way, after setting the white threshold value WSL and the black threshold value BSL, the inspection area DM ₂ is raster-scanned in the same manner as in Example 4, and tracking is started. The edge flag point which becomes a point is detected. When the edge flag point F _M is detected, as shown in FIG. 24, the density measurement points P _R and P _{L are} located symmetrically across the edge flag point in the direction orthogonal to the differential direction value of the position. To set. Here, the density measurement points P _R and P _L are set at positions separated by n pixels from the edge flag point F _M. As is clear from the definition, the differential direction value has the property that the left side of the direction indicated by the differential direction value is dark and the right side is bright on the original image, so the density measurement point P _R on the right side of the direction indicated by the differential direction value Density u in the original image of
The density v in the original image at the density measurement point P _L on the left side is compared with L and the black threshold value B SL is compared. Here, if the condition of u> WSL is satisfied, the edge flag point is set as a white defect candidate point. If v <BSL is satisfied, the edge flag point is set as a black defect candidate point. This process is repeated while tracking the edge flag points connected in series, and the total number of each of the edge flag points regarded as the white defect candidate points and the edge flag points regarded as the black defect candidate points is obtained. The conditions for stopping the tracking of edge flag points are the same as in the fourth embodiment. Let SW be the total number of white defect candidate points and SB be the total number of black defect candidate points at the time when the tracking is completed, and compare them with preset thresholds SL ₃ and SL ₄ . In other words, satisfy at least one condition of _{SW> SL 3 SB> SL 4} , it determines that the defect portion is present in the inspection area DM _2, when neither satisfied, in the examination region DM _2, still raster Raster scanning is performed on the non-scanned portion and the defective portion is inspected by the same procedure. At the time when the raster scan is completed for the entire area of the inspection area DM ₂ , if there is no edge flag point or both the conditions regarding the threshold values SL ₃ and SL ₄ are not satisfied, the inspection area DM ₂ has a defect. Judge that the department does not exist.

【The invention's effect】

As described above, the present invention sets at least one search line inside the contour line of the inspection object for the edge image, and detects the edge flag point that can be regarded as the edge of the defect portion on the search line. The defect portion is identified based on the density change of the original image in the vicinity of, and the edge flag point that can be regarded as a point on the edge of the defect portion is detected, and the original image in the vicinity of the edge flag point is detected. Since the change in density is detected, even a defective portion having a relatively small contrast with the surroundings can be detected, and there is an advantage that the detection accuracy of the defective portion is improved. Further, when at least one inspection area is set inside the contour line of the inspection object for the edge image, and an edge flag point that can be regarded as the edge of the defect portion is detected by searching while scanning the entire inspection area, If the contour line of the defect portion is traced with the edge flag point as the starting point and the defect portion is identified based on the density change of the original image near each pixel on the contour line, the contour line of the defect portion Since the density change in the vicinity of is detected, it is possible to detect even a defective portion having a relatively small contrast.

[Brief description of drawings]

FIG. 1 is a block diagram of a processing circuit corresponding to the method of the first embodiment of the present invention, FIGS. 2 (a) and 2 (b) are explanatory views showing examples of an original image and an edge image in the present invention, and FIG. Is an explanatory view showing a local parallel window in the same as above, FIG. 4 is an explanatory view showing a differential absolute value image in the same as above, and FIG. 5 is a noise removing process at the stage of obtaining an edge image from the differential absolute value image in the same as above. Explanatory diagram, FIG. 6 is an operational explanatory diagram showing the processing procedure of the above, FIG. 7 is an operational explanatory diagram showing the setting state of the search line for the edge image of the above, and FIG. 8 (a) to (d) are the same as above. 9 is an explanatory view showing the form of a search line, FIG. 9 is an operation explanatory view showing the concept of edge flag points in the same as above, FIG. 10 is an operation explanatory view showing a processing procedure in the second embodiment of the present invention, and FIG. a) to (d) are the same as above. Explanatory diagram showing the form of the search line,
FIG. 12 is an operation explanatory view showing a processing procedure in Embodiment 3 of the present invention, FIG. 13 is an explanatory view showing a form of a search line in the same as above, FIG. 14 is an operation explanatory view in the same as above, and FIG. FIG. 16 is an operation explanatory diagram showing a processing procedure in the fourth embodiment, FIG. 16 is an operation explanatory diagram of the same as above, FIG. 17 is an operation explanatory diagram of raster scanning in the same as above, and FIG. 18 is an operation explanation showing a stick mask setting state in the same as above. FIG. 19 is an operation explanatory view showing an example of the stick mask in the above, FIG. 20 is an operation explanatory view showing a processing procedure in the fifth embodiment of the present invention, and FIG. 21 is an operation explanatory view showing an inspection area in the same. FIG. 22 is an operation explanatory view showing a processing procedure in the sixth embodiment of the present invention, FIG. 23 is an operation explanatory view showing a relationship between a white threshold value and a black threshold value and an offset value in the same as above, FIG. Is an operation explanatory diagram showing the concentration measurement points in the same as above. That. l _{0 to} l ₂ ... contour line, O ... inspection object, X ₁ ... defect, X ₂ ... foreign matter.

Claims (8)

[Claims] 1. An image input means captures an image of a spatial region including an inspection target, and then converts the image into an edge image including an outline of the inspection target based on the density of each pixel of the original image obtained by the image input means. Then, in the defect inspection method for inspecting a defective portion such as a chip, a crack or a stain existing inside the contour line of the inspection object based on the original image and the edge image, the contour line of the inspection object based on the edge image When at least one search line is set on the inside of the line, and an edge flag point that can be regarded as an edge of the defect portion is detected on the search line, the defect portion is identified based on the density change in the vicinity of the edge flag point in the original image. A defect inspection method characterized by the above. 2. The edge image is obtained by a thinning process based on a differential absolute value image in which each pixel is associated with a differential absolute value representing the maximum change rate of the density value in the neighborhood area of each pixel of the original image. , When an edge flag point is detected on the search line set in the contour line of the inspection object based on the edge image, a stick mask composed of a plurality of pixels centered on the edge flag point is set on the search line, 2. The defect inspection method according to claim 1, wherein the sum of the density differences of the pixels at symmetrical positions in the stick mask is calculated based on the original image, and if the sum exceeds a predetermined value, it is judged that there is a defective portion. . 3. The edge image is obtained by a thinning process based on a differential absolute value image in which each pixel is associated with a differential absolute value representing the maximum change rate of the density value in the neighborhood area of each pixel of the original image. , When an edge flag point is detected on the search line set in the contour line of the inspection object based on the edge image, a stick mask composed of a plurality of pixels centered on the edge flag point is set on the search line, Based on the original image, the sum of the density differences of the pixels at symmetrical positions in the stick mask is obtained, and when the sum exceeds a predetermined value, the edge flag point is set as a defect candidate point, and the total number of defect candidate points on the search line is The defect inspection method according to claim 1, wherein it is determined that a defective portion is present when a predetermined value is exceeded. 4. A plurality of parallel search lines are set within the contour line of the inspection object for the edge image, and
When it is determined that there is a defective portion for each search line by the defect inspection method according to claim 2 or 3, the search line is set as a defect candidate line, and a plurality of adjacent search lines are continuously defined as defect candidate lines. In addition, the defect inspection method is characterized in that when the total number of defect candidate lines exceeds a predetermined value, it is judged that there is a defective portion. 5. The image inputting means captures an image of a spatial region containing the inspection object, and then converts it into an edge image containing the contour line of the inspection object based on the density of each pixel of the original image obtained by the image inputting means. Then, in the defect inspection method for inspecting a defective portion such as a chip, a crack, and a stain existing inside the contour line of the inspection object based on the original image and the edge image, in the edge image, the inside of the contour line of the inspection object If an edge flag point that can be regarded as an edge of a defective portion is detected by setting at least one inspection area in the scanning area and searching the entire scanning area while scanning, the contour line of the defective portion is determined using the edge flag point as a starting point. A defect inspection method characterized in that a defect portion is identified based on a density change in the vicinity of each pixel on a contour line while tracking. 6. The edge image is obtained by a thinning process based on a differential absolute value image in which each pixel is associated with a differential absolute value representing the maximum change rate of the density value in the neighborhood area of each pixel of the original image. , The reference density setting area and the inspection area are set inside the contour line of the inspection object, and the average value of the densities of the pixels in the reference density setting area of the original image is calculated as the reference density, and the scanning is performed in the inspection area. When an edge flag point is detected in the pixel, the differential direction value image in which the differential direction value that reflects the direction of the maximum change in the density value in the vicinity region of each pixel of the original image is associated with each pixel A stick mask consisting of multiple pixels centered on the edge flag point is set in the direction of maximum change, and the densities of all the pixels in the stick mask for the original image exceed the above-mentioned reference densities, and If the density difference of the pixels of the pixel exceeds a predetermined white threshold value, the density difference is set as the white addition value, and the density of at least one pixel in the stick mask of the original image is equal to or lower than the reference density, and the stick mask When the density difference of pixels at symmetrical positions exceeds the specified black threshold value, the density difference is set as the black addition value, all adjacent edge flag points are tracked, and the white addition value and black addition are performed for each edge flag point. 6. The defect according to claim 5, wherein after detecting the value, if at least one of the white added value and the black added value exceeds a predetermined value set corresponding to each of them, it is judged that a defective portion exists. Inspection method. 7. The edge image is obtained by a thinning process based on a differential absolute value image in which each pixel is associated with a differential absolute value representing the maximum change rate of the density value in the neighborhood area of each pixel of the original image. , If the inspection area is set inside the contour line of the inspection object and an edge flag point is detected during scanning in the inspection area, the direction of the maximum change of the density value in the vicinity area of each pixel of the original image is reflected. Based on the differential direction value image in which the differential direction value is associated with each pixel, a stick mask consisting of a plurality of pixels centered on the edge flag point is set in the direction of the maximum change in the density value, and the original image When the density difference of pixels at symmetrical positions exceeds a predetermined value, the edge flag point is set as a defect candidate point, all the adjacent edge flag points are traced, and then the total number of defect candidate points and the differentiation of all defect candidate points are detected. Absolute value Defect inspection method according to claim 5, characterized in that it is determined that at least one of the sum is defective unit exceeds a predetermined value set in correspondence with each occurrence. 8. The edge image is obtained by a thinning process based on a differential absolute value image in which each pixel is associated with a differential absolute value representing the maximum change rate of the density value in the neighborhood area of each pixel of the original image. , A reference density setting area and an inspection area are set inside the contour line of the inspection object, and a white threshold value larger than the average value of the density of the pixels in the reference density setting area obtained from the original image and the above average value. When a small black threshold value is set and an edge flag point is detected during scanning in the inspection area, the differential direction value that reflects the direction of maximum change in density value in the vicinity area of each pixel of the original image is set to each pixel. Based on the differential direction value image corresponding to, the pair of density measurement points apart from the edge flag point are set at symmetrical positions around the edge flag point in the direction of the maximum change of the density value, and the density of the original image is changed. The density of the density measurement point with the larger value If the value exceeds the white threshold value, it is regarded as a white defect candidate point, and if the density value of the density measurement point with the smaller density value is smaller than the black threshold value, it is regarded as a black defect candidate point, and all adjacent edge flag points are traced. After detecting the white defect candidate points and the black defect candidate points for each edge flag point, at least one of the total number of white defect candidate points and the total number of black defect candidate points is set to a predetermined value corresponding to each. The defect inspection method according to claim 5, wherein it is determined that a defective portion is present when the number exceeds the limit.