JP2008241298A - Flaw detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a flaw detection method capable of detecting a flaw accurately by properly connecting the flaw, even in a special shape, for example, like a line and a space. <P>SOLUTION: The flaw detection method includes an adjacent flaw circumscribed rectangle setting process for setting the circumscribed rectangle of an adjacent detection flaw Dn as an adjacent flaw circumscribed rectangle 52, an adjacent flaw circumscribed rectangle for magnifying a noticeable flaw circumscribed rectangle 52, to obtain a magnified circumscribed rectangle 55 and a determination process for determining whether to connect or not a noticeable detection flaw Dn, adjacent detection flaws D<SB>n-1</SB>, D<SB>n+1</SB>and D<SB>n+2</SB>on the base of whether the magnified circumscribed rectangle 55 and the positions of adjacent flaw circumscribed rectangles 51, 53 and 54 overlap. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a defect detection method, and more particularly, to a defect detection method in which an attention detection defect and an adjacent detection defect adjacent to the attention detection defect are combined under a predetermined condition.

  Conventionally, appearance inspection apparatuses for inspecting defects have been used in the manufacturing process of semiconductor products. In such an appearance inspection apparatus, a phenomenon in which a detection defect detected by the inspection apparatus does not match an actual defect that actually exists due to the influence of noise or the like.

  FIG. 11 is an explanatory diagram schematically showing the existing defect and the detected defect. FIG. 11A shows a real defect, and FIG. 11B shows a detected defect.

  The defect detection target shown in this figure is a line-and-space semiconductor element in which linear conductive portions 100 and linear blank portions are alternately arranged. In FIG. As shown, the open defects 110 in which the conductive parts 100 are largely missing, the short defects 120 in which the conductive parts 100 are short-circuited in a wide range, the open defects 140 in which the conductive parts 100 are missing, and the conductive parts 100 are present as actual defects. There are short defects 130 that are short-circuited.

  When a defect is detected with respect to this defect detection target by an appearance inspection apparatus, as shown in FIG. 11B, the short defect 130 and the open defect 140 are accurately detected as defects 131 and 141, respectively. The However, for example, the open defect 110 may be detected as being divided into four defects 111, 112, 113, and 114, and the short defect 120 may be detected as being divided into three defects 121, 122, and 123.

  FIG. 12 is an explanatory diagram schematically showing a real defect and a detected defect with respect to a defect called a scratch defect. 12A shows a real defect, and FIG. 12B shows a detected defect.

  The defect detection target shown in this figure is also a line-and-space-shaped semiconductor element in which linear conductive portions 100 and linear blank portions are alternately arranged. In FIG. As shown, there are real defects 150 that span the plurality of conductive portions 100. Even when a defect is detected with respect to this defect detection target by a visual inspection apparatus, a single real defect 150 is detected by being divided into three defects 151, 152, and 153 as shown in FIG. May be.

  Thus, when a single actual defect is detected as a plurality of defects, the number of actual defects and the number of detected defects do not match, and accurate defect detection cannot be performed. Such a problem is a phenomenon that occurs not only when a defect detection object is imaged and the defect is detected, but also when a defect is detected by a VC (voltage contrast) inspection using an electron beam.

For this reason, in Patent Document 1, the position information of the detected defect is extracted, the coordinates of each defect and the distance between each defect are obtained based on this position information, and when this distance is smaller than a preset set value A visual inspection method for determining that a single defect is recognized as another defect and combining these defects is disclosed.
Japanese Patent No. 29986868

  In the appearance inspection method described in Patent Document 1, since the determination is performed using the distance between each defect, an erroneous determination is made with respect to a combination of defects characterized by a shape such as a line defect. There is a case.

  FIG. 13 is an explanatory diagram of a conventional appearance inspection method.

  As shown in FIG. 13A, it is assumed that the actual defects Da and Db exist in the line-and-space pattern. At this time, when this pattern is subjected to a defect inspection by an appearance inspection apparatus, due to the influence of noise or the like, as shown in FIG. 13B, four detected defects Da1, Da2, Da3, A case where Da4 is detected and one detected defect Db1 is detected with respect to the actual defect Db is shown.

  In such a case, as shown in FIG. 13C, the distance between the defects is obtained, and when this distance is smaller than a preset value, a process of recognizing a single defect as another defect is executed. In this case, as shown in FIG. 13D, the detected defects Da1, Da2, and Db2 are recognized as a single defect D1, and the detected defects Da3 and Da4 are recognized as single defects D2 and D3, respectively. This causes the problem of being done.

  The present invention has been made to solve the above-described problems. For example, even when a special shape such as a line and space is used, defects can be accurately detected by properly combining defects. It aims to provide a method.

  According to the first aspect of the present invention, in the defect detection method for combining the attention detection defect and the adjacent detection defect adjacent to the attention detection defect under a predetermined condition, the circumscribed rectangle of the attention detection defect is set as the attention defect circumscribed rectangle. Enlarging the attention defect circumscribing rectangle to be set by enlarging the attention defect circumscribing rectangle to the first direction in which the defect to be combined exists and the second direction orthogonal to the first direction, respectively. A circumscribed rectangle enlarging step for obtaining a circumscribed rectangle, and a determining step for determining whether or not the target detected defect and the adjacent detected defect are combined based on the relationship between the enlarged circumscribed rectangle and the position of the adjacent detected defect. It is characterized by that.

  According to a second aspect of the present invention, in the defect detection method for combining the attention detection defect and the adjacent detection defect adjacent to the attention detection defect under a predetermined condition, the circumscribed rectangle of the attention detection defect is set as the attention defect circumscribed rectangle. A target defect circumscribing rectangle setting step to be set; an adjacent defect circumscribing rectangle setting step of setting a circumscribed rectangle of an adjacent detected defect as an adjacent defect circumscribing rectangle; and a first direction in which a defect to be combined with the target defect circumscribing rectangle exists. Enlarging in a second direction orthogonal to the first direction to obtain an enlarged circumscribed rectangle and whether or not the positions of the enlarged circumscribed rectangle and the adjacent defect circumscribed rectangle overlap each other Thus, when the determination step of determining whether to combine the attention detection defect and the adjacent detection defect, and the position of the enlarged circumscribed rectangle and the adjacent defect circumscribed rectangle overlap, Further comprising a coupling step of coupling the serial noted detect defects and with said neighbor detecting defects characterized.

  According to a third aspect of the present invention, in a defect detection method for combining an attention detection defect and an adjacent detection defect adjacent to the attention detection defect under a predetermined condition, the circumscribed rectangle of the attention detection defect is set as the attention defect circumscribed rectangle. Enlarging the attention defect circumscribing rectangle to be set by enlarging the attention defect circumscribing rectangle to the first direction in which the defect to be combined exists and the second direction orthogonal to the first direction, respectively. A circumscribed rectangle enlarging step for obtaining a circumscribed rectangle, the enlarged circumscribed rectangle and the region in the first direction of the adjacent detection defect overlap, and the center of the enlarged circumscribed rectangle and the adjacent defect in the second direction A determination step of determining whether to combine the target detection defect and the adjacent detection defect according to whether or not the position overlaps; and the enlarged circumscribed rectangle and the region in the first direction of the adjacent detection defect And heavy And, when the enlarged circumscribed rectangle and the position of the center of gravity of the adjacent defect in the second direction overlap, a combining step of combining the attention detection defect and the adjacent detection defect is provided. Features.

  According to a fourth aspect of the present invention, in the invention according to any one of the first to third aspects, a straight line connecting two points separated most on a contour line in the attention detection defect is obtained as a maximum length line, and the attention defect The circumscribed rectangle is a rectangle whose long side is parallel to the maximum long line.

  The invention according to claim 5 is the invention according to claim 2, wherein a straight line connecting two points that are most separated on the contour line in the adjacent detection defect is obtained as a maximum long line, and the adjacent defect circumscribed rectangle has a long side thereof Is a rectangle parallel to the maximum long line.

  The invention according to claim 6 is the invention according to any one of claims 1 to 5, wherein the enlargement amount with respect to the first direction is larger than the enlargement amount with respect to the second direction.

  According to a seventh aspect of the present invention, in the invention according to any one of the first to fifth aspects, the enlargement amount with respect to the first direction and the enlargement amount with respect to the second direction are defined as the rectangle of interest circumscribed rectangle. It is determined based on the ratio of the long side to the short side.

  According to an eighth aspect of the present invention, in the invention according to any one of the first to seventh aspects, the defect detection target is a line and in which linear conductive portions and linear blank portions are alternately arranged. It has a space shape, and the first direction is the same direction as the linear conductive portion.

  According to the first to third aspects of the invention, even if the shape is a special shape such as a line and space, it is possible to accurately detect the defect by properly combining the defects.

  According to the fourth aspect of the present invention, it is possible to obtain an appropriate target defect circumscribed rectangle corresponding to the shape of the target detection defect.

  According to the fifth aspect of the present invention, an appropriate adjacent defect circumscribed rectangle corresponding to the shape of the adjacent detection defect can be obtained.

  According to the invention described in claim 6 and claim 7, it becomes possible to appropriately combine necessary defects.

  According to the eighth aspect of the present invention, it is possible to appropriately combine defects generated corresponding to the conductive portions in the line and space shape.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic view of an appearance inspection apparatus to which the present invention is applied.

  This appearance inspection apparatus is an apparatus for inspecting a pattern on a semiconductor substrate (hereinafter referred to as “substrate”) W, and acquires a multi-tone object image data by imaging a predetermined area on the substrate W. An imaging unit 2 that holds the substrate W, and a stage moving mechanism 4 that moves the stage 3 relative to the imaging unit 2.

  The imaging unit 2 is connected by a light source 21 that emits illumination light, an optical system 22 that irradiates the substrate W with illumination light emitted from the light source 21, and an optical system 22 that receives reflected light from the substrate W. An imaging device 23 that converts an image of the imaged substrate W into an electrical signal is included.

  Further, the stage moving mechanism 4 includes an X-direction moving mechanism 41 that moves the stage 3 in the X direction perpendicular to the paper surface in FIG. 1 and a Y-direction moving mechanism 42 that moves in the Y direction that is the left-right direction in FIG. Have. The X-direction moving mechanism 41 has a configuration in which a ball screw (not shown) is connected to the motor 43, and the entire Y-direction moving mechanism 42 moves in the X direction along the guide rail 44 as the motor 43 rotates. . The Y-direction moving mechanism 42 has the same configuration as the X-direction moving mechanism 41. When the motor 45 rotates, the stage 3 moves in the Y direction along the guide rail 46 by driving a ball screw (not shown). To do.

  The visual inspection apparatus further includes a test output unit 5 configured by an electrical circuit, and a computer 6 configured by a CPU that performs various arithmetic processes, a memory that stores various types of information, and the like. The inspection output unit 5 acquires data indicating the object image from the imaging unit 2 and performs image processing. Further, the computer 6 performs determination and the like to be described later based on the image data after the image processing, and also functions as a control unit that controls other configurations of the appearance inspection apparatus.

  Next, the defect detection operation by this appearance inspection apparatus will be described. FIG. 2 is a flowchart showing a defect detection operation by the appearance inspection apparatus.

  When executing the defect inspection operation, first, the defect detection operation is executed (step S1). In this case, the image obtained by imaging the surface of the substrate W by the imaging unit 2 is compared with the reference image, and an area that is equal to or greater than the threshold value is detected as a defect. As the reference image, a pattern image having the same shape existing on the substrate W to be inspected may be used, or a reference image may be formed from CAD data and used for comparison. As this defect detection operation, a conventional well-known method can be used.

  When the defect detection is completed, information on the defect is extracted and stored as defect information in a memory or the like in the computer 6 (step S2). The defect information includes, for example, a die chip index where the defect is located, an inspection area identification number where the defect is located, a circumscribed rectangle of the defect, a barycentric coordinate of the defect, a defect area, a defect image, and the like. The die chip index where the defect is located, the inspection area identification number where the defect is located, the circumscribed rectangle of the defect, etc. will be described in detail later.

  Next, a defect size filter is executed (step S3). Here, in order to prevent the noise component of the imaging system from being detected as a minute pseudo defect, when the defect area of the detected defect is smaller than a predetermined threshold area, the detected defect is recognized as a pseudo defect, Execute processing to be excluded from processing targets.

  Next, the detected defects are grouped into processing units (step S4).

  FIG. 3 is an explanatory diagram schematically showing a die map and an inspection area of the substrate W to be inspected.

  As shown in this figure, the substrate W is divided into, for example, seven areas of −3 to 3 in the I and J directions orthogonal to each other, and the die chip index (I , J). Each die chip has two inspection areas, and identification numbers A1 and A2 are assigned to the inspection areas, respectively.

  The number of detected defects on the substrate W may exceed tens of thousands, and when the combination determination described later is executed for all of these, the number of combinations becomes enormous and it takes time to determine. For this reason, detection defects are grouped into processing units to reduce the number of combinations of determinations. More specifically, only detected defects having the same die chip index (I, J) where the detected defect is located and the same identification number of the inspection area are grouped into the same group.

  Then, one of these groups is selected as a processing group (step S5), and the following operations are repeated for each processing group in the following operations.

  When a processing group is selected, detection defect combination determination is performed for the processing group (step S6). This detection defect combination determination is to determine whether or not an attention detection defect and an adjacent detection defect adjacent to this attention detection defect should be combined. In making this determination, the concept of a circumscribed rectangle is used.

  FIG. 4 is an explanatory diagram showing a circumscribed rectangle R for the detection defect D.

  In obtaining the circumscribed rectangle R, a straight line connecting two points that are most separated on the contour line in the detection defect D is obtained as the maximum long line L. Of the rectangles circumscribing the detection defect D, a rectangle whose long side is parallel to the maximum long line L is specified as the circumscribed rectangle R in the present invention.

  FIG. 5 is an explanatory diagram for explaining the determination operation in the detected defect combination determination step.

In the case of detecting defects binding determination, the detected defect certain of the plurality of the detected defect as a target detected defect D _n, adjacent detected defect D _n-1 of the detected defect adjacent _{thereto, D n + 1, D n} + ₂

Then, the circumscribed rectangle for the attention detection defect D _n is set as the attention defect circumscribing rectangle 52. In addition, circumscribed rectangles for adjacent detection defects D _n−1 , D _{n + 1} , and D _{n + 2} are set as adjacent defect circumscribed rectangles 51, 53, and 54.

In FIG. 5, the coordinates of the upper, lower, left and right end portions of the target defect circumscribed rectangle 52 are (RL _N , RT _N , RR _N , RB _N ). By enlarging the target defect circumscribed rectangle 52 by Tx in the first direction and Ty in the second direction, an enlarged circumscribed rectangle 55 is obtained.

Here, the first direction is generally a direction in which defects to be combined exist, and as shown in FIGS. 11 to 13, the conductive part 100 having a linear pattern and a linear blank part are formed. In the case of an alternately arranged line and space shape, the direction is the same as that of the linear conductive portion 100. The second direction is a direction orthogonal to the first direction. Note that the values of T _X and T _Y related to these enlargement amounts are set so that the enlargement amount in the first direction is larger than the enlargement amount in the second direction.

Then, it is determined whether or not the positions of the enlarged circumscribed rectangle 55 and the adjacent defect circumscribed rectangles 51, 53, and 54 overlap. At this time, among the adjacent defect circumscribed rectangles 51, 53, 54, adjacent defects D _n-1 and D _{n + 1} corresponding to the adjacent defect circumscribed rectangles 51, 53 whose positions overlap with the enlarged circumscribed rectangle 55 are determined as follows: determining a neighbor binding defect to be coupled to the target detected defect D _n. Such a combination determination operation is repeatedly executed for each detected defect, with each detected defect as a target detected defect.

  Referring to FIG. 2 again, when the above-described detected defect combination determination process is completed, a combined defect information generation process is executed (step S7). At this time, in addition to information such as the die chip index where the defect extracted in the detected defect information extraction step (step S2) and the inspection area identification number where the defect is located, the circumscribed rectangle of the entire defect, the barycentric coordinates of the entire defect, The area of the entire defect, the image of the entire defect, and the like. At this time, in order to represent the entire defect, for example, as shown in FIG. 6, a minimum rectangle 61 that covers the entire circumscribed rectangles 51, 52, and 53 for all the detected defects after the combination is used.

  Thereafter, it is determined whether or not the join determination has been completed for all groups (step S8). If there is an unprocessed group, the process returns to step S5 and is repeated. On the other hand, when the combination determination is completed for all the groups, the inspection result is stored (step S9), and the process ends.

  FIG. 7 shows the present invention when the real defects Da and Db shown in FIG. 13A are detected as four detected defects Da1, Da2, Da3 and Da4 as shown in FIG. 13B. It is explanatory drawing which shows the result of having couple | bonded this by the external appearance inspection method which concerns on 1st Embodiment. By comparing this figure with FIG. 13, it becomes clear that the detection defects can be appropriately combined according to the first embodiment of the present invention.

In FIG. 5, in the case where the pattern has a line and space shape in which the linear conductive portions 100 and the linear blank portions are alternately arranged, the detected defects D _n−1 , D _n , The case where D _{n + 1} and D _{n + 2} are in the same direction as the linear conductive part 100 has been described. On the other hand, as shown in FIG. 8, there are cases where the attention detection defect D _n and the adjacent detection defects D _n−1 , D _{n + 1} face the direction intersecting with the linear conductive part 100.

In the embodiment shown in FIG. 5, the values of Tx and Ty related to the amount of enlargement are set so that the amount of enlargement in the first direction is larger than the amount of enlargement in the second direction. On the other hand, as shown in FIG. 8, when the target detection defect D _n and the adjacent detection defects D _n−1 , D _{n + 1} face the direction intersecting the linear conductive part 100, the enlargement amount is related. The values of Th and Tv are determined based on the ratio of the long side to the short side of the target defect circumscribed rectangle 58.

  That is, in this embodiment, when T is a predetermined parameter, the values of Th and Tv related to the enlargement amount are determined by the following formula.

Th = T · [Sh / Sv]
Tv = T · [Sv / Sh]
Then, when the positions of the adjacent circumscribed rectangles 57 and 59 overlap with the enlarged circumscribed rectangle 60 enlarged using the Th and Tv values related to the enlargement amount, they correspond to the adjacent defect circumscribed rectangles 57 and 59. Adjacent defects D _n−1 and D _{n + 1} are determined as adjacent coupling defects to be coupled to the target detection defect D _n . By adopting such a configuration, it becomes possible to appropriately combine necessary defects according to the shape of the detected defect.

  Next, another embodiment of the determination operation will be described. FIG. 9 is an explanatory diagram illustrating a determination operation in the second embodiment. In this figure, the first and second directions are the same as those in FIG.

Also in this embodiment, a detection defect of a plurality of detection defects is set as a target detection defect D _n, and detection defects adjacent thereto are adjacent detection defects D _n−1 , D _{n + 1} , D _{n + 2} , D _{n. +3} . Further, with respect to the circumscribed rectangle for the attention detection defect D _n, an enlarged circumscribed rectangle enlarged by Tx in the first direction and Ty in the second direction is assumed. Also at this time, the values of T _X and T _Y relating to these enlargement amounts are set so that the enlargement amount in the first direction is larger than the enlargement amount in the second direction.

In the second embodiment, the assumed enlarged circumscribed rectangle overlaps with the region in the first direction of the adjacent detection defects D _n−1 , D _{n + 1} , D _{n + 2} , D _{n + 3.} And whether or not the assumed enlarged circumscribed rectangle and the position of the center of gravity of the adjacent detection defects D _n−1 , D _{n + 1} , D _{n + 2} , D _{n + 3 in} the second direction overlap. Thus, it is determined whether or not the attention detection defect D _n and the adjacent detection defects D _n−1 , D _{n + 1} , D _{n + 2} , D _{n + 3} and the attention detection defect are combined. In the embodiment in FIG. 9, in the end, the attention detected defect D _n and the adjacent detected defect D _n-1 is to be coupled.

  FIG. 10 shows the case where the above-mentioned real defects Da and Db shown in FIG. 13A are detected as four detected defects Da1, Da2, Da3 and Da4 as shown in FIG. 13B. It is explanatory drawing which shows the result of having couple | bonded this with the external appearance inspection method which concerns on 2nd Embodiment. By comparing this figure with FIG. 13, it becomes clear that the detection defects can be appropriately combined according to the second embodiment of the present invention.

1 is a schematic diagram of an appearance inspection apparatus to which the present invention is applied. It is a flowchart which shows the defect detection operation | movement by an external appearance inspection apparatus. It is explanatory drawing which shows typically the die map and inspection area of the board | substrate W which are inspection object. 4 is an explanatory diagram showing a circumscribed rectangle R for a detection defect D. FIG. It is explanatory drawing explaining the determination operation | movement in a detection defect coupling | bonding determination process. It is explanatory drawing which shows the minimum rectangle 61 which covers the circumscribed rectangles 51, 52, and 53 whole. It is explanatory drawing which shows the result of having combined four detection defects Da1, Da2, Da3, Da4 by the appearance inspection method which concerns on 1st Embodiment of this invention. FIG. 6 is an explanatory diagram showing a noticed detection defect D _n and adjacent detection defects D _n−1 and D _{n + 1} that are arranged to be inclined with respect to a conductive part 100. It is explanatory drawing explaining the determination operation | movement in 2nd Embodiment. It is explanatory drawing which shows the result of having combined four detection defects Da1, Da2, Da3, Da4 by the visual inspection method which concerns on 2nd Embodiment of this invention. It is explanatory drawing which shows a real defect and a detected defect typically. It is explanatory drawing which shows a real defect and a detected defect typically. It is explanatory drawing of the conventional external appearance inspection method.

Explanation of symbols

2 imaging unit 3 stage 4 stage moving mechanism 5 inspection output unit 6 computer 21 light source 22 optical system 23 imaging device 41 X direction moving mechanism 42 Y direction moving mechanism 43 motor 44 guide rail 45 motor 46 guide rail 51 adjacent defect circumscribed rectangle 52 attention Defect circumscribed rectangle 53 Adjacent defect circumscribed rectangle 54 Adjacent defect circumscribed rectangle 100 Conductive part D Detected defect L Maximum long line R circumscribed rectangle W Substrate

Claims (8)

In a defect detection method for combining an attention detection defect and an adjacent detection defect adjacent to the attention detection defect according to a predetermined condition,
Attention defect circumscribing rectangle setting step for setting the circumscribed rectangle of the attention detection defect as the attention defect circumscribing rectangle;
A circumscribed rectangle enlarging step of obtaining an enlarged circumscribed rectangle by enlarging the target defect circumscribed rectangle in a first direction in which a defect to be combined exists and a second direction orthogonal to the first direction,
From the relationship between the enlarged circumscribed rectangle and the position of the adjacent detection defect, a determination step of determining whether to combine the target detection defect and the adjacent detection defect;
A defect detection method comprising: In a defect detection method for combining an attention detection defect and an adjacent detection defect adjacent to the attention detection defect according to a predetermined condition,
Attention defect circumscribing rectangle setting step for setting the circumscribed rectangle of the attention detection defect as the attention defect circumscribing rectangle;
An adjacent defect circumscribed rectangle setting step for setting the circumscribed rectangle of the adjacent detection defect as an adjacent defect circumscribed rectangle;
A circumscribed rectangle enlarging step of obtaining an enlarged circumscribed rectangle by enlarging the target defect circumscribed rectangle in a first direction in which a defect to be combined exists and a second direction orthogonal to the first direction,
A determination step of determining whether to combine the target detection defect and the adjacent detection defect according to whether or not the position of the enlarged circumscribed rectangle and the adjacent defect circumscribed rectangle overlaps;
When the positions of the enlarged circumscribed rectangle and the adjacent defect circumscribed rectangle overlap, a combining step of combining the attention detection defect and the adjacent detection defect;
A defect detection method comprising: In a defect detection method for combining an attention detection defect and an adjacent detection defect adjacent to the attention detection defect according to a predetermined condition,
Attention defect circumscribing rectangle setting step for setting the circumscribed rectangle of the attention detection defect as the attention defect circumscribing rectangle;
A circumscribed rectangle enlarging step of obtaining an enlarged circumscribed rectangle by enlarging the target defect circumscribed rectangle in a first direction in which a defect to be combined exists and a second direction orthogonal to the first direction,
Whether the enlarged circumscribed rectangle and the area of the adjacent detection defect in the first direction overlap, and whether the enlarged circumscribed rectangle and the position of the center of gravity of the adjacent defect in the second direction overlap. A determination step of determining whether to combine the attention detection defect and the adjacent detection defect;
When the enlarged circumscribed rectangle and the region of the adjacent detection defect in the first direction overlap, and the enlarged circumscribed rectangle and the position of the center of gravity of the adjacent defect in the second direction overlap, A combining step of combining the attention detection defect and the adjacent detection defect;
A defect detection method comprising: In the defect detection method in any one of Claims 1 thru | or 3,
A defect detection method in which a straight line connecting two points that are farthest apart on a contour line in the target detection defect is obtained as a maximum long line, and the target defect circumscribed rectangle has a long side parallel to the maximum long line. The defect detection method according to claim 2,
A defect detection method in which a straight line connecting two points that are most separated on the contour line in the adjacent detection defect is obtained as a maximum long line, and the adjacent defect circumscribed rectangle has a long side parallel to the maximum long line. In the defect detection method in any one of Claims 1 thru | or 5,
The defect detection method, wherein an enlargement amount with respect to the first direction is larger than an enlargement amount with respect to the second direction. In the defect detection method in any one of Claims 1 thru | or 5,
A defect detection method for determining an enlargement amount with respect to the first direction and an enlargement amount with respect to the second direction based on a ratio between a long side and a short side of the defect circumscribed rectangle. The defect detection method according to any one of claims 1 to 7,
The defect detection target has a line-and-space shape in which linear conductive portions and linear blank portions are alternately arranged,
The defect detection method, wherein the first direction is the same direction as the linear conductive portion.