Abnormal region screening method, device and equipment, medium and program product

Abstract

The invention provides an abnormal region screening method and device, equipment, a medium and a program product, contour point extraction processing is performed on a to-be-selected image to obtain a plurality of contour points, a target point pair corresponding to the contour points is determined based on the contour points, and the target point pair comprises a first point and a second point. The first point and the second point are located on the two sides of the partial contour where the contour point is located respectively, and the first gray level of the first point and the second gray level of the second point are subjected to gray level comparison to obtain a first comparison result of the target point pair. As the gray scale difference between the real abnormal area and the periphery of the real abnormal area is larger, and the gray scale difference between the false abnormal area and the periphery of the false abnormal area is smaller, when the first comparison results meeting the preset number all indicate that the difference between the first gray scale and the second gray scale reaches the preset difference, the false abnormal area is judged. The gray scale differences between the inner sides and the outer sides of most contour points are large enough, the to-be-selected area is accurately indicated to be the abnormal area, false detection of a false abnormal area as the abnormal area is avoided, and the accuracy of abnormal area screening is improved.

Description

Technical Field

[0001] The present application relates to the field of image detection, and particularly relates to a method, device, equipment, medium and program product for screening abnormal regions. Background Art

[0002] There are candidate regions in the candidate image, and it is necessary to accurately determine whether the candidate region is an abnormal region, such as whether it is the region where the real defect is located. When collecting the candidate image, due to reasons such as optical system imaging, the signal intensity of some real defects is weak and it is difficult to detect.

[0003] In the related art, usually a lower segmentation threshold is used for detection. If the gray value of the candidate region is higher or lower than a certain gray threshold, it indicates that the candidate region is an abnormal region, such as the region where the real defect is located.

[0004] However, while this method detects real defects, it will also misdetect background noise with weak signal intensity as real defects, that is, misdetect the region where the noise is located as an abnormal region, and it is impossible to accurately determine whether the candidate region is really an abnormal region, resulting in misdetection of abnormal regions and low product yield. Therefore, providing a suitable method for screening abnormal regions has become an urgent technical problem to be solved at present. Summary of the Invention

[0005] In view of this, the purpose of the present application is to provide a method, device, equipment, medium and program product for screening normal regions, which greatly improves the accuracy of screening abnormal regions. The specific scheme is as follows: On the one hand, the present application provides a method for screening abnormal regions, including: Performing contour point extraction processing on the candidate image to obtain a plurality of contour points; the plurality of contour points are used to identify the contour of the candidate region of the candidate image; Determining a target point pair corresponding to the contour point based on the contour point; the target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the partial contour where the contour point is located; Performing gray comparison on the first gray value of the first point and the second gray value of the second point to obtain a first comparison result of the target point pair; When a preset number of the first comparison results all indicate that the difference between the first gray value and the second gray value reaches a preset difference, determining that the candidate region is an abnormal region.

[0006] Optionally, before determining the target point pair corresponding to the contour point based on the contour point, the method further includes: Comparing the size parameter of the candidate region with a preset parameter to obtain a second comparison result; When the second comparison result indicates that the size parameter is greater than or equal to the preset parameter, the determining of the target point pair corresponding to the contour point based on the contour point includes: Determine the point located outside the partial contour where the contour point is located as the first point; Determine the point located inside the partial contour where the contour point is located as the second point; the first point, the second point, and the contour point are on the same straight line.

[0007] Optionally, when the second comparison result indicates that the size parameter is less than the preset parameter, the determining of the target point pair corresponding to the contour point based on the contour point includes: Determine the point located outside the partial contour where the contour point is located as the first point; Determine the center point of the candidate area as the second point.

[0008] Optionally, the determining the point located outside the partial contour where the contour point is located as the first point includes: Determine a first reference point and a second reference point on the contour based on the contour point; the first reference point and the second reference point are close to the contour point, and the contour point is located between the first reference point and the second reference point; Determine a second straight line passing through the contour point and perpendicular to the first straight line where the first reference point and the second reference point are located; Determine the point located on the second straight line and outside the contour as the first point.

[0009] Optionally, the determining the point located on the second straight line and outside the contour as the first point includes: Perform an expansion process on the contour based on an expansion parameter to obtain an expanded contour of the candidate area; the expanded contour has the same shape as the contour; Determine the intersection point closer to the contour point among at least one intersection point of the second straight line and the expanded contour as the first point.

[0010] Optionally, there are multiple other contour points between the first reference point or the second reference point and the contour point.

[0011] Optionally, before determining the candidate area as an abnormal area when the first comparison results of the preset quantity all indicate that the difference between the first gray level and the second gray level reaches a preset difference, the method further includes: Perform a gray level comparison between the contour point and the target area where the contour point is located to obtain a third comparison result; When the gray level of the contour point indicated by the third comparison result is greater than or equal to the gray level of the target area, when the differences between the first gray level and the second gray level indicated by the first comparison results that meet the preset quantity reach the preset difference, determining that the candidate area is an abnormal area includes: When the differences between the second gray level and the first gray level indicated by the first comparison results that meet the preset quantity are greater than the preset difference, determining that the candidate area is the abnormal area; the preset difference is greater than 0; When the gray level of the contour point indicated by the third comparison result is less than the gray level of the target area, when the differences between the first gray level and the second gray level indicated by the first comparison results that meet the preset quantity reach the preset difference, determining that the candidate area is an abnormal area includes: When the differences between the second gray level and the first gray level indicated by the first comparison results that meet the preset quantity are less than the preset difference, determining that the candidate area is the abnormal area; the preset difference is less than 0.

[0012] Optionally, when the differences between the first gray level and the second gray level indicated by the first comparison results that meet the preset quantity reach the preset difference, determining that the candidate area is an abnormal area includes: When the ratio of the quantity of the first comparison results indicating that the difference between the first gray level and the second gray level reaches the preset difference to the quantity of all contour points reaches a first ratio, determining that the candidate area is the abnormal area.

[0013] Optionally, when the ratio of the quantity of the first comparison results indicating that the difference between the first gray level and the second gray level reaches the preset difference to the quantity of all contour points reaches a first ratio, determining that the candidate area is an abnormal area includes: Determining that the candidate area is located in the edge area of the candidate image; When the ratio of the quantity of the first comparison results indicating that the difference between the first gray level and the second gray level reaches the preset difference to the quantity of all contour points reaches a second ratio, determining that the candidate area is the abnormal area; the second ratio is less than the first ratio.

[0014] Optionally, before performing the contour point extraction process on the candidate image to obtain a plurality of contour points, the method further includes: Performing defect detection on a semiconductor structure to obtain the candidate image of the semiconductor structure; the candidate image includes the candidate area, and the candidate area is the area where a real defect or a false defect is located; When it is determined that the differences between the first grayscale and the second grayscale reach a preset difference in all of the first comparison results that meet a preset quantity, determining that the candidate area is an abnormal area, includes: When it is determined that the differences between the first grayscale and the second grayscale reach a preset difference in all of the first comparison results that meet a preset quantity, determining that the candidate area is the area where the true defect is located.

[0015] Optionally, the extracting contour points from the candidate image to obtain a plurality of contour points includes: Performing image enhancement processing on the candidate image to obtain an enhanced image of the candidate image; Performing contour point extraction processing on the enhanced image to obtain the plurality of contour points; the plurality of contour points are used to identify the contour of the candidate area of the enhanced image.

[0016] In another aspect, an abnormal area screening device provided by an embodiment of the present application includes: An extraction unit, configured to perform contour point extraction processing on a candidate image to obtain a plurality of contour points; the plurality of contour points are used to identify the contour of the candidate area of the candidate image; A first determination unit, configured to determine a target point pair corresponding to the contour points based on the contour points; the target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the partial contour where the contour points are located; A comparison unit, configured to perform grayscale comparison on the first grayscale of the first point and the second grayscale of the second point to obtain a first comparison result of the target point pair; A second determination unit, configured to determine that the candidate area is an abnormal area when it is determined that the differences between the first grayscale and the second grayscale reach a preset difference in all of the first comparison results that meet a preset quantity.

[0017] In another aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory: The memory is configured to store program code and transmit the program code to the processor; The processor is configured to execute the method described in the above aspects according to the instructions in the program code.

[0018] In another aspect, an embodiment of the present application provides a computer-readable storage medium, which is configured to store a computer program, and the computer program is used to execute the method described in the above aspects.

[0019] In another aspect, an embodiment of the present application provides a computer program product including a computer program, and when the computer program product runs on a computer device, the computer device is enabled to execute the method described above.

[0020] An embodiment of the present application provides an abnormal area screening method, device, equipment, medium and program product. The candidate image is subjected to contour point extraction processing to obtain a plurality of contour points, and the plurality of contour points are used to reflect the contour of the candidate area of the candidate image. In order to accurately determine whether the candidate area is really an abnormal area, target point pairs corresponding to the contour points are determined based on the contour points. The target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the partial contour where the contour point is located, so as to be able to represent the inner side and the outer side of the contour respectively. The first gray level of the first point and the second gray level of the second point can represent the gray level distribution of the inner and outer sides of the contour. The first gray level of the first point and the second gray level of the second point are compared in gray level to obtain a first comparison result of the target point pair, and the first comparison result can reflect the gray level difference between the inner and outer sides of the contour. Since the gray level difference between the real abnormal area and its periphery is larger, while the gray level difference between the false abnormal area and its periphery is smaller, when the first comparison results that meet the preset quantity all indicate that the difference between the first gray level and the second gray level reaches the preset difference, it means that the gray level differences between the inner and outer sides of most of the contour points are large enough, so as to accurately indicate that the candidate area is an abnormal area, avoid misdetecting the false abnormal area as an abnormal area, and greatly improve the accuracy of abnormal area screening. Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0022] Figure 1 A flowchart showing an abnormal area screening method provided by an embodiment of the present application; Figure 2 A related schematic diagram of a candidate image provided by an embodiment of the present application; Figure 3 A schematic diagram of a target point pair provided by an embodiment of the present application; Figure 4 A related schematic diagram of another candidate image provided by an embodiment of the present application; Figure 5 A related schematic diagram of still another candidate image provided by an embodiment of the present application; Figure 6 A structural block diagram of an abnormal area screening device provided by an embodiment of the present application; Figure 7 A structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0023] To make the above objects, features, and advantages of the present application more obvious and understandable, the following will describe the detailed implementation manners of the present application with reference to the accompanying drawings.

[0024] In the following description, many specific details are set forth to facilitate a thorough understanding of the present application. However, the present application may be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present application. Therefore, the present application is not limited by the specific embodiments disclosed below.

[0025] As described in the background art, the method of using a lower segmentation threshold to screen for abnormal regions has low detection accuracy. As an example, in the semiconductor field, it is necessary to accurately detect real defects on a semiconductor structure. Real defects are, for example, particles, scratches, lattice defects, etc. attached to the surface of the semiconductor structure. Due to differences in substrate materials and the processes of each manufacturer, there are certain differences in the imaging characteristics of real defects. Coupled with problems such as the self-stability of the optical system, it will cause uneven illumination of the semiconductor structure, and there are cases where the signal intensity of some real defects in the candidate image is weak. If a lower segmentation threshold is used for detection, both real defects and false defects will be detected simultaneously. False defects are, for example, caused by noise interference or the defect extraction algorithm being too sensitive, resulting in inaccurate defect detection of the semiconductor structure.

[0026] Based on the above technical problems, the embodiments of the present application provide an abnormal region screening method, device, equipment, medium, and program product. The candidate image is subjected to contour point extraction processing to obtain a plurality of contour points, and the plurality of contour points are used to reflect the contour of the candidate region of the candidate image. To accurately determine whether the candidate region is truly an abnormal region, target point pairs corresponding to the contour points are determined based on the contour points. The target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the partial contour where the contour point is located, so as to be able to represent the inner and outer sides of the contour respectively. The first gray level of the first point and the second gray level of the second point can represent the gray level distribution inside and outside the contour. The first gray level of the first point and the second gray level of the second point are compared in gray level to obtain a first comparison result of the target point pair, and the first comparison result can reflect the gray level difference between the inner and outer sides of the contour. Since the gray level difference between a real abnormal region and its periphery is larger, while the gray level difference between a false abnormal region and its periphery is smaller, when the first comparison results of a preset number all indicate that the difference between the first gray level and the second gray level reaches a preset difference, it means that the gray level differences between the inner and outer sides of most of the contour points are large enough, so as to accurately indicate that the candidate region is an abnormal region, avoiding misdetecting a false abnormal region as an abnormal region, and greatly improving the accuracy of abnormal region screening.

[0027] The abnormal area screening method provided by the embodiments of the present application can be implemented by a computer device, which can be a terminal device or a server. Among them, the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Terminal devices include, but are not limited to, mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, etc. The terminal device and the server can be directly or indirectly connected through wired or wireless communication methods, and the present application does not make any restrictions on this.

[0028] For the convenience of understanding, a method, device, equipment, medium and program product for screening abnormal areas provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0029] Refer to Figure 1 As shown, it is a schematic flowchart of a method for screening abnormal areas provided by the embodiments of the present application, and this method can include the following steps.

[0030] S101, perform contour point extraction processing on the candidate image to obtain a plurality of contour points.

[0031] Specifically, the candidate image is an image for which abnormal area screening needs to be performed. The candidate image can be, for example, an image obtained by photographing a semiconductor structure, or other images. The contour points of the candidate image can be extracted to obtain a plurality of contour points. The plurality of contour points are used to identify the contour of the candidate area of the candidate image. Among them, the candidate area can be understood as the area that needs to be screened in the candidate image, and the contour points are the points located on the contour of the candidate area. The plurality of contour points can form the entire contour of the candidate area.

[0032] Refer to Figure 2 As shown, it is a schematic diagram related to a candidate image provided by the embodiments of the present application. (a) is a schematic diagram of the candidate image, and the black area in the figure is the candidate area, and the contour points are the points located on the edge of the black area. (b) is a schematic diagram of the candidate area in (a), and the white area in the figure is the candidate area identified by the contour points, that is, the mask of the extracted candidate area. (c) is a schematic diagram of another candidate image, and (d) is a schematic diagram of the candidate area in (c), and the white area in the figure is misidentified as the candidate area, that is, the mask of the extracted candidate area.

[0033] In practical applications, the candidate image may have one or more candidate areas. For each candidate area, contour point extraction processing can be performed to obtain the contour points of each candidate area, so as to perform abnormal identification on each candidate area subsequently.

[0034] In a possible implementation, in order to improve the accuracy of the collected contour points, S101 performs contour point extraction processing on the candidate image to obtain a plurality of contour points, which may specifically include S1011 - S1012.

[0035] S1011, perform image enhancement processing on the candidate image to obtain an enhanced image of the candidate image.

[0036] Specifically, image enhancement processing can be performed on the candidate image. Image enhancement processing can improve the contrast between the background area and the candidate area of the candidate image, and general image enhancement methods can be used for processing. The candidate image after image enhancement processing can be denoted as the enhanced image. In short, compared with the candidate image, the enhanced image can enhance the signal of the candidate area. Of course, it may also enhance the noise signal in the background area, but the noise signal can be excluded by other means later.

[0037] S1012, perform contour point extraction processing on the enhanced image to obtain a plurality of contour points; the plurality of contour points are used to identify the contour of the candidate area of the enhanced image.

[0038] Specifically, the contour points of the enhanced image can be extracted to obtain a plurality of contour points. At this time, the contour points are the points located on the contour of the candidate area of the enhanced image. The contour points collected from the enhanced image can represent the contour of the candidate area more clearly and accurately than the contour points collected from the candidate image, improving the accuracy of the contour points, and further improving the accuracy of subsequent abnormal area screening.

[0039] S102, determine the target point pair corresponding to the contour point based on the contour point.

[0040] Specifically, for a single contour point, the target point pair corresponding to the contour point can be determined. The target point pair can include a first point and a second point, that is, the target point pair can be a point pair composed of at least two points. The first point and the second point are respectively located on both sides of the partial contour where the contour point is located. That is to say, for a certain contour point, if its first point is located outside the partial contour where the contour point is located, its second point is located inside this partial contour. If the first point is located inside the partial contour where the contour point is located, the second point is located outside this partial contour. Among them, the partial contour where the contour point is located can be understood as a part of the contour near the contour point on the entire contour of the candidate area.

[0041] Reference Figure 2 As shown, in (a), the green points are the set of inner contour points, that is, the set of second points, and the red points are the set of outer contour points, that is, the set of first points. In (c), similarly, the green points are the set of inner contour points, and the red points are the set of outer contour points. Reference Figure 3As shown in the figure, it is a schematic diagram of a target point pair provided by an embodiment of the present application. The closed curve in the figure is the contour line of the area to be selected. For the contour point a located on the contour line, its target point pair includes the first point d located outside the contour and the second point e located inside the contour.

[0042] In a possible implementation manner, before S102 determines the target point pair corresponding to the contour point based on the contour point, the method may further include S1021 comparing the size parameter of the area to be selected with a preset parameter to obtain a second comparison result. When the second comparison result indicates that the size parameter is greater than or equal to the preset parameter, S102 determining the target point pair corresponding to the contour point based on the contour point may include S1021 - S1022.

[0043] S1021, determine the point located outside the part of the contour where the contour point is located as the first point.

[0044] S1022, determine the point located inside the part of the contour where the contour point is located as the second point.

[0045] Since the sizes of the areas to be selected are some large and some small, in order to further improve the detection accuracy, the first point and the second point can be determined according to the size of the area to be selected. Among them, the size parameter of the area to be selected can be understood as a parameter representing the size of the area to be selected. For example, it can be the area of the area to be selected, the aspect ratio of the area to be selected, etc. The preset parameter is a preset size parameter, and the preset parameter can be a preset area, a preset aspect ratio, etc. For the convenience of comparison, the size parameter of the area to be selected and the preset parameter can be the same type of parameter. For example, both are areas, or both are aspect ratios, etc.

[0046] Specifically, comparing the size parameter of the area to be selected with the preset parameter can obtain a second comparison result. When the second comparison result indicates that the size parameter of the area to be selected is not less than the preset parameter, it means that the size of the area to be selected is relatively large. For example, the area of the area to be selected is relatively large, which is a large area to be selected. Also for example, the aspect ratio of the area to be selected is relatively large, which is a long and narrow area to be selected. For such relatively large areas to be selected, due to their large sizes, there may very likely be a situation of local gray - level non - uniformity in the area to be selected, that is, the gray - level values at different positions inside the area to be selected vary greatly. Therefore, it is necessary to determine relatively accurate first and second points for each contour point.

[0047] At this time, for a single contour point, the point located outside the part of the contour where the contour point is located can be determined as the first point, and the point located inside this part of the contour can be determined as the second point. The first point, the second point, and the contour point are on the same straight line. That is to say, both the first point and the second point are points near the contour point and are located on both sides of the contour.

[0048] As an example, refer toFigure 3 As shown, the area of the candidate region is large. Therefore, for the contour point a, the first point d and the second point e are located on both sides of the contour point a, and these three points are on the same straight line.

[0049] In summary, for a candidate region with a large size, by specifically determining the first point and the second point for each contour point, the contour point and the two points on its both sides are on the same straight line. Thus, the second point located inside is always near the contour point, which can accurately represent the gray-scale distribution near the contour point, avoiding the inaccurate situation caused by using an internal point far from the contour point as the second point, overcoming the uneven gray-scale distribution in the internal region due to the large size of the candidate region, improving the accuracy of the target point pair, and further improving the detection accuracy of the abnormal region.

[0050] In a possible implementation manner, when the size parameter of the second comparison result identifier is less than the preset parameter, S102 determining the target point pair corresponding to the contour point based on the contour point may include S1023 - S1024.

[0051] S1023, determining the point located outside the part of the contour where the contour point is located as the first point.

[0052] S1024, determining the center point of the candidate region as the second point.

[0053] Specifically, when the second comparison result indicates that the size parameter of the candidate region is less than the preset parameter, it means that the size of the candidate region is small. For example, it is a candidate region with a small area or a candidate region with a regular shape. To reduce the calculation amount on the premise of ensuring the detection accuracy, there is no need to determine the corresponding second point for each contour point once.

[0054] At this time, for each contour point, the point located outside the contour can be determined as the first point. That is, the first point is still determined based on the position of the contour point, and the first point is located outside the contour point. Since the size of the candidate region is small at this time, there is no need to determine the corresponding second point for each contour point, and the same second point can be used for multiple contour points. That is to say, the center point of the candidate region can be determined and the center point can be determined as the second point. That is, no matter where the contour point is located, its corresponding second point is the center point.

[0055] In this way, for a relatively small candidate region, by using the center point to represent the second point, the center point only needs to be determined once, instead of determining the second point for each contour point. The number of times of determining points is greatly reduced, and the related calculation amount is significantly decreased. In addition, since the size of the candidate region is small, the center point is relatively close to each contour point, and the center point can still accurately represent the gray level distribution inside each contour point. That is, on the premise of ensuring the accuracy of the second point, the calculation amount is greatly reduced.

[0056] In a possible implementation manner, S1021 or S1023 determines the point located outside the partial contour where the contour point is located as the first point, which may specifically include S201 - S203.

[0057] S201, based on the contour point, determines a first reference point and a second reference point on the contour.

[0058] Specifically, the first reference point is a point located on the contour, and the second reference point is another point located on the contour. For each contour point, the corresponding first reference point and second reference point can be determined. The first reference point and the second reference point are close to the contour point, that is, both of these reference points are located near the contour point. In addition, the contour point is located between the first reference point and the second reference point.

[0059] That is to say, for a single contour point, the corresponding first reference point and second reference point can be determined for each contour point. On the contour line of the candidate region, the first reference point can be a point located on the left side of the contour point, and the second reference point can be a point located on the right side of the contour point.

[0060] S202, determines a second straight line that passes through the contour point and is perpendicular to the first straight line where the first reference point and the second reference point are located.

[0061] Specifically, the straight line passing through the first reference point and the second reference point is denoted as the first straight line, and the first straight line is used as the demarcation line between the inside and the outside of the contour line. Based on the first straight line, another straight line that passes through the contour point and is perpendicular to the first straight line is determined, that is, the second straight line. The second straight line is the straight line used to represent the direction from the inside to the outside of the candidate region.

[0062] Reference Figure 3 As shown, for the contour point a, the first reference point can be the point b, and the second reference point can be the point c. Then the first straight line is L 1 , and the second straight line is L 2 , L 2 passes through the inside and the outside of the candidate region.

[0063] S203, determines the point located on the second straight line and outside the contour as the first point.

[0064] Specifically, a point on the second straight line and outside the contour can be determined as the first point. Additionally, when determining the second point corresponding to each contour point, a point on the second straight line and inside the contour can be determined as the second point. As an example, as shown in Figure 3 point d as the first point not only lies on the second straight line but also outside the contour of the candidate area. Point e as the second point also lies on the second straight line.

[0065] In practical applications, taking the contour point as a reference, a point can be taken at a distance of L = gap along the outer direction of the second straight line as the first point, and a point can be taken at a distance of L = gap along the inner direction of the second straight line as the second point. Then, the first point and the second point form a target point pair (i.e., an inner and outer point pair). In addition, all contour points can be traversed in sequence to obtain the inner and outer point pairs corresponding to each contour point.

[0066] In summary, by determining the first straight line representing the approximate orientation of the position where the contour point is located, and then determining the second straight line perpendicular to the first straight line, the direction where the second straight line is located can represent the approximate orientation of the inside and outside of the contour. Furthermore, the outer point (i.e., the first point) is determined on the second straight line. The orientation between the first point and the contour point is more consistent, which can more accurately represent the points outside the contour and improve the accuracy of the first point.

[0067] In a possible implementation, S203 determines the point on the second straight line and outside the contour as the first point, which may specifically include S2031 - S2032.

[0068] S2031, based on the expansion parameter, expands the contour to obtain the expanded contour of the candidate area.

[0069] Specifically, the expansion parameter can be understood as the size parameter for expanding the candidate area, and the expansion parameter can be represented by gap. For example, the expansion parameter can be 2 times, etc. Based on the expansion parameter, the contour of the candidate area can be expanded to obtain the expanded contour of the candidate area. The expanded contour is the boundary contour after the candidate area is expanded. The expanded contour is consistent with the shape of the contour, that is, before and after the expansion process, the boundary shape of the candidate area remains unchanged, thus ensuring the consistency of the shape of the candidate area.

[0070] S2032, determines the intersection point closer to the contour point among at least one intersection point of the second straight line and the expanded contour as the first point.

[0071] Specifically, since the second straight line is in the direction from the inside to the outside (or from the outside to the inside) of the candidate area, the second straight line will pass through the expanded contour, and there will be at least one intersection point between the second straight line and the expanded contour. For example, if the second straight line completely penetrates the expanded contour, there will be at least two intersection points between the second straight line and the expanded contour.

[0072] The intersection point close to the contour point can be determined as the first point. This is because the closer the intersection point is to the contour point, it indicates that the intersection point is more located near the contour point, and thus can best represent the gray-scale distribution outside the contour point. Therefore, it can be used as the first point.

[0073] In practical applications, for each contour point, its corresponding second straight line can be determined, and the intersection points of the second straight line and the enlarged contour are calculated, so as to obtain the first point corresponding to each contour point.

[0074] In short, by calculating the intersection points of the second straight line and the enlarged contour, the corresponding first point can be quickly and accurately determined for each contour point, thereby improving the detection accuracy of the abnormal area.

[0075] In addition, for the second point, the contour can also be shrunk based on the shrinking parameter to obtain a shrunk contour. Among at least one intersection point where the second straight line intersects the shrunk contour, the intersection point close to the contour point can be used as the second point, so as to determine a more accurate inner point for each contour point.

[0076] In a possible implementation manner, there are multiple other contour points between the first reference point or the second reference point and the contour point. That is to say, the first reference point and the second reference point can both be contour points, and there is a certain distance between them and the corresponding contour points. As an example, the first reference point b and the second reference point c of the contour point a can be obtained at intervals. For example, b = a + m, c = a - m, where m is the number of adjacent contour points at intervals. This can avoid the situation that the direction of the first straight line is too skewed due to the two reference points being too close to the contour point, and improve the direction accuracy of the first straight line.

[0077] S103. Compare the first gray scale of the first point with the second gray scale of the second point to obtain the first comparison result of the target point pair.

[0078] Specifically, each point in the candidate image has a gray-scale value, which can be, for example, in the range of 0 - 255. The gray-scale value can reflect the brightness and darkness of the position where the point is located. The first gray scale can be understood as the gray-scale value of the first point, and the second gray scale can be understood as the gray-scale value of the second point. The first gray scale and the second gray scale are compared to obtain the first comparison result of the target point pair. The first comparison result can be used to identify the difference between the first gray scale and the second gray scale. For example, the first gray scale is greater than the second gray scale, and the difference between them reaches 50.

[0079] S104. When the first comparison results meeting the preset quantity all indicate that the difference between the first gray scale and the second gray scale reaches the preset difference, determine the candidate area as the abnormal area.

[0080] Specifically, the preset difference can be understood as a preset gray-scale difference value, for example, it can be 100. For a certain contour point, if its first comparison result indicates that the difference between the first gray scale and the second gray scale is greater than or equal to the preset difference, it means that the difference between the two is large enough, indicating that the brightness and darkness difference between the inside and outside of the candidate area is large, then the candidate area is very likely to be an abnormal area. Among them, the abnormal area can be understood as an area different from other areas, and the gray scale changes significantly inside the abnormal area.

[0081] To improve the detection accuracy, a preset quantity can be set. If for the contour points that meet the preset quantity, their first comparison results all indicate that the difference between the first gray scale and the second gray scale is the preset difference, it means that at the positions of these contour points, the gray-scale difference between the inside and outside of the contour is large, and the situation where the candidate area is the area where the noise is located is excluded, further increasing the possibility that the candidate area is an abnormal area, so that the candidate area can be determined to be an abnormal area. As an example, the preset quantity is 200, then if the first comparison results corresponding to 200 contour points all indicate that the difference between the first gray scale and the second gray scale is large enough, it means that the candidate area is an abnormal area.

[0082] Reference Figure 2 As shown, in (b), the first comparison results that meet the preset quantity indicate that the gray-scale difference reaches the preset difference, so the candidate area in the figure is an abnormal area. In (d), the condition is not met, so the candidate area in the figure is not an abnormal area. In addition, when the candidate image is an image of a semiconductor structure, (b) indicates that the candidate area is a real defect, and (d) indicates a false defect.

[0083] Reference Figure 4 As shown, this is a schematic diagram related to another candidate image provided by the embodiment of the present application. (a) is the candidate image, the white area in (b) is the mask of each candidate area, and there is a misdetection of noise in it. (c) is the mask of the abnormal area screened after the gray-scale difference comparison, and the noise mask has been deleted. Reference Figure 5 As shown, this is a schematic diagram related to another candidate image provided by the embodiment of the present application. (a) is another candidate image, the white area in (b) is the mask of each candidate area, and there is a misdetection of noise in it. (c) is the mask of the abnormal area screened after the gray-scale difference comparison, and the noise mask has been deleted.

[0084] In summary, since the gray-level difference between the real abnormal area and its periphery is greater, while the gray-level difference between the false abnormal area and its periphery is smaller, when the first comparison results of the preset quantity all indicate that the difference between the first gray level and the second gray level reaches the preset difference, it shows that the gray-level differences between the inner and outer sides of most contour points are large enough, so that it can accurately indicate that the area to be selected is an abnormal area, avoiding misdetecting a false abnormal area as an abnormal area, such as the area where noise is located, and greatly improving the accuracy of abnormal area screening.

[0085] In a possible implementation manner, before performing the contour point extraction process on the image to be selected in S101 to obtain a plurality of contour points, the method may further include S301 performing defect detection on the semiconductor structure to obtain the image to be selected of the semiconductor structure. Then, when the first comparison results of the preset quantity all indicate that the difference between the first gray level and the second gray level reaches the preset difference in S104, determining that the area to be selected is an abnormal area may specifically be that in S1041, when the first comparison results of the preset quantity all indicate that the difference between the first gray level and the second gray level reaches the preset difference, determining that the area to be selected is the area where the real defect is located.

[0086] Specifically, defect detection may be performed on the semiconductor structure to obtain the image to be selected of the semiconductor structure. The defect detection method may be dark field detection, bright field detection, photoluminescence detection, etc. Among them, dark field detection is to detect defects such as particles using the scattering principle, bright field detection is mainly used to detect defects on the surface of the wafer, and photoluminescence detection is used to identify lattice defects. Since the emission wavelengths of different defects are different, photoluminescence detection can be further divided into near-ultraviolet photoluminescence (NUVPL) and visible photoluminescence (VISPL) channels.

[0087] The image to be selected includes the area to be selected, and the area to be selected is the area where the real defect or the false defect is located. For the image to be selected of the semiconductor structure, the area to be selected therein may be the area where the real defect caused by process problems or imaging system optical path problems is located, or may be the area where the false defect caused by some noise is located. In order to accurately distinguish the real defect, when determining that the area to be selected is an abnormal area, if the first comparison results of the preset quantity indicate that the gray-level difference meets the preset difference, it may be determined that the area to be selected is the area where the real defect is located, rather than the area where the false defect is located. In this way, for the semiconductor detection field, the present application can accurately and quickly identify the real defects in the semiconductor structure, providing an accurate adjustment basis for subsequent process adjustment.

[0088] In a possible implementation manner, before determining that the area to be selected is an abnormal area when the first comparison results of the preset quantity all indicate that the difference between the first gray level and the second gray level reaches the preset difference, it may further include S401 comparing the gray level of the contour point with the target area where the contour point is located to obtain a third comparison result; When the gray value of the contour point marked by the third comparison result is greater than or equal to the gray value of the target area, and when the differences between the first gray value and the second gray value marked by a preset number of first comparison results all reach a preset difference, it is determined that the candidate area is an abnormal area, which may specifically include that in S1042, when the differences between the second gray value and the first gray value marked by a preset number of first comparison results are all greater than the preset difference, it is determined that the candidate area is an abnormal area; the preset difference is greater than 0. When the gray value of the contour point marked by the third comparison result is less than the gray value of the target area, in S104, when the differences between the first gray value and the second gray value marked by a preset number of first comparison results all reach a preset difference, it is determined that the candidate area is an abnormal area, which may specifically include that in S1043, when the differences between the second gray value and the first gray value marked by a preset number of first comparison results are all less than the preset difference, it is determined that the candidate area is an abnormal area; the preset difference is less than 0.

[0089] Specifically, the target area can be understood as the surrounding area of the position where the contour point is located. The gray value of the target area can be calculated, and this gray value can reflect the brightness and darkness of the target area. The gray value of the contour point and the gray value of the target area can be compared in terms of gray value, so as to obtain the third comparison result. The third comparison result can represent the comparison situation of the gray values of the contour point and the target area.

[0090] In addition, due to the different ways of obtaining the candidate image, the candidate area in the candidate image may be black or white. As an example, in the candidate image obtained by dark field detection, the candidate area is brighter and more inclined to white, and in the candidate image obtained by bright field detection, the candidate area is darker and more inclined to black.

[0091] Next, the situation where the third comparison result indicates that the gray value of the contour point is not less than the gray value of the target area will be described. When the gray value of the contour point is not less than the gray value of the target area, it indicates that the position where the contour point is located is brighter than the surrounding area. For example, the contour point is a point on the contour of the white candidate area, and it is necessary to accurately identify the white candidate area.

[0092] Since when the candidate area is white, the gray value inside the candidate area will be greater than the external gray value, that is, the second gray value of the second point is greater than the first gray value of the first point. Therefore, it can be set that the preset difference is greater than 0. If the first comparison result indicates that the difference between the second gray value and the first gray value is greater than the preset difference, it means that the second gray value is greater than the first gray value, that is, the inner point of the contour point is brighter than the outer point, so as to be able to identify the white candidate area as abnormal.

[0093] As an example, the process of determining that the difference between the first gray value and the second gray value reaches the preset difference can be specifically calculated by the following formula: gray inside -grayoutside ≥ D, where gray inside is the second gray level of the second point, gray outside is the first gray level of the first point, and D represents a preset difference. If a preset number of contour points all satisfy this formula, it indicates that the candidate area is an abnormal area.

[0094] Next, the case where the gray level value of the contour point indicated by the third comparison result is less than the gray level value of the target area will be described. When the gray level value of the contour point is less than the gray level value of the target area, it means that the position of the contour point is darker than the brightness of its surrounding area. For example, for a point on the contour of a candidate area that is black, it is necessary to accurately identify the black candidate area.

[0095] Since when the candidate area is black, the gray level value inside the candidate area will be smaller than the gray level value outside, that is, the second gray level of the second point is less than the first gray level of the first point. Therefore, it can be set that the preset difference is less than 0. If the first comparison result indicates that the difference between the second gray level and the first gray level is less than the preset difference, it means that the second gray level is less than the first gray level, that is, the inner point of the contour point is darker than the outer point, so that the black candidate area can be abnormally identified.

[0096] As an example, the process of determining that the difference between the first gray level and the second gray level reaches the preset difference can be specifically calculated by the following formula: gray inside - gray outside ≤ D, where gray inside is the second gray level of the second point, gray outside is the first gray level of the first point, and D represents a preset difference. If a preset number of contour points all satisfy this formula, it indicates that the candidate area is an abnormal area.

[0097] In summary, by selecting the corresponding formula based on the light and dark conditions of the candidate area for the identification of abnormal areas, the identification accuracy is improved, and misdetection caused by incorrect formula selection is avoided.

[0098] In a possible implementation manner, when S104 determines that the candidate area is an abnormal area when a preset number of first comparison results all indicate that the difference between the first gray level and the second gray level reaches the preset difference, specifically, when the ratio of the number of first comparison results indicating that the difference between the first gray level and the second gray level reaches the preset difference to the number of all contour points reaches the first ratio, S104 determines that the candidate area is an abnormal area.

[0099] Specifically, the first ratio can be a preset ratio value, such as 3 / 4. The number of first comparison results indicating that the difference between the first gray level and the second gray level reaches a preset difference can be counted, and the ratio of this number to the number of all contour points is calculated. If this ratio is not less than the first ratio, it means that a large number of contour points among all contour points satisfy the condition that the gray level difference reaches the preset difference, and then the candidate area can be determined as an abnormal area.

[0100] As an example, it can be expressed by the formula count≥M*count edge , where count is the number of first comparison results indicating that the difference between the first gray level and the second gray level reaches a preset difference, count edge is the number of all contour points, and M is the first ratio.

[0101] In summary, by counting the number of cases where the gray level difference between the second gray level and the first gray level satisfies the preset difference and calculating the ratio of this number to the total number of all contour points, it is possible to more accurately determine whether it is an abnormal area, avoiding the situation where the total number of all contour points is large and the number of contour points satisfying the difference condition is too small, resulting in inaccurate detection.

[0102] In a possible implementation, when the ratio of the number of first comparison results indicating that the difference between the first gray level and the second gray level reaches a preset difference to the number of all contour points reaches the first ratio, S1044 determines the candidate area as an abnormal area, which may specifically include S10441 - S10442.

[0103] S10441, determining that the candidate area is located in the edge area of the candidate image.

[0104] Specifically, it is determined that the candidate area is located in the edge area of the candidate image. If the candidate area is located in the edge area of the candidate image, then when determining the target point pairs of each contour point, the contour points located on the edge of the candidate image cannot collect target point pairs. Therefore, during subsequent abnormal detection, the corresponding difference satisfaction conditions need to be adjusted.

[0105] S10442, when the ratio of the number of first comparison results indicating that the difference between the first gray level and the second gray level reaches a preset difference to the number of all contour points reaches the second ratio, determining the candidate area as an abnormal area.

[0106] Specifically, when the candidate region is located in the edge region, a part of the contour of the candidate region is the contour of its own region shape, and the other part of the contour is the boundary contour of the candidate image, such as a part of the side length of a rectangular candidate image. Therefore, for the contour points located on the boundary contour of the candidate image, the outer point, i.e., the first point, cannot be determined, that is, the corresponding target point pair cannot be determined. Furthermore, subsequent differential comparison based on the gray values of the inner and outer point pairs cannot be performed. Therefore, the number of contour points whose gray difference satisfies the preset difference is less, and its proportion in the total number of all contour points is also lower.

[0107] Therefore, in order to accurately identify such candidate images for anomalies, a lower second ratio can be set, that is, the second ratio is less than the first ratio. For example, the second ratio can be 1 / 2, so that the abnormal region can more easily meet the condition, that is, the number of first comparison results indicating that the difference between the first gray value and the second gray value reaches the preset difference, and the ratio of it to the total number of all contour points can more easily reach the second ratio, so that it can be more accurately identified whether the candidate region located in the edge region is an abnormal region.

[0108] This application also provides an abnormal region screening device. Refer to Figure 6 As shown, it is a structural block diagram of an abnormal region screening device provided by an embodiment of this application. The device includes: An extraction unit 201, configured to perform contour point extraction processing on a candidate image to obtain a plurality of contour points; the plurality of contour points are used to identify the contour of the candidate region of the candidate image; A first determination unit 202, configured to determine a target point pair corresponding to the contour point based on the contour point; the target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the partial contour where the contour point is located; A comparison unit 203, configured to perform gray comparison between the first gray value of the first point and the second gray value of the second point to obtain a first comparison result of the target point pair; A second determination unit 204, configured to determine that the candidate region is an abnormal region when a preset number of first comparison results all indicate that the difference between the first gray value and the second gray value reaches the preset difference.

[0109] Optionally, the device further includes: A comparison unit, configured to compare the size parameter of the candidate region with a preset parameter to obtain a second comparison result; When the second comparison result indicates that the size parameter is greater than or equal to the preset parameter, the first determination unit is configured to: Determine the point located outside the partial contour where the contour point is located as the first point; Determine the point inside the partial contour where the contour point is located as the second point; the first point, the second point, and the contour point are on the same straight line.

[0110] Optionally, when the second comparison result indicates that the size parameter is less than the preset parameter, the first determination unit is configured to: Determine the point outside the partial contour where the contour point is located as the first point; Determine the center point of the candidate area as the second point.

[0111] Optionally, the first determination unit is configured to: Determine a first reference point and a second reference point on the contour based on the contour point; the first reference point and the second reference point are close to the contour point, and the contour point is located between the first reference point and the second reference point; Determine a second straight line passing through the contour point and perpendicular to the first straight line where the first reference point and the second reference point are located; Determine the point on the second straight line and outside the contour as the first point.

[0112] Optionally, the first determination unit is configured to: Enlarge the contour based on an enlargement parameter to obtain an enlarged contour of the candidate area; the enlarged contour has the same shape as the contour; Determine the intersection point closer to the contour point among at least one intersection point of the second straight line and the enlarged contour as the first point.

[0113] Optionally, there are multiple other contour points between the first reference point or the second reference point and the contour point.

[0114] Optionally, the device further includes: A comparison unit, configured to perform a grayscale comparison between the contour point and the target area where the contour point is located to obtain a third comparison result; When the third comparison result indicates that the grayscale of the contour point is greater than or equal to the grayscale of the target area, the second determination unit is configured to: When a preset number of the first comparison results all indicate that the difference between the second grayscale and the first grayscale is greater than the preset difference, determine the candidate area as the abnormal area; the preset difference is greater than 0; When the third comparison result indicates that the grayscale of the contour point is less than the grayscale of the target area, the second determination unit is configured to: When the difference between the second grayscale value and the first grayscale value is less than the preset difference as indicated by all of the first comparison results that meet the preset quantity, determine the candidate region as the abnormal region; the preset difference is less than 0.

[0115] Optionally, the second determination unit is configured to: When the ratio of the quantity of the first comparison results indicating that the difference between the first grayscale value and the second grayscale value reaches the preset difference to the quantity of all contour points reaches a first ratio, determine the candidate region as the abnormal region.

[0116] Optionally, the second determination unit is configured to: Determine that the candidate region is located in the edge region of the candidate image; When the ratio of the quantity of the first comparison results indicating that the difference between the first grayscale value and the second grayscale value reaches the preset difference to the quantity of all contour points reaches a second ratio, determine the candidate region as the abnormal region; the second ratio is less than the first ratio.

[0117] Optionally, the apparatus further includes: A detection unit, configured to perform defect detection on a semiconductor structure to obtain the candidate image of the semiconductor structure; the candidate image includes the candidate region, and the candidate region is the region where a real defect or a false defect is located; The second determination unit is configured to: When all of the first comparison results that meet the preset quantity indicate that the difference between the first grayscale value and the second grayscale value reaches the preset difference, determine the candidate region as the region where the real defect is located.

[0118] Optionally, the extraction unit is configured to: Perform image enhancement processing on the candidate image to obtain an enhanced image of the candidate image; Perform contour point extraction processing on the enhanced image to obtain the plurality of contour points; the plurality of contour points are used to identify the contour of the candidate region of the enhanced image.

[0119] In another aspect, an embodiment of the present application provides a computer device. Refer to Figure 7 As shown, it is a structural diagram of a computer device provided by an embodiment of the present application. The computer device includes a processor 310 and a memory 320: The memory 320 is configured to store program codes and transmit the program codes to the processor 310; The processor 310 is configured to execute the method provided by the above embodiment according to the instructions in the program codes.

[0120] The computer device may include a terminal device or a server, and the foregoing apparatus may be configured in the computer device.

[0121] In another aspect, an embodiment of the present application further provides a storage medium for storing a computer program for executing the method provided in the foregoing embodiment.

[0122] Those of ordinary skill in the art can understand that all or part of the steps of implementing the foregoing method embodiments can be completed by program instructions and hardware. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the foregoing method embodiments; and the foregoing storage medium can be at least one of the following media: read-only memory (English: Read-only Memory, abbreviation: ROM), RAM, magnetic disk, or optical disk, etc., various media that can store program codes.

[0123] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the partial description of the method embodiments for the relevant parts.

[0124] The foregoing is only the preferred embodiment of the present application. Although the present application has been disclosed above with preferred embodiments, it is not intended to limit the present application. Any person skilled in the art can make many possible changes and modifications to the technical solution of the present application, or modify it into an equivalent embodiment with equivalent changes, without departing from the scope of the technical solution of the present application. Therefore, any simple modification, equivalent change, and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application still fall within the scope of the protection of the technical solution of the present application.

Claims

1. A method for screening abnormal areas, characterized in that: include: Perform contour point extraction processing on the selected image to obtain multiple contour points; The plurality of contour points are used to identify the contour of the to-be-selected area of ​​the to-be-selected image; Determine a target point pair corresponding to the contour point based on the contour point; the target point pair includes a first point and a second point, the first point and the second point are respectively located on both sides of the contour of the portion where the contour point is located; Performing a grayscale comparison on the first grayscale of the first point and the second grayscale of the second point to obtain a first comparison result of the target point pair; When a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, the to-be-selected region is determined to be an abnormal region.

2. The method according to claim 1, characterized in that Before determining the target point pair corresponding to the contour point based on the contour point, the method further includes: Comparing the size parameter of the to-be-selected area with the preset parameter to obtain a second comparison result; When the second comparison result indicates that the size parameter is greater than or equal to the preset parameter, determining the target point pair corresponding to the contour point based on the contour point includes: Determine a point located outside the portion of the contour where the contour point is located as the first point; A point located inside the portion of the contour where the contour point is located is determined as the second point; the first point, the second point and the contour point are located on the same straight line.

3. The method according to claim 2, characterized in that When the second comparison result indicates that the size parameter is less than the preset parameter, determining the target point pair corresponding to the contour point based on the contour point includes: Determine a point located outside the portion of the contour where the contour point is located as the first point; The center point of the to-be-selected area is determined as the second point.

4. The method according to claim 2 or 3, characterized in that: The step of determining a point located outside the portion of the contour where the contour point is located as the first point comprises: Determine a first reference point and a second reference point on the contour based on the contour point; the first reference point and the second reference point are close to the contour point, and the contour point is located between the first reference point and the second reference point; Determine a second straight line passing through the contour point and perpendicular to the first straight line where the first reference point and the second reference point are located; A point located on the second straight line and outside the contour is determined as the first point.

5. The method according to claim 4, characterized in that The step of determining a point located on the second straight line and outside the contour as the first point comprises: The outline is enlarged based on the enlargement parameter to obtain an enlarged outline of the selected area; the enlarged outline is consistent with the shape of the outline; Among at least one intersection point of the second straight line and the expanded outline, an intersection point close to the outline point is determined as the first point.

6. The method according to claim 4, characterized in that There are a plurality of other contour points between the first reference point or the second reference point and the contour point.

7. The method according to claim 1, characterized in that Before determining that the selected region is an abnormal region when a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, the method further includes: Performing a grayscale comparison between the contour point and the target area where the contour point is located to obtain a third comparison result; When the third comparison result indicates that the grayscale of the contour point is greater than or equal to the grayscale of the target area, when a preset number of the first comparison results indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, determining that the selected area is an abnormal area includes: When a preset number of the first comparison results all indicate that the difference between the second grayscale and the first grayscale is greater than the preset difference, determining that the selected area is the abnormal area; and the preset difference is greater than 0; When the third comparison result indicates that the grayscale of the contour point is less than the grayscale of the target area, when a preset number of the first comparison results indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, determining that the selected area is an abnormal area includes: When a preset number of the first comparison results all indicate that the difference between the second grayscale and the first grayscale is less than the preset difference, the to-be-selected region is determined to be the abnormal region; and the preset difference is less than 0.

8. The method according to claim 1, characterized in that When a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, determining that the selected area is an abnormal area includes: When the ratio of the number of first comparison results indicating that the difference between the first grayscale and the second grayscale reaches the preset difference to the number of all contour points reaches a first ratio, the to-be-selected region is determined to be the abnormal region.

9. The method according to claim 8, characterized in that When the ratio of the number of first comparison results indicating that the difference between the first grayscale and the second grayscale reaches the preset difference to the number of all contour points reaches a first ratio, determining that the selected area is the abnormal area includes: Determining that the to-be-selected area is located in an edge area of ​​the to-be-selected image; When the ratio of the number of first comparison results indicating that the difference between the first grayscale and the second grayscale reaches the preset difference to the number of all contour points reaches a second ratio, the selected area is determined to be the abnormal area; and the second ratio is smaller than the first ratio.

10. The method according to claim 1, characterized in that Before performing contour point extraction processing on the selected image to obtain a plurality of contour points, the method further includes: Performing defect detection on the semiconductor structure to obtain the image to be selected of the semiconductor structure; the image to be selected includes the area to be selected, and the area to be selected is the area where the real defect or the false defect is located; When a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, determining that the selected area is an abnormal area includes: When a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference, the to-be-selected region is determined to be the region where the real defect is located.

11. The method according to claim 1, characterized in that: The contour point extraction process is performed on the selected image to obtain a plurality of contour points, including: Performing image enhancement processing on the image to be selected to obtain an enhanced image of the image to be selected; The enhanced image is subjected to contour point extraction processing to obtain the plurality of contour points; the plurality of contour points are used to identify the contour of the to-be-selected area of ​​the enhanced image.

12. An abnormal area screening device, characterized in that: include: An extraction unit is used to perform contour point extraction processing on the selected image to obtain multiple contour points; The plurality of contour points are used to identify the contour of the to-be-selected area of ​​the to-be-selected image; A first determining unit is used to determine a target point pair corresponding to the contour point based on the contour point; the target point pair includes a first point and a second point, and the first point and the second point are respectively located on both sides of the contour of the portion where the contour point is located; A comparison unit, configured to perform a grayscale comparison on a first grayscale of the first point and a second grayscale of the second point to obtain a first comparison result of the target point pair; The second determining unit is configured to determine that the to-be-selected region is an abnormal region when a preset number of the first comparison results all indicate that the difference between the first grayscale and the second grayscale reaches a preset difference.

13. A computer device, characterized in that: The computer device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 11 according to the instructions in the program code.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method according to any one of claims 1 to 11.

15. A computer program product comprising a computer program, which, when executed on a computer device, causes the computer device to execute the method according to any one of claims 1 to 11.

Images