Wafer image processing method and device, electronic equipment and storage medium

Abstract

The invention provides a wafer image processing method and device, electronic equipment and a storage medium. The wafer image processing method comprises the following steps: determining a plurality of target areas in a wafer image comprising a wafer and first area categories corresponding to the plurality of target areas respectively; for each target area in the plurality of target areas, based on the position of the target area in the wafer image, determining a second area category corresponding to the target area in the first wafer map of the wafer; and for each target region in the plurality of target regions, determining an anomaly detection result of the target region based on the first region category corresponding to the target region and the second region category corresponding to the target region. According to the scheme, the efficiency of performing anomaly detection on the sorted wafers can be improved, and the labor cost can be reduced. Especially, under the condition that residual bare chips and empty areas exist in sorted wafers, manual proofreading is extremely tedious. In addition, the accuracy of an anomaly detection result is improved.

Description

Technical Field

[0001] This invention relates to the field of wafer inspection, and more specifically to a wafer image processing method, a wafer image processing apparatus, an electronic device, a storage medium, and a computer program product. Background Technology

[0002] Wafer surface defect inspection is a critical process for ensuring the quality of bare dies within a wafer. With the continuous advancement of semiconductor technology, the efficiency requirements for wafer inspection are also increasing.

[0003] Typically, defects can be determined manually by comparing physical images and spectrograms of sorted wafers. However, manual inspection is inefficient and inaccurate. Therefore, how to better detect anomalies in sorted wafers is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The present invention was proposed in view of the above-mentioned problems. The present invention provides a method for processing wafer images, a device for processing wafer images, an electronic device, a storage medium, and a computer program product.

[0005] According to one aspect of the present invention, a method for processing a wafer image is provided. The method includes: determining a plurality of target regions in a wafer image including a wafer and a first region category corresponding to each of the target regions, wherein the wafer includes a plurality of residual dies and a plurality of empty areas formed by removing the dies, and the first region category corresponding to a target region is used to indicate that the region corresponding to the target region in the wafer image is either the region corresponding to the residual die or the region corresponding to the empty area; for each target region among the plurality of target regions, based on the position of the target region in the wafer image, determining a second region category corresponding to the target region in a first wafer pattern of the wafer, wherein the second region category is used to indicate that the region corresponding to the target region in the first wafer pattern is either the region corresponding to the residual die or the region corresponding to the empty area; and determining an anomaly detection result for the target region based on the first region category and the second region category corresponding to the target region, the anomaly detection result indicating whether an anomaly exists in the target region.

[0006] For example, the above processing method further includes: Acquire a first image including the wafer; Determine the orientation mark region in the first image, wherein the orientation mark region is the region in the first image corresponding to the orientation mark of the wafer; Based on the relative positional relationship between the orientation mark area and the vertical axis, the first rotation angle of the wafer is determined, wherein the orientation mark of the wafer faces downwards towards the vertical axis after the wafer rotates by the first rotation angle, and the wafer image is an image acquired for the wafer rotated based on the first rotation angle.

[0007] For example, based on the first region category corresponding to the target region and the second region category corresponding to the target region, the anomaly detection result of the target region is determined, including at least one of the following: If the first region category corresponding to the target region is the same as the region represented by the second region category corresponding to the target region, the anomaly detection result of the target region is determined to indicate that no anomaly exists; If the first region category corresponding to the target region is different from the second region category corresponding to the target region, the anomaly detection result of the target region is used to indicate the presence of an anomaly.

[0008] For example, based on the first region category corresponding to the target region and the second region category corresponding to the target region, the anomaly detection result of the target region is determined, including: The first data and the second data corresponding to the target region are determined. The first region category corresponding to the target region is used to indicate that when the region corresponding to the target region in the wafer image is the region corresponding to the residual die, the first data corresponding to the target region is a first natural number. The first region category corresponding to the target region is used to indicate that when the region corresponding to the target region in the wafer image is the region corresponding to the empty region, the first data corresponding to the target region is a second natural number, and the sum of the second natural number and the first natural number is odd. The second region category corresponding to the target region is used to indicate that when the region corresponding to the target region in the first wafer image is the region corresponding to the residual die, the second data corresponding to the target region is a third natural number, and the sum of the third natural number and the first natural number is even. The second region category corresponding to the target region is used to indicate that when the region corresponding to the target region in the first wafer image is the region corresponding to the empty region, the second data corresponding to the target region is a fourth natural number, and the sum of the fourth natural number and the second natural number is even. If the sum of the first data corresponding to the target area and the second data corresponding to the target area is odd, the anomaly detection result of the target area is used to indicate the presence of an anomaly. If the sum of the first data corresponding to the target area and the second data corresponding to the target area is even, the anomaly detection result of the target area is used to indicate that no anomaly exists.

[0009] For example, the above processing method further includes: A second wafer map is determined based on multiple target regions in a wafer image including the wafer and the first region category corresponding to each of the multiple target regions; Based on the second wafer pattern, the matching result between the wafer and the first wafer pattern is determined.

[0010] For example, determining the matching result between the wafer and the first wafer pattern based on the second wafer pattern includes: Based on the region corresponding to the wafer in the second wafer map and the preset size corresponding to each image block in the second wafer map, the predicted size of the wafer is determined; Based on the preset size corresponding to the first wafer pattern and the predicted size of the wafer, the matching result between the wafer and the first wafer pattern is determined.

[0011] For example, the anomaly detection result of the target region includes anomaly result information indicating whether an anomaly exists, and, when the anomaly result information indicates that an anomaly exists in the target region, position information of the target region in at least one of the first wafer map, the second wafer map, and the wafer image, wherein the second wafer map is determined based on multiple target regions in the wafer image including the wafer and the first region category corresponding to each of the multiple target regions.

[0012] For example, for each of multiple target regions, where the anomaly detection result of that target region is used to indicate the presence of an anomaly, the method further includes: If it is determined that the target area corresponds to a defective die area in the first wafer map, a first prompt message is generated. The defective die area is the area corresponding to the defective residual die in the first wafer map. The first prompt message is used to indicate that the target area is a defective die area and that there is an anomaly. If the target area is determined to be a functional die area in the first wafer map, a second prompt message is generated. The functional die area is the area in the first wafer map corresponding to the residual die other than the defective die. The second prompt message is used to indicate that the target area is a functional die area and that there is an anomaly. If the target region is determined to be an empty region in the first wafer pattern, a third prompt message is generated. The empty region is the region corresponding to the empty region in the first wafer pattern. The third prompt message is used to indicate that the target region is an empty region and that there is an anomaly.

[0013] For example, the above processing method further includes: Based on the anomaly detection results of multiple target regions, the anomaly detection results of the wafer are determined.

[0014] According to another aspect of the present invention, a wafer image processing apparatus is also provided, the processing apparatus comprising a first region category determination module, a second region category determination module, and a region detection result determination module. The first region category determination module is used to determine multiple target regions in a wafer image including a wafer and a first region category corresponding to each of the multiple target regions, wherein the wafer includes multiple residual dies and multiple empty areas formed by removing the dies, and the first region category corresponding to the target region is used to indicate that the region corresponding to the target region in the wafer image is either the region corresponding to the residual die or the region corresponding to the empty area; the second region category determination module is used to, for each of the multiple target regions, determine a second region category corresponding to the target region in a first wafer map of the wafer based on the position of the target region in the wafer image, wherein the second region category is used to indicate that the region corresponding to the target region in the first wafer map is either the region corresponding to the residual die or the region corresponding to the empty area; the region detection result determination module is used to, for each of the multiple target regions, determine an anomaly detection result for the target region based on the first region category and the second region category corresponding to the target region, wherein the anomaly detection result is used to indicate whether an anomaly exists in the target region.

[0015] According to another aspect of the present invention, an electronic device is also provided, the electronic device including a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the above-described wafer image processing method.

[0016] According to another aspect of the present invention, a storage medium is also provided, which stores computer program instructions that, when executed, are used to perform the above-described wafer image processing method.

[0017] According to another aspect of the present invention, a computer program product is also provided, the computer program product comprising computer program instructions which, when run by a processor, are used to execute the above-described wafer image processing method.

[0018] According to the above-described scheme of the present invention, multiple target regions in a wafer image including a wafer and a first region category corresponding to each of the multiple target regions can be determined. Then, for each target region among the multiple target regions, a second region category corresponding to the target region in a first wafer map of the wafer is determined based on the position of the target region in the wafer image. Finally, for each target region among the multiple target regions, an anomaly detection result for the target region is determined based on the first region category and the second region category corresponding to the target region. The above scheme can improve the efficiency of anomaly detection on sorted wafers by determining the anomaly detection result of the target region through the first region category and the second region category corresponding to the target region, which is beneficial to reducing labor costs. Especially when there are residual dies and empty areas in the sorted wafers, manual verification is extremely tedious. In addition, compared with the manual verification scheme, the above-described scheme of this application is not affected by the experience of manual verification, which is beneficial to improving the accuracy of anomaly detection results. Attached Figure Description

[0019] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0020] Figure 1 A schematic flowchart of a wafer image processing method according to an embodiment of the present invention is shown; Figure 2 A schematic block diagram of a wafer image processing apparatus according to an embodiment of the present invention is shown; Figure 3 A schematic block diagram of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0022] To at least partially solve the above problems, embodiments of the present invention provide a method for processing wafer images. Figure 1A schematic flowchart illustrating a wafer image processing method according to an embodiment of the present invention is shown. Figure 1 As shown, the processing method may include steps S110 to S130.

[0023] In step S110, multiple target regions in a wafer image including the wafer and a first region category corresponding to each of the multiple target regions are determined.

[0024] An image acquisition device can be used to acquire images of the wafer to obtain a wafer image. The wafer includes multiple residual dies and multiple empty areas formed by the removal of the dies.

[0025] Residual dies in a wafer may include dies that fail to meet electrical performance standards, dies that meet electrical performance standards and need to be retained, and dies that meet electrical performance standards but were not removed due to incomplete dicing. Empty areas in a wafer include empty areas formed by the accidental removal of dies that meet electrical performance standards, empty areas formed by the accidental removal of dies that fail to meet electrical performance standards, and empty areas formed by the accidental removal of dies that meet electrical performance standards and need to be retained.

[0026] The specific size of the wafer is not limited in this embodiment of the invention. In a specific example, the wafer can be a 6-inch to 12-inch wafer.

[0027] In some embodiments, the wafer image can be an image containing the entire wafer obtained from a single imaging session. In some alternative embodiments, if the field of view of a single imaging session is limited and it is not possible to obtain an image containing the entire wafer from a single imaging session, an automatic displacement stage can be used to photograph the wafer in sections to obtain multiple partitioned images belonging to different parts of the wafer. These partitioned images can then be stitched together to obtain a wafer image containing the entire wafer. It is understood that the wafer image may include the wafer region corresponding to the wafer. The wafer region can be used to represent the wafer's position information within the wafer image.

[0028] In one example, the target region can be determined by the user tracing a path. For instance, a wafer image can be displayed on a screen. The user can then use an input device (such as a mouse or keyboard) to trace multiple closed regions in the wafer image as target regions. In another example, the target region can be automatically determined using a trained target region detection model or algorithm. In a specific example, the target region can represent the area in the wafer image corresponding to a residual die or an empty area formed by the removal of a die.

[0029] Optionally, wafer regions within a wafer image can be identified, and then target regions can be determined based on these wafer regions. For example, wafer regions within a wafer image can be determined using a trained wafer region detection model. Specifically, the wafer image can be directly input into the trained wafer region detection model to obtain the wafer regions. Another example is that the wafer image can be preprocessed (e.g., converted to grayscale, feature enhancement, etc.) and then input into the trained wafer region detection model to obtain the wafer regions. Yet another example is that the wafer image can be displayed on a screen. Then, the user can trace a closed region in the wafer image using an input device to represent the wafer region. Combining this with the example of the user tracing to determine the target region, the wafer region can be marked on the displayed wafer image for the user's tracing reference. Combining this with the example of determining the target region using a trained target region detection model or algorithm, the image corresponding to the wafer region in the wafer image can be used as input to the trained target region detection model or algorithm to obtain a more accurate target region. The trained wafer region detection model described above can be trained based on multiple first training images, including a training wafer (which may be a sorted wafer used for training, and different from the wafer in step S110; different first training images may include different training wafers), and the wafer annotation region corresponding to the training wafer in each first training image. The wafer annotation region can be used to indicate the position of the training wafer in the first training image. Similarly, the trained target region detection model described above can be trained based on multiple second training images, including a training wafer (which may be a sorted wafer used for training, and different from the wafer in step S110; different second training images may include different training wafers, and may be different from the training wafer in the first training image), and the target annotation regions corresponding to the residual die or the empty area formed by the removal of the die in each second training image. The target annotation region can be used to indicate the position of the residual die or the empty area formed by the removal of the die in the second training image. The specific training process and model structure of the wafer region detection model and the target region detection model are not limited in this embodiment of the invention and can be determined according to the actual needs of the developers.

[0030] The first region category corresponding to the target region is used to indicate whether the region corresponding to the target region in the wafer image is the region corresponding to the residual die or the region corresponding to the empty area.

[0031] The first region category can be automatically determined using a trained region category detection model or a region category detection algorithm. For example, the first region category of a target region can be determined by identifying the image grayscale of each target region in a wafer image. Specifically, the second region category of the target region is either an empty area or a residual die. The trained region category detection model can be trained based on multiple third training images that include residual dies or empty areas formed by the removal of dies, and the training region category corresponding to each third training image. The training region category corresponding to the third training image can be used to represent that the third training image includes residual dies or empty areas formed by the removal of dies. The specific training process and specific model structure of the region category detection model are not limited in this embodiment of the invention and can be determined according to the actual needs of the developers.

[0032] In step S120, for each of the multiple target regions, based on the position of the target region in the wafer image, the second region category corresponding to the target region in the first wafer map of the wafer is determined.

[0033] The first wafer map can be used to visually represent the location of each remaining die or the empty area formed by each die being removed within the wafer. In one example, the first wafer map can be used to guide wafer sorting and update the second region category. The wafer map in this embodiment of the invention is also called a wafer map.

[0034] The second region category corresponding to the target region in the first wafer map is used to indicate whether the region corresponding to the target region in the first wafer map is the region corresponding to the residual die or the region corresponding to the empty region.

[0035] In the first wafer map, the regions corresponding to residual dies may include regions corresponding to dies with substandard electrical performance and regions corresponding to dies with satisfactory electrical performance that need to be retained. In the first wafer map, the regions corresponding to empty areas are those formed when electrically satisfactory dies are removed.

[0036] In some embodiments, for each of the multiple target regions, the position of the target region in the first wafer map can be determined based on its position in the wafer image. For example, a correspondence between the position in the wafer image and the position in the first wafer map can be established in advance, and then the corresponding position can be determined using this correspondence. If the corresponding position is located in the region corresponding to an empty area in the first wafer map, then the second region category of the target region is used to indicate that the region corresponding to the target region in the first wafer map is the region corresponding to an empty area (hereinafter abbreviated as the second region category of the target region is the empty area category). If the corresponding position is located in the region corresponding to a residual die in the first wafer map, then the second region category of the target region is used to indicate that the region corresponding to the target region in the first wafer map is the region corresponding to a residual die (hereinafter abbreviated as the second region category of the target region is the residual die category).

[0037] In some other embodiments, multiple partitioned regions can be defined in the first wafer map, such that a target region in the wafer image corresponds one-to-one with a partitioned region in the first wafer map based on their positions in the image and map. For each target region among the multiple target regions, a second region category corresponding to that target region in the first wafer map is determined based on the partitioned region corresponding to that target region in the first wafer map. For example, the second region category corresponding to the partitioned region can be used as the second region category corresponding to that target region in the first wafer map.

[0038] In step S130, for each target region among multiple target regions, the anomaly detection result of the target region is determined based on the first region category and the second region category corresponding to the target region.

[0039] Anomaly detection results are used to indicate whether anomalies exist in the target area.

[0040] It is understandable that if the first region category corresponding to the target region is the residual die category and the second region category corresponding to the target region is the empty region category, then the target region may correspond to a die with qualified electrical properties that was not removed due to incomplete cutting in the wafer, that is, the position of the target region in the wafer is abnormal.

[0041] Similarly, based on whether the first region category and the second region category corresponding to the target region represent the same or different regions, the anomaly detection result for the target region can be determined. For example, if they are the same, it can be determined that the anomaly detection result for the target region indicates that there is no anomaly in the target region; if they are different, it can be determined that the anomaly detection result for the target region indicates that there is anomaly in the target region. As another example, if they are the same, it can be determined that the anomaly detection result for the target region indicates that there is no anomaly in the target region; if they are different, the first region category and / or the corresponding second region category for the target region can be redefined to reduce the probability of false anomaly detection results.

[0042] In some embodiments, the anomaly detection result for the target region includes whether an anomaly exists. In some alternative embodiments, the anomaly detection result for the target region also includes the location information of the target region. In other alternative embodiments, the anomaly detection result for the target region also includes a first region category and a second region category corresponding to the target region.

[0043] According to the above-described scheme of the present invention, multiple target regions in a wafer image including a wafer and a first region category corresponding to each of the multiple target regions can be determined. Then, for each target region among the multiple target regions, a second region category corresponding to the target region in a first wafer map of the wafer is determined based on the position of the target region in the wafer image. Finally, for each target region among the multiple target regions, an anomaly detection result for the target region is determined based on the first region category and the second region category corresponding to the target region. The above scheme can improve the efficiency of anomaly detection on sorted wafers by determining the anomaly detection result of the target region through the first region category and the second region category corresponding to the target region, which is beneficial to reducing labor costs. Especially when there are residual dies and empty areas in the sorted wafers, manual verification is extremely tedious. In addition, compared with the manual verification scheme, the above-described scheme of this application is not affected by the experience of manual verification, which is beneficial to improving the accuracy of anomaly detection results.

[0044] For example, the above processing method further includes steps S210 to S230.

[0045] In step S210, a first image including the wafer is obtained.

[0046] An image acquisition device can be used to acquire images of the wafer to obtain a first image including the wafer. The first image may include the sorted wafer. If the field of view of a single imaging is limited, multiple partition images can be acquired and then stitched together. For details, please refer to the relevant content in step S110. This embodiment of the invention will not be described in detail here. It can be understood that the first image of the wafer may also include the wafer region corresponding to the wafer.

[0047] In step S220, the directional marker region in the first image is determined.

[0048] The orientation marking area is the region in the first image corresponding to the orientation marking on the wafer. The orientation marking can be a notch (a groove on a wafer used for orientation) or a flat edge (or flat edge) of the wafer.

[0049] In some embodiments, a trained orientation marker detection model can be used to identify orientation markers in a first image to determine orientation marker regions. The trained orientation marker detection model can be trained based on multiple fourth training images, including a training wafer (which may be a sorted wafer used for training, and different from the wafer in step S110; the training wafers included in different fourth training images may be different, and may be different from the training wafers in the first or second training images), and the corresponding orientation marker training regions in each fourth training image. The orientation marker training regions can be the regions in the fourth training images corresponding to the orientation markers of the training wafers. The first image can be input into the trained orientation marker detection model to obtain the orientation marker regions in the first image.

[0050] In some other embodiments, the directional marking area can be determined by a user tracing a line. For example, a first image can be displayed on a screen. Then, the user can trace a closed area in the first image using an input device (e.g., a mouse or keyboard) to serve as the directional marking area.

[0051] In some other embodiments, a matching directional marker region can also be determined in the first image by using a reference template of the directional marker region.

[0052] If the orientation mark area is not identified in the first image or the identified orientation mark is incomplete, a prompt message can be generated, and the orientation mark area in the first image can be re-identified after the user manually or the robotic arm automatically adjusts or replaces the wafer.

[0053] In step S230, the first rotation angle of the wafer is determined based on the relative positional relationship between the orientation marking area and the vertical axis.

[0054] The orientation mark on the wafer is oriented downwards towards the vertical axis after the wafer has rotated by the first rotation angle, and the wafer image is an image acquired for the wafer rotated based on the first rotation angle.

[0055] It is understandable that the first rotation angle can also have a directional sign. The specific adjustment angle of the wafer can be determined by the angle value of the first rotation angle, and the specific adjustment direction of the wafer can be determined by the directional sign of the first rotation angle, thus enabling preliminary adjustment of the wafer. The first wafer map can be the map corresponding to the adjusted wafer, and thus the orientation of the wafer in the wafer image can be adapted to the orientation of the wafer in the first wafer map.

[0056] In some embodiments, when the orientation mark of the wafer is a flat edge, the first image can be adjusted until the flat edge is perpendicular to the vertical axis in the first image and the orientation mark area corresponding to the flat edge is at the bottom of the wafer area, and the rotation angle used to adjust the first image is determined as the first rotation angle.

[0057] In some other embodiments, when the wafer's orientation marker is a Notch, the first image can be adjusted until the Notch is oriented downwards towards the vertical axis in the first image, and the rotation angle used to adjust the first image is determined as the first rotation angle. Considering practical scenarios, the Notch is typically located at the edge of the wafer; therefore, the first image can also be adjusted until the orientation marker area corresponding to the Notch is at the bottom of the wafer area, and the rotation angle used to adjust the first image is determined as the first rotation angle.

[0058] After rotating the wafer according to the first rotation angle, a wafer image can be obtained.

[0059] According to the above-described scheme of the present invention, a first image including a wafer can be acquired. Then, an orientation mark region in the first image is determined. Finally, based on the relative positional relationship between the orientation mark region and the vertical axis, a first rotation angle of the wafer is determined. The above scheme, by adjusting the orientation of the wafer through the first rotation angle, can obtain a wafer image more adapted to the first wafer pattern, which is beneficial for improving the accuracy of anomaly detection results.

[0060] For example, step S130, which determines the anomaly detection result of the target region based on the first region category and the second region category corresponding to the target region, includes step S131a.

[0061] In step S131a, if the first region category corresponding to the target region is the same as the region represented by the second region category corresponding to the target region, the anomaly detection result of the target region is determined to indicate that no anomaly exists.

[0062] If the first region category corresponding to the target region and the second region category corresponding to the target region represent the same region, it can be determined that there is no anomaly at the corresponding location of the target region on the wafer. For example, if the first region category corresponding to the target region represents the region corresponding to the target region in the wafer image as the region corresponding to the residual die, and the second region category corresponding to the target region represents the region corresponding to the target region in the first wafer atlas as the region corresponding to the residual die, then the regions represented can be considered to be the same. In one example, the anomaly detection result for the target region can be the text information "No anomaly". In another example, the anomaly detection result for the target region can include the marker information of the target region in the first wafer atlas or wafer image, which indicates that there is no anomaly at the corresponding location of the target region. In yet another example, the anomaly detection result for the target region can be empty. It can be understood that if the anomaly detection result is not empty when used to indicate the presence of an anomaly, then an empty anomaly detection result can be used to indicate that there is no anomaly.

[0063] According to the above-described solution of the present invention, when the first region category corresponding to the target region and the second region category corresponding to the target region represent the same category, the anomaly detection result of the target region can be determined to indicate that no anomaly exists. Compared with the solution based on manual verification, the above-described solution of this application is not affected by the experience of manual verification, which is beneficial to improving the accuracy of the anomaly detection result used to indicate that no anomaly exists.

[0064] For example, step S130, which determines the anomaly detection result of the target region based on the first region category and the second region category corresponding to the target region, includes step S131b.

[0065] In step S131b, if the first region category corresponding to the target region is different from the region represented by the second region category corresponding to the target region, the anomaly detection result of the target region is determined to indicate the existence of an anomaly.

[0066] If the first region category corresponding to the target region and the second region category corresponding to the target region represent different regions, it can be determined that there is an anomaly at the corresponding location of the target region on the wafer. For example, if the first region category corresponding to the target region represents the region corresponding to the residual die in the wafer image, and the second region category corresponding to the target region represents the region corresponding to the empty area in the first wafer map, then the regions represented can be considered to be different. In one example, the anomaly detection result for the target region can be the text message "Anomaly exists". In another example, the anomaly detection result for the target region can also include the marker information of the target region in the first wafer map or wafer image, which is used to indicate that there is an anomaly at the corresponding location of the target region. In yet another example, the anomaly detection result for the target region may not be empty. It can be understood that if the anomaly detection result is empty when used to indicate that there is no anomaly, then a non-empty anomaly detection result can be used to indicate that there is an anomaly.

[0067] According to the above-described solution of the present invention, when the categories represented by the first region category corresponding to the target region and the second region category corresponding to the target region are different, the anomaly detection result of the target region can be determined to indicate the presence of anomalies. Compared with the solution based on manual verification, the above-described solution of this application is not affected by the experience of manual verification, which is beneficial to improving the accuracy of the anomaly detection result used to indicate the presence of anomalies.

[0068] For example, step S130, which determines the anomaly detection result of the target region based on the first region category and the second region category corresponding to the target region, includes steps S131c to S133c.

[0069] In step S131c, the first data corresponding to the target area and the second data corresponding to the target area are determined.

[0070] The first region category corresponding to the target region indicates that, in the case where the target region corresponds to a residual die in the wafer image, the first data corresponding to the target region is a first natural number. The first region category corresponding to the target region indicates that, in the case where the target region corresponds to an empty region in the wafer image, the first data corresponding to the target region is a second natural number. The sum of the second and first natural numbers is odd. The second region category corresponding to the target region indicates that, in the case where the target region corresponds to a residual die in the first wafer image, the second data corresponding to the target region is a third natural number. The sum of the third and first natural numbers is even. The second region category corresponding to the target region indicates that, in the case where the target region corresponds to an empty region in the first wafer image, the second data corresponding to the target region is a fourth natural number. The sum of the fourth and second natural numbers is even.

[0071] In some embodiments, the first natural number is odd, and it is understood that the third natural number is also odd, while the second and fourth natural numbers are both even. In yet other embodiments, the first natural number is even, and it is understood that the third natural number is also even, while the second and fourth natural numbers are both odd.

[0072] In some embodiments, the first and third natural numbers may be equal, and the second and fourth natural numbers may be equal. Specifically, for example, the first and third natural numbers may both be 1, and the second and fourth natural numbers may both be 0.

[0073] Based on the first region category and the second region category of the target region, the first data and the second data corresponding to the target region can be determined.

[0074] In step S132c, if the sum of the first data corresponding to the target area and the second data corresponding to the target area is odd, the anomaly detection result of the target area is determined to indicate the presence of an anomaly.

[0075] It is understandable that if the sum of the first data and the second data corresponding to the target area is odd, the first area category and the second area category of the target area are different, which indicates that the target area is abnormal.

[0076] In step S133c, if the sum of the first data corresponding to the target area and the second data corresponding to the target area is even, the anomaly detection result of the target area is determined to indicate that no anomaly exists.

[0077] It is understandable that if the sum of the first data and the second data corresponding to the target area is even, the first area category and the second area category of the target area are the same, and it can be determined that there is no anomaly in the target area.

[0078] According to the above-described scheme of the present invention, a first data point corresponding to the target region and a second data point corresponding to the target region can be determined. Then, if the sum of the first data point and the second data point corresponding to the target region is odd, an anomaly detection result for the target region is determined to indicate the presence of an anomaly. If the sum of the first data point and the second data point corresponding to the target region is even, an anomaly detection result for the target region is determined to indicate the absence of an anomaly. The above scheme, by converting the first region category and the second region category of the target region into one-dimensional information, can more efficiently determine the anomaly detection result of the target region.

[0079] For example, the above processing method further includes steps S210 and S220.

[0080] In step S210, a second wafer map is determined based on multiple target regions in a wafer image including the wafer and the first region category corresponding to each of the multiple target regions.

[0081] The location of each target region in the second wafer atlas can be determined based on its position in the wafer image. Target regions of the same first region category are represented by the same type of image block, while target regions of different first region categories are represented by different types of image blocks. This yields the second wafer atlas. It can be understood that one target region corresponds to one image block. It can also be understood that an image block can correspond to a non-wafer portion. Specifically, for example, the position information of the target region in the wafer image can be used to represent the number of rows and columns corresponding to the target region in the wafer image. Similarly, the position information of the target region in the second wafer atlas can be used to represent the number of rows and columns of the image block corresponding to the target region in the second wafer atlas. The number of rows and columns corresponding to the target region in the wafer image can be directly used as the number of rows and columns of the image block corresponding to the target region in the second wafer atlas. For example, if the total number of target regions in each row is 4 and the total number of target regions in each column is 3, then there can be a total of 4*3 target regions. The target region located in the first row and first column of the wafer image corresponds to the image block located in the first row and first column of the second wafer image. It is understood that the row and column numbers of the image block corresponding to non-wafer portions are not considered here; if they were, an offset value could be added.

[0082] Step S220: Based on the second wafer pattern, determine the matching result between the wafer and the first wafer pattern.

[0083] In some embodiments, if the image block sizes in the first wafer pattern and the image blocks in the second wafer pattern are the same, the overall size of the first wafer pattern and the overall size of the second wafer pattern can be compared to determine the matching result between the wafer and the first wafer pattern. For example, if the overall sizes are the same, the matching result can be determined as a match; if the overall sizes are different, the matching result can be determined as a mismatch. In still other embodiments, the matching result between the wafer and the first wafer pattern can be determined based on the total number of image blocks in the first wafer pattern and the total number of image blocks in the second wafer pattern. For example, if the total number of horizontal image blocks in the first wafer pattern is equal to the total number of horizontal image blocks in the second wafer pattern, and the total number of vertical image blocks in the first wafer pattern is equal to the total number of vertical image blocks in the second wafer pattern, then the above matching result is a match; if the total number of horizontal image blocks in the first wafer pattern is not equal to the total number of horizontal image blocks in the second wafer pattern and / or the total number of vertical image blocks in the first wafer pattern is not equal to the total number of vertical image blocks in the second wafer pattern, then the above matching result is a mismatch.

[0084] In one example, if the matching result between the wafer and the first wafer pattern is a match, steps S120 and S130 can be continued. This allows for the determination of a more accurate anomaly detection result when the wafer matches the first wafer pattern. If the matching result between the wafer and the first wafer pattern is a mismatch, the user can be prompted, the user can request a switch to the first wafer pattern, and the matching result between the wafer and the switched first wafer pattern can be determined.

[0085] According to the above-described scheme of the present invention, a second wafer map can be determined based on multiple target regions in a wafer image including the wafer and a first region category corresponding to each of the multiple target regions. Then, based on the second wafer map, the matching result between the wafer and the first wafer map is determined. The above scheme can determine the matching result between the wafer and the first wafer map through the second wafer map, which can improve the accuracy of anomaly detection results.

[0086] For example, step S220, based on the second wafer pattern, determines the matching result between the wafer and the first wafer pattern, including steps S221 and S222.

[0087] In step S221, the predicted size of the wafer is determined based on the region corresponding to the wafer in the second wafer map and the preset size corresponding to each image block in the second wafer map.

[0088] The predicted wafer size can be determined based on the preset size corresponding to each image block in the second wafer map and the number of image blocks in the region corresponding to the wafer. Specifically, the preset size of the image block can be the size of a single die in the wafer. It can be understood that the first and second wafer maps may include non-wafer regions in addition to the region corresponding to the wafer. Therefore, the predicted wafer size can be determined based on the number of image blocks in the region corresponding to the wafer in the second wafer map. For example, the predicted lateral size of the wafer can be determined by multiplying the total number of the most abundant image blocks in the region corresponding to the wafer by the preset lateral size corresponding to the image blocks. Similarly, the predicted lateral size of the wafer can be determined by multiplying the total number of the most abundant image blocks in the region corresponding to the wafer by the preset lateral size corresponding to the image blocks.

[0089] In step S222, the matching result between the wafer and the first wafer pattern is determined based on the preset size corresponding to the first wafer pattern and the predicted size of the wafer.

[0090] If the difference between the preset size corresponding to the first wafer pattern and the predicted size of the wafer is less than a first threshold, the matching result between the wafer and the first wafer pattern is determined to be a match. Otherwise, the matching result between the wafer and the first wafer pattern is determined to be a mismatch. Specifically, if the preset size corresponding to the first wafer pattern includes a first length (which can be considered as a horizontal dimension) and a first width (which can be considered as a vertical dimension), and the predicted size of the wafer includes a second length (which can be considered as a horizontal dimension) and a second width (which can be considered as a vertical dimension), then if the absolute value of the difference between the first length and the second length is less than the first length threshold and the absolute value of the difference between the first width and the second width is less than the first width threshold, it can be considered that the difference between the preset size corresponding to the first wafer pattern and the predicted size of the wafer is less than the first threshold. For example, the first length threshold can be four times the length of a die in the wafer, and the first width threshold can be four times the width of a die in the wafer. In one example, the sign of the above difference can also be used to determine whether the wafer is too large or too small compared to the preset size of the first wafer pattern. The specific values ​​of the first threshold, the first length threshold, and the first width threshold mentioned above are not limited in this embodiment of the invention and can be determined according to the actual needs of the developers.

[0091] According to the above-described scheme of the present invention, the predicted size of the wafer can be determined based on the region corresponding to the wafer in the second wafer map and the preset size corresponding to each image block in the second wafer map. Then, based on the preset size corresponding to the first wafer map and the predicted size of the wafer, the matching result between the wafer and the first wafer map is determined. The above scheme can determine the matching result between the wafer and the first wafer map based on the preset size corresponding to the first wafer map and the predicted size of the wafer, which is beneficial to improving the matching efficiency and thus improving the overall efficiency of the processing method.

[0092] For example, the anomaly detection result of the target region includes anomaly result information indicating whether an anomaly exists, and, when the anomaly result information indicates that an anomaly exists in the target region, position information of the target region in at least one of the first wafer map, the second wafer map, and the wafer image, wherein the second wafer map is determined based on multiple target regions in the wafer image including the wafer and the first region category corresponding to each of the multiple target regions.

[0093] When the anomaly detection result of the target area is used to indicate that there is an anomaly at the corresponding location of the target area on the wafer, the anomaly detection result includes anomaly result information indicating the existence of an anomaly and location information indicating the location of the target area. Specifically, the location information of the target area can be marked in at least one of the first wafer map, the second wafer map, and the wafer image.

[0094] When the anomaly detection result of the target area is used to indicate that there is no anomaly at the corresponding position of the target area on the wafer, the anomaly detection result includes anomaly result information indicating that there is no anomaly. In some embodiments, the position information of the target area may also be marked in at least one of the first wafer map, the second wafer map, and the wafer image, and the visualization effect may be different from the position information marked in the above example for distinction.

[0095] According to the above-described scheme of the present invention, the anomaly detection result of the target region is used to indicate whether an anomaly exists. When the anomaly result information indicates that an anomaly exists in the target region, the anomaly detection result of the target region is further used to indicate the location information of the target region in at least one of the first wafer map, the second wafer map, and the wafer image. The above scheme can provide the user with the location of the target region with an anomaly in at least one of the first wafer map, the second wafer map, and the wafer image, providing the user with more comprehensive information and facilitating the user to locate the target region with an anomaly in a timely manner for subsequent wafer processing.

[0096] For example, for each of the multiple target regions, if the anomaly detection result of that target region is used to indicate the presence of an anomaly, the above processing method further includes steps S310 to S330.

[0097] In step S310, if it is determined that the region corresponding to the target region in the first wafer pattern is a defective bare wafer region, a first prompt message is generated.

[0098] The defective wafer area is the region in the first wafer pattern corresponding to the defective residual wafer.

[0099] The region corresponding to the residual die in the first wafer map includes a defective die region and a functional die region. When the anomaly detection result of the target region indicates the presence of an anomaly, if the second region category corresponding to the target region is a residual die, the region corresponding to the target region in the first wafer map can be determined to be either a defective die region or a functional die region based on the location information of the target region in the first wafer map. If it is determined that the region corresponding to the target region in the first wafer map is a defective die region, a first prompt message is generated. The first prompt message indicates that the target region is a defective die region and that an anomaly exists. For example, it can be used to indicate that a defective die has been mistakenly removed from the wafer. In some embodiments, the first prompt message also includes the location information of the defective die in the wafer. For example, it can be the location information in at least one of the first wafer map, the second wafer map, and the wafer image.

[0100] In step S320, if it is determined that the region corresponding to the target region in the first wafer pattern is a functional die region, a second prompt message is generated.

[0101] The functional die region is the area in the first wafer pattern corresponding to the remaining dies other than the defective dies.

[0102] If the target region is determined to be a functional die region in the first wafer map, a second prompt message is generated. The second prompt message indicates that the target region is a functional die region and that an anomaly exists. For example, it can indicate that a functional die has been mistakenly removed from the wafer. In some embodiments, the second prompt message also includes the location information of the functional die within the wafer. For example, this could be the location information in at least one of the first wafer map, the second wafer map, and the wafer image.

[0103] In step S330, if it is determined that the region corresponding to the target region in the first wafer pattern is an empty region, a third prompt message is generated.

[0104] The empty region is the region corresponding to the empty region in the first wafer pattern.

[0105] If the anomaly detection result of the target area indicates the presence of an anomaly, and the second region category corresponding to the target area is an empty region, it can be determined that the region corresponding to the target area in the first wafer map is an empty region, and a third prompt message is generated. The third prompt message indicates that the target area is an empty region and that an anomaly exists. For example, it can be used to indicate that there is a bare die with acceptable electrical performance that was not removed due to incomplete cutting. In some embodiments, the third prompt message also includes the location information of the bare die with acceptable electrical performance that was not removed due to incomplete cutting within the wafer. For example, it can be the location information in at least one of the first wafer map, the second wafer map, and the wafer image.

[0106] The aforementioned first, second, and third prompts may be presented as visual or audio prompts, and this embodiment of the invention does not impose any limitations on them.

[0107] According to the above-described scheme of the present invention, for each of the multiple target regions, when the anomaly detection result of the target region indicates the presence of an anomaly, a first prompt message can be generated if the region corresponding to the target region in the first wafer map is determined to be a defective die region. If the region corresponding to the target region in the first wafer map is determined to be a functional die region, a second prompt message is generated. If the region corresponding to the target region in the first wafer map is determined to be an empty region, a third prompt message is generated. This scheme, by generating corresponding prompt messages for defective die regions, functional die regions, and empty regions, can provide more accurate prompts to users for different anomalies, facilitating timely location of the target region with the anomaly for subsequent wafer processing.

[0108] For example, the above processing method further includes: determining the anomaly detection result of the wafer based on the anomaly detection results of each of the multiple target regions.

[0109] The anomaly detection results of the wafer can be determined by combining the anomaly detection results of multiple target regions. For example, the anomaly detection result of the wafer includes the anomaly detection results of all target regions. Another example is that the anomaly detection result of the wafer also includes the number of target regions corresponding to the anomaly detection results indicating the presence of anomalies. Yet another example is that when the anomaly detection results of all target regions are used to indicate the absence of anomalies, the anomaly detection result of the wafer includes "the wafer has no anomalies." It can be understood that if the total number of target regions whose anomaly detection results indicate the presence of anomalies is less than a preset total threshold, the anomaly detection result of the wafer can also be considered as including "the wafer has no anomalies." The specific value of the aforementioned preset total threshold is not limited in this embodiment of the invention.

[0110] According to the above-described solution of the present invention, the anomaly detection result of the wafer can be determined based on the anomaly detection results of multiple target regions. The anomaly detection result of the wafer can more comprehensively reflect the overall condition of the wafer, providing users with a more comprehensive reference.

[0111] This invention also provides a wafer image processing apparatus. Figure 2 A schematic block diagram of a processing apparatus 400 according to an embodiment of the present invention is shown. (In conjunction with...) Figure 2 As shown, the processing device 400 may include a first region category determination module 410, a second region category determination module 420, and a region detection result determination module 430.

[0112] For example, the first region category determination module 410 is used to determine multiple target regions in a wafer image including a wafer and a first region category corresponding to each of the multiple target regions. The wafer includes multiple residual dies and multiple empty areas formed by removing the dies. The first region category corresponding to the target region indicates that the region corresponding to the target region in the wafer image is either the region corresponding to the residual die or the region corresponding to the empty area. The second region category determination module 420 is used, for each target region in the multiple target regions, to determine a second region category corresponding to the target region in a first wafer map of the wafer based on the position of the target region in the wafer image. The second region category indicates that the region corresponding to the target region in the first wafer map is either the region corresponding to the residual die or the region corresponding to the empty area. The region detection result determination module 430 is used, for each target region in the multiple target regions, to determine an anomaly detection result for the target region based on the first region category and the second region category corresponding to the target region. The anomaly detection result indicates whether an anomaly exists in the target region.

[0113] For example, the processing device 400 further includes a first image acquisition module, a directional marker region determination module, and a first rotation angle determination module.

[0114] The first image acquisition module is used to acquire a first image including the wafer. The orientation mark region determination module is used to determine the orientation mark region in the first image, wherein the orientation mark region is the region in the first image corresponding to the orientation mark of the wafer. The first rotation angle determination module is used to determine a first rotation angle of the wafer based on the relative positional relationship between the orientation mark region and the vertical axis, wherein the orientation mark of the wafer faces downwards towards the vertical axis after the wafer rotates by the first rotation angle, and the wafer image is an image acquired for the wafer rotated based on the first rotation angle.

[0115] For example, the area detection result determination module 430 includes at least one of the following: a first area detection result determination module and a second area detection result determination module.

[0116] The first region detection result determination module is used to determine the anomaly detection result of the target region as indicating that no anomaly exists when the first region category corresponding to the target region and the second region category corresponding to the target region represent the same category. The second region detection result determination module is used to determine the anomaly detection result of the target region as indicating that anomaly exists when the first region category corresponding to the target region and the second region category corresponding to the target region represent different categories.

[0117] For example, the region detection result determination module 430 includes a third region detection result determination module, a fourth region detection result determination module, and a fifth region detection result determination module.

[0118] The third determination module for the region detection result is used to determine the first data and the second data corresponding to the target region. The first region category corresponding to the target region indicates that, in the case where the target region corresponds to a residual die in the wafer image, the first data is a first natural number. The second region category indicates that, in the case where the target region corresponds to an empty region in the wafer image, the first data is a second natural number, and the sum of the second and first natural numbers is odd. The second region category indicates that, in the case where the target region corresponds to a residual die in the first wafer image, the second data is a third natural number, and the sum of the third and first natural numbers is even. The third region category indicates that, in the case where the target region corresponds to an empty region in the first wafer image, the second data is a fourth natural number, and the sum of the fourth and second natural numbers is even. The fourth module for determining the region detection result is used to determine the anomaly detection result of the target region when the sum of the first data and the second data corresponding to the target region is odd, indicating the presence of an anomaly. The fifth module for determining the region detection result is used to determine the anomaly detection result of the target region when the sum of the first data and the second data corresponding to the target region is even, indicating the absence of an anomaly.

[0119] For example, the processing device 400 further includes a second wafer pattern determination module and a matching result determination module.

[0120] The second wafer map determination module is used to determine a second wafer map based on multiple target regions in a wafer image including the wafer and the first region category corresponding to each of the multiple target regions. The matching result determination module is used to determine the matching result between the wafer and the first wafer map based on the second wafer map.

[0121] For example, the matching result determination module includes a first matching result determination module and a second matching result determination module.

[0122] The first matching result determination module is used to determine the predicted size of the wafer based on the region corresponding to the wafer in the second wafer map and the preset size corresponding to each image block in the second wafer map. The second matching result determination module is used to determine the matching result between the wafer and the first wafer map based on the preset size corresponding to the first wafer map and the predicted size of the wafer.

[0123] For example, the anomaly detection result of the target region includes anomaly result information indicating whether an anomaly exists, and, when the anomaly result information indicates that an anomaly exists in the target region, position information of the target region in at least one of the first wafer map, the second wafer map, and the wafer image, wherein the second wafer map is determined based on multiple target regions in the wafer image including the wafer and the first region category corresponding to each of the multiple target regions.

[0124] For example, for each of the multiple target areas, when the anomaly detection result of the target area is used to indicate the presence of an anomaly, the processing device 400 further includes a first prompt information generation module, a second prompt information generation module, and a third prompt information generation module.

[0125] The first prompt information generation module generates a first prompt information when it is determined that the target region corresponds to a defective die region in the first wafer pattern, wherein the defective die region is the region corresponding to a defective residual die in the first wafer pattern. The second prompt information generation module generates a second prompt information when it is determined that the target region corresponds to a functional die region in the first wafer pattern, wherein the functional die region is the region corresponding to a residual die other than a defective die in the first wafer pattern. The third prompt information generation module generates a third prompt information when it is determined that the target region corresponds to an empty region in the first wafer pattern, wherein the empty region is the region corresponding to an empty region in the first wafer pattern.

[0126] For example, the processing device 400 also includes a wafer inspection result determination module.

[0127] The wafer inspection result determination module is used to determine the wafer's anomaly detection result based on the anomaly detection results of multiple target areas.

[0128] According to another aspect of the present invention, an electronic device is also provided. Figure 3 A schematic block diagram of an electronic device 500 according to an embodiment of the present invention is shown. Figure 3As shown, the electronic device 500 includes a processor 510 and a memory 520. The memory 520 stores a computer program, and the computer program instructions are executed by the processor 510 to perform the above-described wafer image processing method.

[0129] Furthermore, according to another aspect of the present invention, a storage medium is provided, on which program instructions are stored. When the program instructions are executed by a computer or processor, the computer or processor performs corresponding steps of the wafer image processing method described in the embodiments of the present invention, and is used to implement corresponding modules in the wafer image processing apparatus or the electronic device described in the embodiments of the present invention. The storage medium may, for example, include a memory card of a smartphone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media. According to yet another aspect of the present invention, a computer program product is also provided, including computer program instructions. When the computer program instructions are executed by a computer or processor, the computer or processor performs corresponding steps of the wafer image processing method described above.

[0130] Those skilled in the art can understand the specific implementation schemes of the above-mentioned electronic devices and storage media by reading the relevant descriptions of the wafer image processing methods. For the sake of brevity, they will not be described in detail here.

[0131] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0132] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0133] In the several embodiments provided by this invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0134] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0135] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of the invention. However, this approach of the invention should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with fewer features than all of those in a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0136] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or elements of any method or apparatus so disclosed may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0137] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments can be used in any combination.

[0138] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules in the wafer image processing apparatus according to embodiments of the present invention. The present invention can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing some or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0139] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0140] The above description is merely a specific embodiment of the present invention or an explanation of that embodiment. The scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this invention should be included within the scope of protection of this invention. The scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for processing wafer images, characterized in that, The method includes: Determine multiple target regions in a wafer image including a wafer and a first region category corresponding to each of the multiple target regions, wherein the wafer includes multiple residual dies and multiple empty areas formed by the removal of dies, and the first region category corresponding to the target region is used to indicate whether the region corresponding to the target region in the wafer image is the region corresponding to the residual die or the region corresponding to the empty area; For each of the plurality of target regions, Based on the location of the target region in the wafer image, a second region category corresponding to the target region in the first wafer map of the wafer is determined, wherein the second region category is used to indicate that the region corresponding to the target region in the first wafer map is the region corresponding to the residual die or the region corresponding to the empty area; Based on the first region category and the second region category corresponding to the target region, the anomaly detection result of the target region is determined, wherein the anomaly detection result is used to indicate whether there is an anomaly in the target region.

2. The method as described in claim 1, characterized in that, The method further includes: Obtain a first image including the wafer; Determine the orientation mark region in the first image, wherein the orientation mark region is the region in the first image corresponding to the orientation mark of the wafer; Based on the relative positional relationship between the orientation mark area and the vertical axis, a first rotation angle of the wafer is determined, wherein the orientation mark of the wafer faces downward toward the vertical axis after the wafer rotates by the first rotation angle, and the wafer image is an image acquired for the wafer rotated based on the first rotation angle.

3. The method as described in claim 1, characterized in that, The determination of the anomaly detection result of the target region based on the first region category and the second region category of the target region includes at least one of the following: If the first region category corresponding to the target region is the same as the region represented by the second region category corresponding to the target region, the anomaly detection result of the target region is determined to indicate that no anomaly exists; If the first region category corresponding to the target region is different from the second region category corresponding to the target region, the anomaly detection result of the target region is used to indicate the presence of an anomaly.

4. The method as described in claim 1, characterized in that, The step of determining the anomaly detection result of the target region based on the first region category and the second region category corresponding to the target region includes: The first data and the second data corresponding to the target region are determined. The first region category of the target region indicates that, in the case where the region corresponding to the target region in the wafer image is a region corresponding to a residual die, the first data corresponding to the target region is a first natural number. The first region category of the target region indicates that, in the case where the region corresponding to the target region in the wafer image is a region corresponding to an empty region, the first data corresponding to the target region is a second natural number, and the sum of the second natural number and the first natural number is odd. The second region category of the target region indicates that, in the case where the region corresponding to the target region in the first wafer image is a region corresponding to a residual die, the second data corresponding to the target region is a third natural number, and the sum of the third natural number and the first natural number is even. The second region category of the target region indicates that, in the case where the region corresponding to the target region in the first wafer image is a region corresponding to an empty region, the second data corresponding to the target region is a fourth natural number, and the sum of the fourth natural number and the second natural number is even. If the sum of the first data corresponding to the target area and the second data corresponding to the target area is odd, the anomaly detection result of the target area is used to indicate the presence of an anomaly. If the sum of the first data corresponding to the target area and the second data corresponding to the target area is even, the anomaly detection result of the target area is used to indicate that no anomaly exists.

5. The method as described in claim 1, characterized in that, The method further includes: A second wafer map is determined based on multiple target regions in a wafer image including the wafer and the first region category corresponding to each of the multiple target regions; Based on the second wafer map, the matching result between the wafer and the first wafer map is determined.

6. The method as described in claim 5, characterized in that, The step of determining the matching result between the wafer and the first wafer map based on the second wafer map includes: Based on the region corresponding to the wafer in the second wafer map and the preset size corresponding to each image block in the second wafer map, the predicted size of the wafer is determined; Based on the preset size corresponding to the first wafer pattern and the predicted size of the wafer, the matching result between the wafer and the first wafer pattern is determined.

7. The method as described in claim 1, characterized in that, The anomaly detection result of the target region includes anomaly result information indicating whether an anomaly exists, and, when the anomaly result information indicates that an anomaly exists in the target region, position information of the target region in at least one of the first wafer map, the second wafer map, and the wafer image. The second wafer map is determined based on multiple target regions in the wafer image including the wafer and the first region category corresponding to each of the multiple target regions.

8. The method as described in claim 1, characterized in that, For each of the plurality of target regions, where the anomaly detection result of that target region is used to indicate the presence of an anomaly, the method further includes: If it is determined that the target area corresponds to a defective die area in the first wafer map, a first prompt message is generated, wherein the defective die area is the area corresponding to the defective residual die in the first wafer map, and the first prompt message is used to indicate that the target area is a defective die area and that there is an anomaly. If it is determined that the target area corresponds to a functional die area in the first wafer map, a second prompt message is generated, wherein the functional die area is the area in the first wafer map corresponding to residual dies other than unqualified dies, and the second prompt message is used to indicate that the target area is a functional die area and that there is an anomaly. If it is determined that the target region corresponds to an empty region in the first wafer map, a third prompt message is generated, wherein the empty region is the region corresponding to the empty region in the first wafer map, and the third prompt message is used to indicate that the target region is an empty region and that there is an anomaly.

9. The method as described in claim 1, characterized in that, The method further includes: Based on the anomaly detection results of the multiple target regions, the anomaly detection results of the wafer are determined.

10. A wafer image processing apparatus, characterized in that, The device includes: The first region category determination module is used to determine multiple target regions in a wafer image including a wafer and the first region category corresponding to each of the multiple target regions. The wafer includes multiple residual dies and multiple empty areas formed by removing the dies. The first region category corresponding to the target region is used to indicate that the region corresponding to the target region in the wafer image is the region corresponding to the residual die or the region corresponding to the empty area. The second region category determination module is used to determine the second region category corresponding to each target region in the first wafer map of the wafer based on the position of the target region in the wafer image for each of the plurality of target regions. The second region category is used to indicate that the region corresponding to the target region in the first wafer map is the region corresponding to the residual die or the region corresponding to the empty area. The region detection result determination module is used to determine the anomaly detection result of each target region in the plurality of target regions based on the first region category and the second region category corresponding to the target region. The anomaly detection result is used to indicate whether there is an anomaly in the target region.

11. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the wafer image processing method as described in any one of claims 1-9.

12. A storage medium storing computer program instructions, characterized in that, The computer program instructions, when executed, are used to perform the wafer image processing method as described in any one of claims 1-9.

13. A computer program product comprising computer program instructions, characterized in that, The computer program instructions, when executed by a processor, are used to perform the wafer image processing method as described in any one of claims 1-9.

Images