Wafer defect detection preprocessing method and computer program product

Abstract

The application provides a wafer defect detection preprocessing method and a computer program product. The wafer defect detection preprocessing method comprises the following steps: obtaining a metal layer layout of a wafer to be detected; extracting an entity region to be processed from a layout pattern of the metal layer layout; performing a breaking processing on the boundary of the entity region to generate a plurality of broken edges; generating a corresponding type of rectangular unit for different types of broken edges to split the entity region; and inputting the generated rectangular units of various types into a wafer defect detection database. The advantage of the application is that the entity region of the metal layer layout can be converted into a standardized rectangular unit, the detection interference caused by irregular patterns can be eliminated, and the accuracy of defect detection can be improved.

Description

Technical Field

[0001] This invention relates to the field of wafer defect detection technology, and in particular to a preprocessing method and computer program product for wafer defect detection. Background Technology

[0002] In the semiconductor manufacturing industry, as integrated circuit manufacturing processes continue to shrink, wafer inspection is becoming increasingly critical for ensuring chip performance and yield. Tiny defects can severely impact chip quality. Timely identification of manufacturing defects through wafer inspection helps manufacturers optimize process parameters, reduce defect rates, improve production efficiency and product quality, while also reducing costs and shortening time to market.

[0003] Currently, D2D, C2C, and D2DB are the mainstream wafer inspection technologies. D2D identifies defects by comparing adjacent die patterns on the same wafer, but obtaining stable reference images is difficult; C2C is suitable for regions with repeating cells, but its detection range is limited; D2DB compares wafer images with patterns in a design database, but it is easily affected by process variations, resulting in noise and false alarms. These traditional inspection technologies have shortcomings in terms of accuracy, adaptability of inspection parameters, and universality of inspection patterns, making it difficult to meet the increasingly complex needs of semiconductor manufacturing. There is an urgent need to innovate inspection methods to improve wafer inspection capabilities. Summary of the Invention

[0004] One object of the present invention is to overcome at least one technical defect in the prior art and to provide a preprocessing method and computer program product for wafer defect detection.

[0005] A further objective of this invention is to transform the solid regions of the metal layer layout into standardized rectangular units, thereby eliminating detection interference caused by irregular patterns and improving the accuracy of defect detection.

[0006] Another further objective of this invention is to improve the versatility of the wafer defect detection preprocessing method, enabling it to be widely applied to semiconductor wafer manufacturing scenarios with different process and design requirements.

[0007] Specifically, according to a first aspect of the present invention, the present invention provides a preprocessing method for wafer defect detection, comprising:

[0008] Obtain the metal layer layout of the wafer to be inspected;

[0009] Extract the entity region to be processed from the layout pattern of the metal layer layout;

[0010] The boundaries of the physical region are broken to generate multiple broken edges;

[0011] For different types of broken edges, corresponding rectangular units are generated to divide the solid region;

[0012] The generated rectangular cells are input into the wafer defect detection database.

[0013] Optionally, in the metal layer layout, a portion of the boundary of the solid region along the first coordinate direction intersects with a portion of the boundary along the second coordinate direction to form a corner vertex. The corner vertex is divided into an inner concave corner vertex and an outer convex corner vertex according to the concavity and convexity characteristics of the boundary at its location.

[0014] The steps for breaking the boundary of the physical region include:

[0015] Identify each concave corner vertex in the solid region, and for each concave corner vertex, project it onto its opposite side along the mutually perpendicular first coordinate direction and second coordinate direction to obtain the projection breakpoint;

[0016] For any two opposite sides of any set of boundaries along the first coordinate direction and any two opposite sides of any set of boundaries along the second coordinate direction in the entity region, multiple pairs of evenly divided breakpoints are configured according to a preset interval distance.

[0017] The boundary of the solid region is divided into multiple broken edges by passing through each concave corner vertex and its corresponding projection breakpoint, as well as through each pair of equally divided breakpoints.

[0018] Optionally, the type of the broken edge includes an evenly divided broken edge, which refers to a broken edge whose two endpoints are both the evenly divided break points;

[0019] The steps for generating corresponding rectangular elements for different types of broken edges include:

[0020] Select two corresponding equally divided edges from the same set of boundaries of the entity region;

[0021] The two equally divided broken edges are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two equally divided broken edges is used as another set of parallel edges, thereby generating an equally divided rectangular unit.

[0022] Optionally, the type of the broken edge includes a folded broken edge, which refers to a broken edge with one end being the vertex of the concave corner or the vertex of the convex corner, and the other end being the evenly divided break point;

[0023] The steps for generating corresponding rectangular elements for different types of broken edges include:

[0024] Identify the included angle formed by a pair of broken edges at the same corner in the physical region, wherein the pair of broken edges have a common concave or convex vertex.

[0025] Based on the positional relationship of each included angle, a folded rectangle unit is generated, and the folded rectangle unit completely covers all the included angles at the same fold.

[0026] Optionally, the type of broken edge includes an end broken edge, which refers to a broken edge where both endpoints are vertices of the outward convex corner;

[0027] The steps for generating corresponding rectangular elements for different types of broken edges include:

[0028] Identify the length of the broken end edge, generate two parallel edges that are parallel to and of equal length to the broken end edge, and make the two generated parallel edges symmetrically distributed about the broken end edge;

[0029] The line connecting the corresponding endpoints of the two generated parallel sides is used as another set of parallel sides, thereby generating the end rectangular unit.

[0030] Optionally, the evenly divided rectangular unit includes a first rectangular unit whose length direction is along the first coordinate direction, and a second rectangular unit whose length direction is along the second coordinate direction;

[0031] After the step of generating corresponding rectangular elements for different types of broken edges, the following steps are also included:

[0032] The first directional rectangular units that are tangent only to the terminal rectangular units, and tangent only to one of the terminal rectangular units, are classified into the first directional single-terminal cluster group;

[0033] The first directional rectangular units that are tangent only to the terminal rectangular units, and only to two of the terminal rectangular units, are classified into the first directional double-terminal cluster group.

[0034] The first directional rectangular unit that is tangent only to the corner rectangular unit and only to one of the corner rectangular units is classified into the first directional single-corner cluster group;

[0035] The first directional rectangular units that are tangent only to the corner rectangular units, and only to two of the corner rectangular units, are classified into the first directional double-corner cluster group.

[0036] Optionally, after the step of generating corresponding types of rectangular elements for different types of broken edges, the method further includes:

[0037] For the second directional rectangular unit, if there are gaps between it and other types of rectangular units on both sides in the first coordinate direction, and the gap width on both sides is less than the preset width, then the second directional rectangular unit is classified into the second directional dense cluster group; if the gap width on both sides is greater than or equal to the preset width, then the second directional rectangular unit is classified into the second directional sparse cluster group.

[0038] Optionally, after the step of generating the folded rectangular unit for the folded edge, the method further includes:

[0039] The folded rectangular units that are tangent to or overlap with the two rectangular units are classified into a double-adjacent folded corner cluster group;

[0040] The folded rectangular units that are tangent to or overlap with the three rectangular units are classified into three adjacent folded cluster groups.

[0041] Optionally, the type of broken edge includes a branched broken edge, which refers to a broken edge with one endpoint being the vertex of the concave angle and the other endpoint being the vertex of the convex angle;

[0042] The steps for generating corresponding rectangular elements for different types of broken edges include:

[0043] Select two corresponding branch breaks from the same set of boundaries of the entity region;

[0044] Two of the branch breaks are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two branch breaks is used as another set of parallel edges, thereby generating a branch rectangular unit.

[0045] Optionally, after the step of generating branch rectangular units for the branch break edge, the method further includes:

[0046] The branch rectangle cells containing two broken edges along the first coordinate direction are classified into the first directional branch cluster group;

[0047] Branch rectangular cells containing two broken branches along the second coordinate direction are classified into the second-direction branch cluster group.

[0048] Optionally, the type of broken edge includes a breakable broken edge, which refers to a broken edge with one endpoint being the vertex of the concave angle and the other endpoint being the projection point of the concave angle. The breakable broken edges appear in pairs in the same set of boundaries of the solid region. The projection point obtained by projecting from the vertex of the concave angle of one of the breakable broken edges along the first coordinate direction or the second coordinate direction to the other breakable broken edge is the projection point of the concave angle of the other breakable broken edge.

[0049] The steps for generating corresponding rectangular elements for different types of broken edges include:

[0050] Select two corresponding easily broken edges from the same set of boundaries of the entity region;

[0051] Two of the easily broken edges are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two easily broken edges is used as another set of parallel edges, thereby generating an easily broken rectangular unit.

[0052] According to a second aspect of the present invention, the present invention provides a computer program product comprising a computer program, which, when executed by a processor, is used to implement the preprocessing method for wafer defect detection as described above.

[0053] The preprocessing method for wafer defect detection of the present invention first obtains the metal layer layout of the wafer to be inspected and extracts the solid regions to be processed from the layout pattern. Then, the boundaries of the solid regions are broken to generate multiple broken edges. Next, corresponding types of rectangular cells are generated for different types of broken edges to segment the solid regions. Finally, the generated rectangular cells are input into a wafer defect detection database. The present invention utilizes the above method to segment complex and irregular solid regions into standardized rectangular cells, eliminating the interference of irregular patterns on the detection algorithm, enabling subsequent detection algorithms to analyze and process with consistent logic, and reducing detection errors caused by pattern differences. Simultaneously, inputting the generated rectangular cells into the wafer defect detection database can also provide standardized and efficient data resources for subsequent detection.

[0054] Furthermore, the preprocessing method for wafer defect detection in this invention significantly improves universality through an innovative boundary breaking strategy. At the concave corner vertices of the solid region, projection breakpoints are obtained by projecting inwards along two mutually perpendicular coordinate directions, accurately locating key boundary turning points. Simultaneously, breakpoints are evenly distributed at preset intervals along the boundaries in each direction, achieving standardized boundary segmentation. The two breakpoints work synergistically, decomposing the solid region boundary into independent line segments regardless of the regularity of the layout shape, generating universal data. This processing method is independent of specific layout shapes and process designs, providing a reliable foundation for subsequent generation of standardized rectangular cells. This makes the preprocessing method applicable to various semiconductor wafer manufacturing scenarios, effectively enhancing the universality and reliability of wafer defect detection.

[0055] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0056] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0057] Figure 1This is a flowchart of a preprocessing method for wafer defect detection according to an embodiment of the present invention;

[0058] Figure 2 This is a schematic structural diagram of the layout pattern of a metal layer layout according to an embodiment of the present invention;

[0059] Figure 3 This is a flowchart of a process for breaking up a physical region according to an embodiment of the present invention;

[0060] Figure 4 Is Figure 2 A schematic diagram of the local configuration of projection breakpoints in the layout graphic shown;

[0061] Figure 5 Is Figure 2 A schematic diagram showing the arrangement of evenly divided breakpoints in a local area of ​​the layout graphic;

[0062] Figure 6 Yes Figure 2 A schematic flowchart illustrating the processing of independent rectangles in the aforementioned layout graphic;

[0063] Figure 7 yes Figure 2 The diagram shown contains a partial view of the layout graphic with independent rectangular units generated.

[0064] Figure 8 yes Figure 2 The diagram shown is a partial view of the layout graphic with end rectangular cells generated.

[0065] Figure 9 yes Figure 2 The diagram shown is a partial illustration of the layout graphic in which branched rectangular cells are generated.

[0066] Figure 10 yes Figure 2 The diagram shown is a partial illustration of a layout graphic in which fragile rectangular cells are generated.

[0067] Figure 11 yes Figure 2 The diagram shown is a partial illustration of a layout graphic containing angled rectangular cells.

[0068] Figure 12 yes Figure 2 The diagram shown is a partial schematic of the layout graphic in which first-direction rectangular cells are generated;

[0069] Figure 13 yes Figure 2 A partial schematic diagram showing the generation of second-direction rectangular cells in the layout graphic shown;

[0070] Figure 14 yes Figure 2The diagram shown includes a partial schematic of the first-direction single-angle cluster group;

[0071] Figure 15 yes Figure 2 The diagram shown includes a partial schematic of the first-direction bi-angled cluster group;

[0072] Figure 16 yes Figure 2 The layout diagram shown includes a partial schematic of the first-direction single-terminal cluster group;

[0073] Figure 17 yes Figure 2 The layout diagram shown includes a partial schematic of the first-direction two-terminal cluster group;

[0074] Figure 18 yes Figure 2 The diagram shown contains a partial schematic of the second-dimensional dense clusters;

[0075] Figure 19 yes Figure 2 The layout diagram shown includes a partial schematic of the second-dimensional sparse clustering group;

[0076] Figure 20 yes Figure 2 The diagram shown includes partial schematics of the first-direction branching cluster group and the second-direction branching cluster group;

[0077] Figure 21 yes Figure 2 The diagram shown contains a partial schematic of a double-adjacent corner cluster group;

[0078] Figure 22 yes Figure 2 The diagram shown contains a partial schematic of a three-adjacent corner cluster group;

[0079] Figure 23 This is a schematic diagram of a computer program product according to an embodiment of the present invention.

[0080] Figure label:

[0081] 10. Metal layer layout; 101. Projected point break; 102. Evenly divided break point; 110. Independent rectangular element; 111. Independent broken edge; 120. End rectangular element; 121. End broken edge; 130. Branch rectangular element; 131. Branch broken edge; 140. Easily broken rectangular element; 141. Easily broken edge; 150. Corner rectangular element; 151. Corner broken edge; 160. First-direction rectangular element; 161. First-direction broken edge; 170. Second-direction rectangular element; 17 1. Second-direction broken edge; 200. First-direction single-angle cluster group; 210. First-direction double-angle cluster group; 220. First-direction single-terminal cluster group; 230. First-direction double-terminal cluster group; 240. Second-direction dense cluster group; 250. Second-direction sparse cluster group; 260. First-direction branching cluster group; 270. Second-direction branching cluster group; 280. Double-adjacent angled cluster group; 290. Triple-adjacent angled cluster group; 30. Computer program product; 31. Machine-executable program. Detailed Implementation

[0082] Reference will now be made in detail to embodiments of the invention, one or more of which are illustrated in the accompanying drawings. The various embodiments provided are intended to explain the invention and not to limit it. In fact, various modifications and variations to the invention will be apparent to those skilled in the art without departing from the scope or spirit of the invention. For example, a feature illustrated or described as part of one embodiment may be used with another embodiment to produce yet another embodiment. Therefore, the invention is intended to cover such modifications and variations within the scope of the appended claims and their equivalents.

[0083] The following reference Figures 1 to 23 This invention describes a preprocessing method and computer program product for wafer defect detection according to embodiments of the present invention. The terms "inner," "outer," "upper," "lower," "top," "bottom," "lateral," and "longitudinal," etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention.

[0084] In the description of this embodiment, it should be understood that the term "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. When a feature "includes or contains" one or more of the features it covers, unless otherwise specifically described, this indicates that other features are not excluded and may be further included.

[0085] In the description of this embodiment, the terms "one embodiment," "some embodiments," "some examples," "one example," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0086] This invention first provides a preprocessing method for wafer defect detection. Figure 1 This is a flowchart of a preprocessing method for wafer defect detection according to an embodiment of the present invention, as follows: Figure 1 As shown, the preprocessing method for wafer defect detection includes at least the following steps S102 to S110.

[0087] Step S102: Obtain the metal layer slab of the wafer to be inspected. Figure 10 Metallic layer plate Figure 10 Obtained from the design layout, metal layer plate Figure 10 This data contains design information about the metal circuitry on the wafer and is fundamental for defect detection. By obtaining this layout, key information such as the arrangement, shape, and size of the metal circuitry on the wafer can be determined, providing a primary reference for subsequent inspection work.

[0088] Step S104, from the metal layer plate Figure 10 Extract the entity region to be processed from the layout graphic. Metal layer plate Figure 10 It may contain a lot of redundant information. By extracting the entity area to be processed, it can locate the part that actually needs to be detected, eliminate interference from irrelevant areas, narrow the detection range, and improve detection efficiency.

[0089] Step S106 involves breaking the boundary of the solid region to generate multiple broken edges. The boundary of the solid region refers to the wafer metal layer. Figure 10 The set of contour lines defines the scope of the entity region to be processed. These contour lines serve as the boundary between the entity region and the surrounding non-detection areas, outlining the target area for defect detection and are the core objects of subsequent boundary decomposition and defect detection. By breaking down continuous boundaries into independent line segments, the boundary structure can be effectively simplified, laying the foundation for the subsequent generation of standardized rectangular units.

[0090] Step S108: Generate corresponding rectangular units for different types of broken edges to split the solid region. After the boundary of the solid region is broken, the boundary is divided into various types of independent line segments, i.e., broken edges. According to the type of broken edge, generate corresponding rectangular units for each type of broken edge, thereby splitting the complex solid region into multiple regular rectangular unit combinations.

[0091] Step S110: Input the generated rectangular cells into the wafer defect detection database. The wafer defect detection database is a dedicated database system used in the semiconductor manufacturing process to store, manage, and analyze wafer inspection data. It is a core hub connecting design data, inspection equipment, and process optimization.

[0092] Using the preprocessing method described above, complex and irregular solid regions can be divided into standardized rectangular units, eliminating the interference of graphic irregularities on the detection algorithm. This allows subsequent detection algorithms to analyze and process data with consistent logic, reducing detection errors caused by graphic differences. Furthermore, inputting the generated rectangular units into the wafer defect detection database provides standardized and efficient data resources for subsequent inspections.

[0093] Figure 2 It is a metal layer plate according to an embodiment of the present invention. Figure 10 A schematic structural diagram of the map, such as Figure 2 As shown, metal layer plate Figure 10 It is in GDS (Graphic Data System) format and consists of various design graphics. The shaded areas in the diagram represent the solid areas of the layout, which are the target objects for defect detection. Their outer contours form the boundaries of the solid areas, clearly defining the scope to be inspected.

[0094] In the metal layer layout, the partial boundary of the solid region along the first coordinate direction and the partial boundary along the second coordinate direction intersect to form a corner vertex. The corner vertex is divided into concave corner vertex and convex corner vertex according to the concavity and convexity of the boundary at its location.

[0095] In this embodiment, the first coordinate direction and the second coordinate direction respectively correspond to Figure 4 The horizontal and vertical directions presented on the paper. For determining the vertex of a fold, there is a method based on the size of the included angle at the boundary:

[0096] When two boundaries meet at a bend vertex, and the angle formed by the two boundaries without passing through the interior of the solid region is greater than 180°, the bend vertex is called the convex corner vertex. In this embodiment, the angle formed by the two boundaries connected to the convex corner vertex is 270°.

[0097] Conversely, if two boundaries meet at a corner vertex, and the angle formed by them without passing through the interior of the solid region is less than 180°, then the corner vertex is a concave corner vertex. In this embodiment, the angle formed by the two boundaries connected to the concave corner vertex is 90°.

[0098] Figure 3 This is a schematic flowchart illustrating the interruption process of a physical region according to an embodiment of the present invention, such as... Figure 3 As shown, breaking the boundary of a physical region may include the following steps S302 to S306.

[0099] Step S302: Identify each concave corner vertex in the solid region. For each concave corner vertex, project it onto its opposite side along the mutually perpendicular first and second coordinate directions to obtain projection breakpoint 101. The aforementioned concave corner vertices exist at the corners of the solid region, such as L-shaped bends, T-shaped intersections, and cross-shaped intersections.

[0100] Step S304: For any two opposite edges of a set of boundaries along the first coordinate direction and any two opposite edges of a set of boundaries along the second coordinate direction in the entity region, multiple pairs of evenly divided breakpoints 102 are configured according to a preset interval distance. The purpose of configuring the evenly divided breakpoints 102 in pairs is to ensure that the subsequent dividing lines formed based on the breakpoints are perpendicular to the corresponding boundaries, thereby providing a precise positioning reference for the regular segmentation of the entity region.

[0101] Step S306: Divide the boundary of the solid region into multiple broken edges by passing through each concave corner vertex and its corresponding projection breakpoint 101, and through each pair of equally divided breakpoints 102. The endpoints of each broken edge must be predefined key locations, including concave corner vertices, convex corner vertices, projection breakpoints 101, or equally divided breakpoints 102, to ensure that the start and end points of the line segments are located at these key locations. All breakpoints must participate in the boundary segmentation to ensure that the boundary of the solid region is completely and without omission decomposed into a set of line segments.

[0102] Using the above method, the innovative boundary breaking strategy significantly improves universality. The projected breakpoint 101 and the evenly divided breakpoint 102 work together to decompose the boundaries of solid regions into independent line segments, generating universal data, regardless of the regularity of the layout shape. This processing method is independent of specific layout shapes and process designs, providing a reliable foundation for the subsequent generation of standardized rectangular cells. This makes the preprocessing method of this embodiment applicable to various semiconductor wafer manufacturing scenarios, effectively enhancing the universality and reliability of wafer defect detection.

[0103] Figure 4 Is Figure 2 The diagram shows a partial configuration of projection breakpoint 101 in the layout graphic, as shown below. Figure 4 As shown, taking a concave vertex P as an example, starting from this point, projections are made along the first coordinate direction (corresponding to the horizontal X-axis in the figure) and the second coordinate direction (corresponding to the vertical Y-axis in the figure) to its opposite side. Projection breakpoints 101 are obtained on the boundaries of the two opposite sides.

[0104] Among them, the projection breakpoint 101 obtained along the first coordinate direction is denoted as Px, and the projection breakpoint 101 obtained along the second coordinate direction is denoted as Py.

[0105] Figure 5 Is Figure 2 The diagram shows a local area of ​​the layout graphic with evenly distributed breakpoints 102, as shown below. Figure 5 As shown, a uniformly divided edge layer is further generated based on the already generated projected edge layer. Specifically, in the straight line segment of the entity region boundary, the boundary can be uniformly divided at a preset interval. The boundary line segment between adjacent uniformly divided breakpoints 102 is also called an edge break. By combining the projected breakpoint 101 and the uniformly divided breakpoint 102, the structured segmentation of the entity region boundary is achieved.

[0106] Figure 6 Yes Figure 2 The flowchart for processing the independent rectangles in the layout graphic is as follows: Figure 6 As shown, in the metal layer plate Figure 10 Before the step of extracting the entity region to be processed, the following steps S602 to S604 may also be performed.

[0107] Step S602, Identify the metal layer plate Figure 10 An independent area within the metal layer plate. Figure 10 In the middle, there are rectangular solid regions that are spaced apart from other solid regions.

[0108] Step S604: Using the four boundaries of the independent region as independent cut edges 111, generate independent rectangular units 110.

[0109] Figure 7 yes Figure 2 The diagram shown contains a partial view of the layout graphic where independent rectangular units 110 are generated, as follows: Figure 7 As shown in the figure, the shaded areas represent the identified independent regions. Each independent region has a long side of B1nm–B2nm and a short side of B3nm–B4nm, with a length ratio of B5–B6. B1, B2, B3, B4, B5, and B6 can be determined based on the actual metal layer to be processed. Figure 10 Depending on the design features, the rectangular units generated by using the four boundaries of each independent region as independent cut edges 111 are called independent rectangular units 110.

[0110] In some optional embodiments, the type of broken edge includes an end broken edge 121, which refers to a broken edge where both endpoints are convex corner vertices. When generating corresponding rectangular units for different types of broken edges, the length of the end broken edge 121 can be identified, and two parallel edges of equal length parallel to the end broken edge 121 can be generated, with the two generated parallel edges symmetrically distributed about the end broken edge 121. Then, the line connecting the corresponding endpoints of the two generated parallel edges is used as another set of parallel edges, thereby generating the end rectangular unit 120.

[0111] Figure 8 yes Figure 2 The layout graphic shown is a partial schematic diagram with end rectangular cells 120 generated, as shown below. Figure 8 As shown, the shaded area represents the end rectangular unit 120, the line segment shown in bold is the end cut-off edge 121, and the point marked by the solid circle is the vertex of the convex corner. The end cut-off edge 121 has a significant characteristic: both its endpoints are convex corner vertices. The length of the end cut-off edge 121 can be between 0 and C1 nm, providing some flexibility. The distance between the other two sides parallel to the end cut-off edge 121 and the end cut-off edge 121 is set to C2 nm.

[0112] The end rectangular unit 120 generated by the above rules can accurately locate the geometric features at the end of the layout pattern, and is suitable for detecting process defects at the ends of metal lines. C1 and C2 are adjustable parameters that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0113] In some optional embodiments, the type of broken edge includes a branch broken edge 131, which refers to a broken edge with one endpoint being a concave corner vertex and the other endpoint being a convex corner vertex. When generating corresponding rectangular units for different types of broken edges, two corresponding branch broken edges 131 can be selected from the same set of boundaries of the solid region; the two branch broken edges 131 are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two branch broken edges 131 is used as another set of parallel edges, thereby generating a branch rectangular unit 130.

[0114] Figure 9 yes Figure 2 The diagram shown contains a partial schematic of a branched rectangular cell 130 generated in the layout graphic, as follows: Figure 9 As shown, the shaded area represents the branch rectangle unit 130, and the line segments shown in bold are the branch breaks 131. The points marked by solid circles are the convex corner vertices, and the points marked by dashed circles are the concave corner vertices. Branch breaks 131 usually appear in pairs, with one endpoint of the branch break 131 being the convex corner vertex and the other endpoint being the concave corner vertex. The length of the branch break 131 can be between 0 and D1nm.

[0115] The branch rectangle unit 130 generated by the above rules can accurately locate the geometric features at the branch points of the layout pattern, making it suitable for detecting process defects in metal line branches. Here, D1 is an adjustable parameter that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0116] In some embodiments, the type of broken edge includes a breakable broken edge 141, which refers to a broken edge with one endpoint being a concave corner vertex and the other endpoint being a concave corner projection point. Breakable broken edges 141 appear in pairs within the same set of boundaries of the solid region. Projection is made from the concave corner vertex of one breakable broken edge 141 along a first coordinate direction or a second coordinate direction to the other breakable broken edge 141, and the resulting projection point is the concave corner projection point of the other breakable broken edge 141. When generating corresponding rectangular elements for different types of broken edges, two corresponding breakable broken edges 141 can be selected from the same set of boundaries of the solid region. These two breakable broken edges 141 are then used as a set of parallel edges, and the line connecting the corresponding endpoints of the two breakable broken edges 141 is used as another set of parallel edges, thereby generating a breakable rectangular element 140.

[0117] In the solid region, the fragile edge 141 is usually located in a relatively weak structure that is more prone to breakage than other regions during subsequent wafer manufacturing processes (such as etching, deposition, etc.) or actual use.

[0118] Figure 10 yes Figure 2 The diagram shown contains a partial schematic of a fragile rectangular cell 140 generated within the layout graphic, as follows: Figure 10 As shown, the shaded area represents the fragile rectangular unit 140, the line segments shown in bold are the fragile edges 141, and the points marked by the dashed circles are the concave corner vertices. Fragile edges 141 typically appear in pairs. Specifically, the length of each fragile edge 141 is between 0 and E1 nm, while the distance between two fragile edges 141 is between 0 and E2 nm. Structurally, each fragile edge 141 has one endpoint as a concave corner vertex, as marked by the dashed circle in the figure, and the other endpoint is the projection point of the concave corner vertex of another fragile edge 141 onto it, i.e., the concave corner projection point, as marked by the dashed box in the figure.

[0119] The fragile rectangular unit 140 generated by the above rules can accurately locate the geometric features of the fragile parts of the layout pattern, and is suitable for detecting process defects of fragile metal lines. E1 and E2 are adjustable parameters that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0120] In some optional embodiments, the type of broken edge includes a folded broken edge 151, which refers to a broken edge with one end being a concave corner vertex or a convex corner vertex and the other end being a break point 102. When generating corresponding rectangular units for different types of broken edges, the included angle formed by pairs of folded broken edges 151 at the same corner in the solid region can be identified, wherein the pairs of folded broken edges 151 have a common concave corner vertex or a convex corner vertex. Then, based on the positional relationship of each included angle, a folded rectangular unit 150 is generated, and the folded rectangular unit 150 completely covers all included angles at the same corner.

[0121] Figure 11 yes Figure 2 The diagram shown contains a partial illustration of a bent rectangular cell 150 generated within the layout graphic, as follows: Figure 11 As shown, the shaded area represents the angled rectangular unit 150, and the line segments shown in bold are the angled breaks 151. The length of the angled breaks 151 is between A1 and A2 nm, with one endpoint being a convex or concave corner vertex and the other endpoint being the dividing point 102. The angled breaks 151 are usually marked in pairs, with each pair sharing a convex or concave corner vertex, forming a 90° angle between the two angled breaks 151. The angle formed by all the angled breaks 151 at the same angle is covered by the angled rectangular unit 150. For example, in... Figure 11 In the middle, the corner rectangle 150 located at the upper left covers one concave corner vertex and one convex corner vertex, the corner rectangle 150 located at the upper right covers two concave corner vertices, and the corner rectangle 150 located at the bottom covers four concave corner vertices.

[0122] The angled rectangular unit 150 generated by the above rules can accurately locate the geometric features at the angles of the layout pattern, making it suitable for detecting process defects at the angles of metal lines. A1 and A2 are adjustable parameters that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0123] In one optional embodiment, the type of broken edge includes an evenly divided broken edge, which refers to a broken edge where both endpoints are evenly divided break points. When generating corresponding rectangular elements for different types of broken edges, two corresponding evenly divided broken edges can be selected from the same set of boundaries of the solid region, and the two evenly divided broken edges can be used as a set of parallel edges. Then, the line connecting the corresponding endpoints of the two evenly divided broken edges can be used as another set of parallel edges, thereby generating an evenly divided rectangular element.

[0124] The evenly divided edge includes a first-direction edge 161 with its length direction along the first coordinate direction and a second-direction edge 171 with its length direction along the second coordinate direction. Correspondingly, the evenly divided rectangular element includes a first-direction rectangular element 160 with its length direction along the first coordinate direction and a second-direction rectangular element 170 with its length direction along the second coordinate direction.

[0125] Figure 12 yes Figure 2 The layout graphic shown contains a partial schematic diagram of a first-direction rectangular cell 160, as follows: Figure 12 As shown, the shaded area is the first rectangular unit 160, and the line segment shown in bold is the first broken edge 161. The first broken edge 161 extends along the first coordinate direction, and its length is between F1 and F2nm, with both endpoints being equally divided breakpoints 102.

[0126] The first rectangular unit 160 generated by the above rules can accurately locate the geometric features of the layout pattern in the first coordinate direction, and is suitable for detecting process defects of metal lines along the first coordinate direction. F1 and F2 are adjustable parameters that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0127] Figure 13 yes Figure 2 The layout graphic shown includes a partial schematic diagram of a second-direction rectangular cell 170, as follows: Figure 13 As shown, the shaded area is the second rectangular unit 170, and the line segment shown in bold is the second broken edge 171. The second broken edge 171 extends along the second coordinate direction, its length is between G1 and G2nm, and both endpoints are evenly divided breakpoints 102.

[0128] The second rectangular unit 170 generated by the above rules can accurately locate the geometric features of the layout pattern in the second coordinate direction, and is suitable for detecting process defects of metal lines along the second coordinate direction. G1 and G2 are adjustable parameters that can be combined with the actual metal layer layout. Figure 10 The design rules and testing accuracy requirements were adjusted.

[0129] In some alternative embodiments, the first directional rectangular unit 160 that is tangent only to the corner rectangular unit 150 and only to one corner rectangular unit 150 can be further classified into the first directional single-corner cluster group 200.

[0130] Figure 14 yes Figure 2 The illustrated layout includes a partial schematic diagram of the first-direction single-angle cluster group 200, as shown below. Figure 14As shown, the black area is the corner rectangle unit 150, and the shaded area is the first-direction rectangle unit 160. Each first-direction rectangle unit 160 is tangent to only one corner rectangle unit 150. The first-direction rectangle units 160 with this positional relationship are further classified into the first-direction single-corner cluster group 200.

[0131] In some alternative embodiments, the first directional rectangular unit 160 that is tangent only to the corner rectangular unit 150 and only to two corner rectangular units 150 can be further classified into the first directional double-corner cluster group 210.

[0132] Figure 15 yes Figure 2 The diagram shown contains a partial schematic of the second-direction bi-fold clustering group, such as... Figure 15 As shown, the black area is the corner rectangle unit 150, and the shaded area is the first-direction rectangle unit 160. Each first-direction rectangle unit 160 is only tangent to two corner rectangle units 150. The first-direction rectangle units 160 with this positional relationship are further classified into the first-direction double-corner cluster group 210.

[0133] In some alternative embodiments, the first directional rectangular unit 160 that is tangent only to the end rectangular unit 120 and only to one end rectangular unit 120 can be further classified into the first directional single-terminal cluster group 220.

[0134] Figure 16 yes Figure 2 The layout diagram shown includes a partial schematic of the first-direction single-terminal cluster group 220, as shown below. Figure 16 As shown, the black area is the end rectangular unit 120, and the shaded area is the first directional rectangular unit 160. Each first directional rectangular unit 160 is tangent to only one end rectangular unit 120. The first directional rectangular units 160 with this positional relationship are further classified into the first directional single-end cluster group 220.

[0135] In some alternative embodiments, the first directional rectangular unit 160 that is tangent only to the end rectangular unit 120 and only to two end rectangular units 120 can be further classified into a first directional dual-terminal cluster group 230.

[0136] Figure 17 yes Figure 2 The layout diagram shown includes a partial schematic of the first-direction two-terminal cluster group 230, as follows: Figure 17 As shown, the black area is the end rectangular unit 120, and the shaded area is the first directional rectangular unit 160. Each first directional rectangular unit 160 is only tangent to two end rectangular units 120. The first directional rectangular units 160 with this positional relationship are further classified into the first directional double-end cluster group 230.

[0137] It needs to be clarified that the "tangency" mentioned above specifically refers to the relationship between the first horizontal rectangular unit and other types of rectangular units, not the relationship between the first horizontal rectangular units themselves. Figure 16 For example, although the two middle first-direction rectangular units 160 are adjacent to each other, each of them is only tangent to one end rectangular unit 120, which satisfies the rule requirement of "only tangent to one end rectangular unit 120".

[0138] In some optional embodiments, for the second directional rectangular unit 170, if there are gaps between its two sides and other types of rectangular units in the first coordinate direction, and the gap width on both sides is less than a preset width, then the second directional rectangular unit 170 is classified into the second directional dense cluster group 240.

[0139] Figure 18 yes Figure 2 The layout diagram shown includes a partial schematic of the second-direction dense cluster group 240, as shown below. Figure 18 As shown, the shaded area represents the second-direction rectangular unit 170. L1 represents the left and right width of the second-direction rectangular unit 170 itself, and L2 represents the spacing between its left and right sides and other types of rectangular units. When L2 is less than 4 times L1 (here, 4 times L1 is the preset width), this type of second-direction rectangular unit 170 is classified into the second-direction dense cluster group 240.

[0140] In some optional embodiments, for the second directional rectangular unit 170, if there are gaps between its two sides and other types of rectangular units in the first coordinate direction, and the gap width on both sides is greater than or equal to a preset width, then the second directional rectangular unit 170 is classified into the second directional sparse cluster group 250.

[0141] Figure 19 yes Figure 2 The layout diagram shown includes a partial schematic of the second-direction sparse cluster group 250, as follows: Figure 19 As shown, the shaded area represents the second rectangular unit. L1 represents the left and right width of the second rectangular unit 170 itself, and L2 represents the spacing between its left and right sides and other types of rectangular units. When L2 is greater than or equal to 4 times L1 (here, 4 times L1 is the preset width), this type of second rectangular unit 170 is classified into the second sparse cluster group 250.

[0142] It should be clarified that, in the first coordinate direction, if the spacing width between the second rectangular unit 170 and other types of rectangular units on one side is greater than the preset width, and the spacing width on the other side is less than the preset width, then the second rectangular unit 170 will be classified as the second sparse cluster group 250.

[0143] In some alternative embodiments, the branch rectangle unit 130 containing two branch breaks 131 along the first coordinate direction can be further classified into a first branch cluster group 260.

[0144] In some alternative embodiments, the branch rectangular unit 130 containing two branch breaks 131 along the second coordinate direction can be further classified into a second-direction branch cluster group 270.

[0145] Figure 20 yes Figure 2 The diagram shown includes partial schematics of the first branching cluster group 260 and the second branching cluster group 270. The shaded and gray areas represent branching rectangular units 130. The thick solid lines represent branch breaks 131 along the first coordinate direction, and the thick dashed lines represent branch breaks 131 along the second coordinate direction. For a branching rectangular unit 130 in the shaded area, if one set of opposite sides coincides with two branch breaks 131 along the first coordinate direction, and the length of this set of opposite sides is equal to the length of the branch breaks 131, it is classified into the first branching cluster group 260. For a branching rectangular unit 130 in the gray area, if one set of opposite sides overlaps with two branch breaks 131 along the second coordinate direction, and the length of this set of opposite sides is equal to the length of the branch breaks 131, it is classified into the second branching cluster group 270.

[0146] In some alternative embodiments, the folded rectangular units 150 that are tangent to or overlap with two rectangular units can be further classified into double-adjacent folded cluster groups 280.

[0147] Figure 21 yes Figure 2 The diagram shown includes a partial schematic of a double-adjacent corner cluster group 280. The rectangular units in the gray area include, but are not limited to, a set consisting of a first-direction rectangular unit 160, a second-direction rectangular unit 170, an end rectangular unit 120, a branch rectangular unit 130, a fragile rectangular unit 140, and an independent rectangular unit 110. This set forms a pattern called a channel. The corner rectangular units 150 in the shaded area are tangent to or overlap with two channels. These corner rectangular units 150 are classified as double-adjacent corner cluster group 280.

[0148] In some alternative embodiments, the folded rectangular units 150 that are tangent to or overlap with the three rectangular units can be classified into a three-adjacent folded cluster group 290.

[0149] Figure 22 yes Figure 2 The layout diagram shown includes a partial schematic of the three-adjacent corner cluster group 290, as follows: Figure 22As shown, the rectangular units in the gray area of ​​the figure include, but are not limited to, a set consisting of first-direction rectangular units 160, second-direction rectangular units 170, end rectangular units 120, branch rectangular units 130, fragile rectangular units 140, and independent rectangular units 110. The graphic formed by this set is called a channel. The folded rectangular units 150 in the shaded area are tangent to or overlap with three channels. These folded rectangular units 150 are classified as three-adjacent folded cluster group 290.

[0150] It is understood that this invention achieves the modification of metal layer plates. Figure 10 The alignment and segmentation of the solid region divides the solid region into various rectangular units, including independent rectangular units 110, end rectangular units 120, branch rectangular units 130, fragile rectangular units 140, corner rectangular units 150, first-direction rectangular units 160, and second-direction rectangular units 170. Units of the same type share design commonalities, while units of different types exhibit significant geometric differences. This segmentation decomposes complex patterns into smaller, more comparable parts. Based on the unit classification, similarity is further explored, grouping rectangular units with similar OPE (Optical Proximity Effect) results into the same cluster. This ensures that the pattern structures within each cluster are similar and that the data volume is sufficient, providing homogeneous data for subsequent statistical analysis and improving the accuracy and reliability of defect detection.

[0151] This invention uses a metal layer plate Figure 10 The geometric properties of the design graphic are determined by dividing the layout into rectangular units as the smallest unit. Based on the geometric information and relative positional relationships of these rectangular units, units with similar OPE results are clustered. This process is crucial for subsequent alignment of rectangular units with SEM or optical image contours and for automatically identifying defect contours. When statistically analyzing image contour information at the rectangular unit level, the statistical values ​​of defect contours differ significantly from those of units within the same cluster, thus avoiding errors in subjective human judgment.

[0152] Specifically, the technical advantages of the present invention are reflected in the following aspects:

[0153] (1) Expansion of detection range: By clustering rectangular units with similar OPE results, statistical analysis is used to feed back outlier contour images. The detection process does not need to rely on fixed reference patterns, breaking through the limitations of traditional methods on periodic or repetitive pattern regions. It can be widely applied to any region on the wafer, whether it is a regular unit pattern or a complex logic pattern, which significantly expands the scope of detection.

[0154] (2) Automated optimization of the detection process: The introduction of adaptive detection thresholds replaces the manual parameter settings for different pattern types and process conditions in traditional methods, avoiding cumbersome operation procedures and human error settings, improving the efficiency and accuracy of the detection process, and reducing the cost of manual intervention.

[0155] (3) Improved defect detection accuracy: Defect detection is achieved by comparing the similarity of similar patterns, rather than relying on absolute reference images. This can effectively tolerate the impact of manufacturing process fluctuations, reduce false defect alarms caused by process changes, significantly reduce the false alarm rate and improve the accuracy of detection.

[0156] (4) Enhanced detection sensitivity: The complex pattern is divided into smaller basic components (rectangular units) to form multiple comparable target units. Through statistical analysis of the small components, the ability to capture subtle defects is effectively improved, especially in the unit area where smaller abnormal structures can be detected.

[0157] (5) Compatibility of the testing platform: It is not limited to specific types of testing equipment (such as scanning electron microscopes (SEM) or optical imaging equipment), and can be compatible with a variety of testing platforms to meet the diverse testing needs of enterprises and comprehensively improve the flexibility and adaptability of the overall testing capabilities.

[0158] (6) Data-driven process optimization: It can establish a basic pattern component database containing design layout and defect information, providing solid data support for subsequent defect analysis, process improvement and intelligent process control, and promoting the development of the inspection process towards automation and data.

[0159] This invention also provides a computer program product 30. Figure 23 This is a schematic diagram of a computer program product 30 according to an embodiment of the present invention, as shown below. Figure 23 As shown, the computer program product 30 includes a machine-executable program 31, which, when executed by a processor, implements the preprocessing method for wafer defect detection according to any of the above embodiments.

[0160] The machine-executable program 31 used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages ​​and procedural programming languages. The machine-executable program 31 may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can connect to the user's computer via any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it can connect to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, to perform aspects of the invention, electronic circuits including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) can execute computer-readable program instructions to personalize the electronic circuits by utilizing state information of computer-readable program instructions.

[0161] For the purposes of this embodiment, computer program product 10 is a related product that includes machine executable program 31.

[0162] Therefore, those skilled in the art should recognize that although numerous exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications conforming to the principles of the present invention can be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Thus, the scope of the present invention should be understood and construed as covering all such other variations or modifications.

Claims

1. A preprocessing method for wafer defect detection, comprising: Obtain the metal layer layout of the wafer to be inspected; Extract the entity region to be processed from the layout pattern of the metal layer layout; The boundaries of the physical region are broken to generate multiple broken edges; For different types of broken edges, corresponding rectangular units are generated to divide the solid region; The generated rectangular cells are input into the wafer defect detection database; In the metal layer layout, the partial boundary of the solid region along the first coordinate direction and the partial boundary along the second coordinate direction intersect to form a corner vertex. The corner vertex is divided into concave corner vertex and convex corner vertex according to the concave and convex characteristics of the boundary at its location. The steps for breaking the boundary of the physical region include: Identify each concave corner vertex in the solid region, and for each concave corner vertex, project it onto its opposite side along the mutually perpendicular first coordinate direction and second coordinate direction to obtain the projection breakpoint; For any two opposite sides of any set of boundaries along the first coordinate direction and any two opposite sides of any set of boundaries along the second coordinate direction in the entity region, multiple pairs of evenly divided breakpoints are configured according to a preset interval distance. The boundary of the solid region is divided into multiple broken edges by passing through each concave corner vertex and its corresponding projection breakpoint, as well as through each pair of equally divided breakpoints.

2. The preprocessing method for wafer defect detection according to claim 1, wherein, The types of broken edges include evenly divided broken edges, which refer to broken edges whose two endpoints are both evenly divided broken points; The steps for generating corresponding rectangular elements for different types of broken edges include: Select two corresponding equally divided edges from the same set of boundaries of the entity region; The two equally divided broken edges are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two equally divided broken edges is used as another set of parallel edges, thereby generating an equally divided rectangular unit.

3. The preprocessing method for wafer defect detection according to claim 2, wherein, The types of broken edges include angled broken edges, which refer to broken edges with one endpoint being the vertex of the concave corner or the vertex of the convex corner, and the other endpoint being the evenly divided break point; The steps for generating corresponding rectangular elements for different types of broken edges include: Identify the included angle formed by a pair of broken edges at the same corner in the physical region, wherein the pair of broken edges have a common concave or convex vertex. Based on the positional relationship of each included angle, a folded rectangle unit is generated, and the folded rectangle unit completely covers all the included angles at the same fold.

4. The preprocessing method for wafer defect detection according to claim 3, wherein, The types of broken edges include end broken edges, which refer to broken edges where both endpoints are vertices of the outward convex corner; The steps for generating corresponding rectangular elements for different types of broken edges include: Identify the length of the broken end edge, generate two parallel edges that are parallel to and of equal length to the broken end edge, and make the two generated parallel edges symmetrically distributed about the broken end edge; The line connecting the corresponding endpoints of the two generated parallel sides is used as another set of parallel sides, thereby generating the end rectangular unit.

5. The preprocessing method for wafer defect detection according to claim 4, wherein, The evenly divided rectangular unit includes a first rectangular unit whose length direction is along the first coordinate direction, and a second rectangular unit whose length direction is along the second coordinate direction; After the step of generating corresponding rectangular elements for different types of broken edges, the following steps are also included: The first directional rectangular units that are tangent only to the terminal rectangular units, and tangent only to one of the terminal rectangular units, are classified into the first directional single-terminal cluster group; The first directional rectangular units that are tangent only to the terminal rectangular units, and only to two of the terminal rectangular units, are classified into the first directional double-terminal cluster group. The first directional rectangular unit that is tangent only to the corner rectangular unit and only to one of the corner rectangular units is classified into the first directional single-corner cluster group; The first directional rectangular units that are tangent only to the corner rectangular units, and only to two of the corner rectangular units, are classified into the first directional double-corner cluster group.

6. The preprocessing method for wafer defect detection according to claim 5, wherein, After the step of generating corresponding rectangular elements for different types of broken edges, the following steps are also included: For the second directional rectangular unit, if there are gaps between it and other types of rectangular units on both sides in the first coordinate direction, and the gap width on both sides is less than the preset width, then the second directional rectangular unit is classified into the second directional dense cluster group; if the gap width on both sides is greater than or equal to the preset width, then the second directional rectangular unit is classified into the second directional sparse cluster group.

7. The preprocessing method for wafer defect detection according to claim 5, wherein, After the step of generating a folded rectangular unit for the folded edge, the method further includes: The folded rectangular units that are tangent to or overlap with the two rectangular units are classified into a double-adjacent folded corner cluster group; The folded rectangular units that are tangent to or overlap with the three rectangular units are classified into three adjacent folded cluster groups.

8. The preprocessing method for wafer defect detection according to claim 1, wherein, The types of broken edges include branched broken edges, which are broken edges with one endpoint being the vertex of the concave angle and the other endpoint being the vertex of the convex angle. The steps for generating corresponding rectangular elements for different types of broken edges include: Select two corresponding branch breaks from the same set of boundaries of the entity region; Two of the branch breaks are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two branch breaks is used as another set of parallel edges, thereby generating a branch rectangular unit.

9. The preprocessing method for wafer defect detection according to claim 8, wherein, After the step of generating branch rectangular units for the branch break edge, the method further includes: The branch rectangle cells containing two broken edges along the first coordinate direction are classified into the first directional branch cluster group; Branch rectangular cells containing two broken branches along the second coordinate direction are classified into the second-direction branch cluster group.

10. The preprocessing method for wafer defect detection according to claim 1, wherein, The types of broken edges include easily broken edges. An easily broken edge is a broken edge with one endpoint being the vertex of the concave angle and the other endpoint being the projection point of the concave angle. The easily broken edges appear in pairs in the same set of boundaries of the solid region. The projection point obtained by projecting from the vertex of the concave angle of one of the easily broken edges along the first coordinate direction or the second coordinate direction to the other easily broken edge is the projection point of the concave angle of the other easily broken edge. The steps for generating corresponding rectangular elements for different types of broken edges include: Select two corresponding easily broken edges from the same set of boundaries of the entity region; Two of the easily broken edges are used as a set of parallel edges, and the line connecting the corresponding endpoints of the two easily broken edges is used as another set of parallel edges, thereby generating an easily broken rectangular unit.

11. A computer program product comprising a computer program, which, when executed by a processor, is used to implement the preprocessing method for wafer defect detection according to any one of claims 1-10.

Images