Method for de-duplication of photolithography layout and related product

Abstract

The application provides a photolithography layout deduplication method and related products. The photolithography layout deduplication method comprises determining a plurality of target pattern regions in a photolithography layout; clustering the plurality of target pattern regions based on relative distances between the target pattern regions to obtain a plurality of candidate pattern clusters; clustering the plurality of candidate pattern clusters based on pattern similarity of the candidate pattern clusters to obtain a plurality of target pattern classes; determining one of the plurality of candidate pattern clusters in each target pattern class as a representative pattern cluster, and removing the remaining candidate pattern clusters in the photolithography layout except the representative pattern clusters. The deduplication method improves the sensitivity of the clustering result to process constraints, and improves the clustering efficiency and deduplication accuracy.

Description

Technical Field

[0001] This invention relates to the field of integrated circuit technology, and in particular to a method for deduplication of photolithography patterns, a computer-readable storage medium, a computer program product, and a computer device. Background Technology

[0002] As integrated circuit process nodes continue to shrink, layout complexity is increasing exponentially. Layout files containing billions of graphic units pose significant challenges to the storage, processing, and verification of electronic design automation tools. Traditional full layout processing methods are too time-consuming, and redundant calculations of repetitive graphic units (such as memory arrays and standard cells) severely restrict layout processing efficiency.

[0003] While various layout deduplication methods have been proposed, they either suffer from low efficiency or low accuracy. The industry is eager for a layout deduplication method that balances both efficiency and accuracy. Summary of the Invention

[0004] One object of the present invention is to provide a method for deduplication of photolithographic patterns, a computer-readable storage medium, a computer program product, and a computer device to improve the deduplication efficiency of patterns while maintaining high deduplication accuracy.

[0005] Specifically, according to one aspect of the present invention, the present invention provides a method for deduplication of photolithographic patterns, comprising: Multiple target graphic regions are identified in the photolithography pattern; Based on the relative distance between each of the target graphic regions, multiple target graphic regions are clustered to obtain multiple candidate graphic clusters; Based on the graphic similarity of each candidate graphic cluster, multiple candidate graphic clusters are clustered to obtain multiple target graphic classes; One of the candidate graphic clusters in each target graphic class is determined as a representative graphic cluster, and the remaining candidate graphic clusters in the photolithography pattern, excluding each of the representative graphic clusters, are removed.

[0006] Optionally, the photolithographic pattern is used for design rule checking, and the determination of multiple target graphic regions in the photolithographic pattern includes: The photolithography pattern was subjected to design rule checks, resulting in multiple hotspot regions; Each of the aforementioned hotspot areas is treated as a target graphic area.

[0007] After the step of removing the remaining candidate pattern clusters in the photolithography pattern except for each of the representative pattern clusters, the method further includes: The new photolithography pattern is subjected to design rule checks.

[0008] Optionally, clustering the multiple target graphic regions based on the relative distance between each target graphic region includes: Obtain the geometric center of each target graphic region and use each geometric center as an anchor point; Using a defined spatial indexing algorithm, multiple anchor points are clustered to obtain multiple anchor point clusters; The cluster center point of each anchor point cluster is obtained, and the region within a preset range centered on the cluster center point is taken as the candidate graphic cluster.

[0009] Optionally, the spatial indexing algorithm includes a KD-tree index or a quadtree index; The spatial indexing algorithm is configured with clustering parameters, including a query distance threshold and a cluster density threshold, and the preset range is greater than three times the query distance threshold.

[0010] Optionally, the step of clustering multiple candidate graphic clusters based on the graphic similarity of each candidate graphic cluster includes: Obtain the graphic feature descriptors for each of the candidate graphic clusters; Using the aforementioned graphic feature descriptor and a defined graphic matching algorithm, multiple candidate graphic clusters are clustered to obtain multiple target graphic classes.

[0011] Optionally, the set graph matching algorithm is a locality-sensitive hash algorithm, and the graph feature descriptor is the graph hash vector corresponding to the candidate graph cluster.

[0012] Optionally, removing the remaining candidate pattern clusters in the photolithography pattern, excluding each of the representative pattern clusters, includes: Obtain a clustering mapping table that represents the clustering results for each of the target graph classes; Delete the remaining candidate pattern clusters in the photolithography pattern, except for each of the representative pattern clusters. Each of the representative graphic clusters is treated as a new instance, and the hierarchical structure of the lithography pattern is updated in conjunction with the clustering mapping table.

[0013] According to another aspect of the present invention, a computer-readable storage medium is also provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the deduplication method for photolithographic patterns described above.

[0014] According to another aspect of the present invention, a computer program product is also provided, comprising a computer program that, when executed by a processor, implements the steps of the deduplication method for photolithographic patterns described above.

[0015] According to another aspect of the present invention, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the deduplication method for photolithography patterns described above.

[0016] The deduplication method for photolithography patterns of the present invention improves the clustering accuracy by selecting key local regions in the photolithography pattern that have a significant impact on deduplication accuracy as target graphic regions for clustering operations. This enhances the sensitivity of subsequent clustering results to process constraints. Furthermore, by performing geometric clustering on the target graphic regions and then performing graphic similarity clustering on the geometric clustering results, this dual clustering improves both clustering efficiency and deduplication accuracy.

[0017] 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

[0018] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a flowchart illustrating a method for deduplicating photolithographic patterns according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the process of obtaining a target graphic region using a deduplication method according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the geometric clustering process of a deduplication method according to an embodiment of the present invention; Figure 4 This is a partial schematic diagram of a deduplication method according to an embodiment of the present invention, which performs geometric clustering of multiple anchor points in a layout. Figure 5 This is a schematic diagram of the process of performing graphical pattern clustering using a deduplication method according to an embodiment of the present invention; Figure 6 This is a schematic flowchart of a deduplication method for layout deduplication according to an embodiment of the present invention; Figure 7 This is a flowchart illustrating a deduplication method according to another embodiment of the present invention; Figure 8 This is a schematic diagram of a computer program product according to an embodiment of the present invention; Figure 9 This is a schematic diagram of a computer-readable storage medium according to an embodiment of the present invention; and Figure 10This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0019] Electronic Design Automation (EDA) is a design approach that utilizes computer-aided design software to complete the functional design, synthesis, verification, and physical design processes of large-scale integrated circuit chips. In the physical verification process, EDA tools perform checks and verifications on the entire layout, such as Design Rule Check (DRC) and Layout Versus Schematics (LVS).

[0020] Direct Detection (DRC) is typically based on a lithography imaging model. It examines and searches for areas with poor imaging on a lithography pattern that has been processed using optical resolution enhancement techniques. These areas may present manufacturability issues in subsequent lithography processes. Specific aspects of DRC may include critical dimension deformation checks, open circuit and bridging verification, line end deformation checks, line end cap deformation checks, via coverage checks, auxiliary pattern imaging checks, and process window checks.

[0021] Large-scale integrated circuit layouts typically contain a large number of repetitive graphic units (such as memory arrays and standard cells). Directly performing physical verification on the entire layout would generate a large amount of redundant computation, severely limiting verification efficiency. Existing layout clustering and deduplication methods mostly rely on global scanning or single feature matching, and then use clustering techniques to identify and merge identical or similar layout regions, making it difficult to balance deduplication efficiency and accuracy.

[0022] The purpose of the deduplication method for the photolithographic layout in this embodiment is to improve the deduplication efficiency of the layout while maintaining high deduplication accuracy.

[0023] Figure 1 This is a schematic flowchart of a method for deduplicating photolithographic patterns according to an embodiment of the present invention. The method generally includes: S100, which determines multiple target graphic regions in the photolithography pattern; S200: Based on the relative distance between each target graphic region, multiple target graphic regions are clustered to obtain multiple candidate graphic clusters; S300: Based on the graphic similarity of each candidate graphic cluster, multiple candidate graphic clusters are clustered to obtain multiple target graphic classes; S400: Among the multiple candidate graphic clusters of each target graphic class, one is determined as the representative graphic cluster, and the remaining candidate graphic clusters in the photolithography pattern are removed except for the representative graphic clusters.

[0024] Existing layout clustering and deduplication methods primarily calculate accuracy based on all graphics within the entire layout. Deduplication accuracy reflects the quality of clustering and deduplication, typically quantified using metrics such as recall and false merging rate. Recall reflects the proportion of duplicate graphics successfully identified and merged, while false merging rate reflects the proportion of non-duplicate graphics incorrectly identified as duplicates and merged. When the inventors decomposed the deduplication accuracy obtained from existing layout clustering and deduplication methods into various blocks of the entire layout, they found that the deduplication accuracy was not uniformly distributed across the entire layout. In some local areas, the deduplication accuracy was significantly lower than in other areas. In other words, the deduplication accuracy in certain local areas lowered the overall deduplication accuracy of the entire layout. After careful analysis, the inventors discovered that these local areas were mostly high-density wiring areas and densely populated device areas, often located at the edge of process constraints. Existing layout clustering and deduplication methods showed poor sensitivity to these process-constrained areas. Further research by the inventors revealed that by focusing clustering on these local regions, the sensitivity of the clustering results to process constraints can be significantly improved with a relatively small computational burden, thereby enhancing the accuracy of clustering and deduplication. Based on this inventive concept, the inventors propose the following technical solution: First, all target pattern regions are selected from the photolithography layout. The layout can be in GDSII / OASIS format, and the target pattern regions are the local areas with low deduplication accuracy mentioned above. Specifically, the target pattern regions can be local critical areas in the layout, such as high-density wiring areas or device-dense areas. These local critical areas can be determined through pre-marking (e.g., marking device-dense areas during the layout design stage), experience, or automatic identification algorithms (e.g., automatically identifying high-density wiring areas and device-dense areas). Alternatively, they can be determined based on actual inspection, such as by pre-running a fast scan to identify critical areas for physical verification projects in the layout (e.g., high-risk areas for violation of detection rules in DRC). By selecting target pattern regions and then focusing on clustering them, the sensitivity of the clustering results to process constraints can be greatly improved.

[0025] It is important to understand that for the same photolithography layout, the target graphic regions selected may differ for different physical verification projects. For example, a localized area in the layout might be selected as a target graphic region for DRC (Dual Rectification) but not for LVS (Low Performance System). Those skilled in the art should selectively select the corresponding target graphic regions based on the physical verification project to be performed after deduplication of the layout.

[0026] Next, geometric clustering based on relative distance is performed on adjacent target graphic regions. Local key regions are often fragmented local graphics. Taking the physical verification project DRC as an example, its hotspot region might just be a corner of a polygon, or two closely spaced edges of adjacent polygons. Two or more local key regions may also intersect. If we directly perform similarity clustering and deduplication on all target graphic regions in the entire map, the computational load is large, especially for maps with over a billion graphic units, leading to low clustering efficiency.

[0027] In this embodiment, by performing geometric clustering on adjacent target graphic regions, multiple adjacent target graphic regions are focused into a relatively large area to form candidate graphic clusters. This can significantly reduce the number of blocks for subsequent graphic similarity clustering and improve the efficiency of subsequent clustering operations.

[0028] Next, clustering (also known as graph pattern clustering) is performed on each candidate graph cluster based on graph similarity. Specific clustering methods can include, for example, K-means clustering or locality-sensitive hashing. These are not specified here. Through graph pattern clustering, multiple candidate graph clusters with the highest degree of similarity or identical graphs can be clustered into a single target graph class, ultimately resulting in multiple target graph classes.

[0029] Next, image deduplication is performed. Specifically, a candidate image cluster is selected from each target image class as the representative image cluster, and other candidate image clusters are marked as duplicate clusters, resulting in a clustering mapping table labeled with the representative image cluster and duplicate clusters. Then, the candidate image clusters marked as duplicate clusters are deleted from the layout, resulting in a simplified layout file. Compared to the original layout, the deduplicated layout significantly reduces redundant image data, lowers storage costs, and significantly improves the efficiency of subsequent layout verification, data analysis, and other operations.

[0030] The deduplication method for photolithography patterns in this embodiment improves clustering accuracy by selecting key local regions that significantly impact deduplication accuracy as target regions for clustering operations. This enhances the sensitivity of subsequent clustering results to process constraints. Geometric clustering of the target regions, followed by pattern similarity clustering of the geometric clustering results, provides dual clustering, improving both clustering efficiency and deduplication accuracy. In practical applications, the deduplication method in this embodiment achieves a false merging rate of less than 0.1%, maintaining high deduplication accuracy, and its processing speed is 3-5 times faster than traditional clustering and deduplication methods.

[0031] In some embodiments of the deduplication method for photolithographic patterns of the present invention, such as Figure 2 As shown, the photolithography layout is used for design rule checks, and multiple target graphic regions are identified within the photolithography layout, including: S111, the first design rule check is performed on the photolithography pattern, and multiple hot spots are obtained; S113, treat each hotspot area as a target graphic area.

[0032] After removing candidate pattern clusters from the photolithographic pattern except for the representative pattern clusters, the process also includes: A second design rule check is performed on the new photolithography pattern.

[0033] In some embodiments, the accuracy of the first design rule check is lower than that of the second design rule check. The first design rule check is used to perform a quick, coarse scan of the layout, initially identifying hotspots in the layout. The check is fast, but the detection rate of hotspots is relatively low.

[0034] In this embodiment, the distribution of hotspot areas in the layout can be quickly obtained through the first design rule check. By focusing on clustering hotspot areas, the sensitivity of the clustering results to process constraints can be greatly improved, enhancing the accuracy of subsequent clustering and deduplication. After deduplication, a second design rule check is used to perform a fine scan of the new layout, accurately retrieving all hotspot areas in a short time.

[0035] In some embodiments of the deduplication method for photolithographic patterns of the present invention, such as Figure 3 As shown, based on the relative distances between each target graphic region, multiple target graphic regions are clustered, including: S211, Obtain the geometric center of each target graphic region, and use each geometric center as an anchor point; S213, using a defined spatial indexing algorithm, cluster multiple anchor points to obtain multiple anchor point clusters; S215, obtain the cluster center point of each anchor point cluster respectively, and take the area within a preset range centered on the cluster center point as the candidate graphic cluster.

[0036] In this embodiment, an anchor point is placed at the geometric center of each target graphic region, and the anchor point represents the corresponding target graphic region. In this way, when performing geometric clustering, region clustering is transformed into point clustering. Based on the coordinates of each anchor point, the relative distance between the anchor points can be quickly calculated, and clusters can be performed on nearby anchor points, thereby quickly completing the merging of adjacent target graphic regions.

[0037] Taking map layout for DRC as an example, an anchor point generation module can be set up in the EDA tool. This module is used to quickly filter hotspot areas in the map layout using a fast DRC engine, calculate the geometric center of each hotspot area, and place an anchor point at each geometric center. For each anchor point, its attributes can be labeled, such as anchor point coordinates, the layer it belongs to, and the DRC violation risk level (e.g., high-risk, medium-risk). In some embodiments, the anchor point generation module also sets a density upper limit threshold to reasonably control the maximum number of anchor points per unit area when placing anchor points, avoiding excessive concentration of anchor points.

[0038] In this embodiment, a geometric clustering engine can be set in the EDA tool. The geometric clustering engine is used to cluster similar anchor points by setting a spatial indexing algorithm. The spatial indexing algorithm can be, for example, a KD-tree (k-dimensional tree), a quadtree / octree, a hierarchical bounding box tree, etc., and is not required here.

[0039] Taking the KD-tree indexing algorithm as an example, the algorithm can set clustering parameters such as query distance threshold and cluster density threshold. Taking the layout for DRC as an example, the query distance threshold is determined by the minimum process spacing of the layout. For example, the query distance threshold can be set to 3 times the minimum process spacing to include repeating array target patterns with high placement density (such as memory arrays) into candidate pattern clusters as much as possible.

[0040] For example, the main steps of the KD-tree indexing algorithm are as follows: First, construct the KD-tree index structure for all anchor points. Next, for each anchor point P... i Query its radius R g The set of nearest anchor points N(P) within the (i.e., the query distance threshold) i If N(P) i If the set is not empty, then all anchor points in the set are merged into a set of several anchor points G. m Cluster center C m The coordinates are the centroids of the coordinates of all anchor points within the set. Iterate through all anchor points until all anchor points have been processed, outputting the anchor point family set {G1, G2, ..., G...}. M After obtaining the set of anchor point clusters, for each anchor point cluster, take its cluster center C. m Centered on, the extraction radius is R p All graphics within the window (i.e., the preset range) are selected as candidate graphic clusters.

[0041] Please see Figure 4 , Figure 4 This is a partial schematic diagram of geometric clustering of multiple anchor points on the map. In the diagram, the blue fill represents the map shape, the red dots represent the placed anchor points, and the black dots represent the cluster centers after geometric clustering.

[0042] Preset range R p It can be set as the query distance threshold R g 2 to 10 times that of R p / R g The smaller the value, the fewer the number of target graphics within a single candidate graphic cluster. Therefore, it is easier to match the same or similar candidate graphic clusters in subsequent graphic pattern clustering, but it may fragment the target graphics in the repeated array; R p / R g A larger R value, while reducing the aforementioned fragmentation, also results in a larger number of target graphics within a single candidate graphic cluster. Therefore, it becomes more difficult to match identical or similar candidate graphic clusters in subsequent graphic pattern clustering. Those skilled in the art can select and set R as needed. p / R g The range allows the size of candidate graph clusters to be adapted to the size of memory arrays and standard cells in the layout to be processed. For example, the query distance threshold R... g Set to 3 times the minimum process spacing, preset range R p It can then be set to 10 to 20 times the minimum spacing of the process.

[0043] In some embodiments of the deduplication method for photolithographic patterns of the present invention, such as Figure 5 As shown, based on the graphic similarity of each candidate graphic cluster, multiple candidate graphic clusters are clustered, including: S311, Obtain the graphic feature descriptors for each candidate graphic cluster; S313 uses graphic feature descriptors and a set graphic matching algorithm to cluster multiple candidate graphic clusters to obtain multiple target graphic classes.

[0044] In this embodiment, a graph pattern analyzer can be set in the EDA tool. The graph pattern analyzer is used to perform graph pattern clustering on each candidate graph cluster by setting a graph matching algorithm. The graph matching algorithm can be K-means clustering, Locality Sensitive Hashing (LSH), etc. The following explanation uses setting the graph matching algorithm as LSH as an example.

[0045] LSH (Approximate Nearest Neighbor Search) is an efficient algorithm for processing massive amounts of data. Its main idea is to reduce high-dimensional data to low-dimensionality while preserving data similarity as much as possible. Specifically, in this embodiment, image features are first extracted from each candidate graphic cluster to obtain feature vectors representing key points and feature information in the candidate graphic cluster image. Next, the feature vectors of each candidate graphic cluster are mapped into a high-dimensional space, converting them into graphic hash vectors to obtain their graphic feature descriptors. Then, the LSH algorithm is used to perform similarity retrieval on the graphic hash vectors of each candidate graphic cluster, and all candidate graphic clusters are clustered based on the retrieval results.

[0046] In similarity retrieval, LSH buckets can be constructed using the following formula: , Where h is the hash function, F m Let be the extracted feature vector, 'a' be a random vector, 'b' be a random offset, and 'w' be the bucket width. By performing multiple random projections and calculating the hash value corresponding to each candidate graphic cluster, candidate graphic clusters with the same hash value are determined to have the same graphic pattern and clustered into a target graphic class.

[0047] In some embodiments of the deduplication method for photolithographic patterns of the present invention, such as Figure 6 As shown, the remaining candidate pattern clusters in the photolithography pattern, excluding the representative pattern clusters, are removed, including: S411, Obtain the clustering mapping table that serves as the clustering result for each target graph class; S413, Delete the remaining candidate pattern clusters in the photolithography pattern except for the representative pattern clusters; S415, each representative graphic cluster is treated as a new instance, and the hierarchical structure of the lithography layout is updated in conjunction with the clustering mapping table.

[0048] Photolithography layouts typically employ hierarchical design, containing a tree-like hierarchical structure. In this embodiment, a layout output module can be set up in the EDA tool. This module is used to deduplicate the layout based on the graph pattern clustering results and output a simplified layout. When dedupplicating the hierarchical design layout, the layout output module retains a representative graph cluster as a reference instance for each target graph class and deletes the graphs corresponding to instances marked as duplicate candidate graph clusters in the layout. Next, the tree-like hierarchical structure of the layout is updated using a clustering mapping table to ensure that the electrical connectivity of structures referenced through cell instantiation remains unchanged. Afterward, a simplified layout file and an LSH clustering report can be output. The clustering report may include information such as the clustering mapping table, deduplication ratio, and compression rate.

[0049] In some embodiments of the deduplication method for photolithographic patterns of the present invention, such as Figure 7 As shown, the methods for deduplication of photolithographic patterns include: S511, loads the photolithography layout and DRC design rules (such as spacing, width, density, etc.). S512 runs a fast DRC scan to identify high-risk areas for violations; S513, calculate the geometric center of each risk area and place an anchor point at the geometric center; S514, construct the spatial index structure for all anchor points; S515, using a set spatial indexing algorithm to perform geometric clustering of all anchor points, resulting in multiple anchor point clusters; S516, Calculate the center point of each anchor point cluster, obtain the graphic within a preset range centered on each center point, and generate candidate graphic clusters; S517, apply locality-sensitive hashing to perform graph pattern clustering on each candidate graph cluster to obtain multiple target graph classes and generate a clustering mapping table; S518, for each target graphic class, retain a representative instance in the layout and delete its duplicate instances; S519 updates the layout hierarchy and outputs a simplified layout file and clustering report.

[0050] The flowchart provided in this embodiment is not intended to indicate that the operations of the method will be performed in any particular order, or that all operations of the method are included in every case. Furthermore, the method may include additional operations. Within the scope of the technical concept provided by the method in this embodiment, additional variations can be made to the above method.

[0051] It should be understood that in some embodiments, the components may be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods may be implemented using software or firmware stored in memory and executed by a suitable instruction execution system.

[0052] This invention also provides a computer program product 10, a computer-readable storage medium 20, and a computer device 30. Figure 8 This is a schematic diagram of a computer program product 10 according to an embodiment of the present invention. Figure 9 This is a schematic diagram of a computer-readable storage medium 20 according to an embodiment of the present invention. Figure 10This is a schematic diagram of a computer device 30 according to an embodiment of the present invention. The computer program product 10 includes a computer program 11, which, when executed by the processor 32, implements the steps of the deduplication method for photolithography patterns described above. A computer-readable storage medium 20 stores the computer program 11 thereon, which, when executed by the processor 32, implements the steps of the deduplication method for photolithography patterns described above. The computer device 30 may include a memory 31, a processor 32, and the computer program 11 stored in the memory 31 and running on the processor 32.

[0053] The computer program 11 used to perform the operations of this invention may be assembly instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state 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 computer program 11 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 may be connected to the user's computer via any type of network, including a Local Area Network (LAN) or Wide Area Network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, to perform aspects of this invention, electronic circuits, including, for example, programmable logic circuits, Field-Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), may execute computer-readable program instructions to personalize the electronic circuits by utilizing state information from computer-readable program instructions.

[0054] For the purposes of this embodiment, computer program product 10 is a related product that includes computer program 11.

[0055] For the purposes of this embodiment, the computer-readable storage medium 20 is a tangible device capable of holding and storing a computer program 11. It can be any device capable of containing, storing, communicating, propagating, or transmitting the computer program 11 for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable storage medium 20 include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanical encoding device, and any suitable combination thereof.

[0056] Computer device 30 can be, for example, a server, desktop computer, laptop computer, tablet computer, or smartphone. In some examples, computer device 30 can be a cloud computing node. Computer device 30 can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., that perform specific tasks or implement specific abstract data types. Computer device 30 can be implemented in a distributed cloud computing environment where tasks are performed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can reside on local or remote computing system storage media, including storage devices.

[0057] Computer device 30 may include a processor 32 adapted to execute stored instructions and a memory 31 that provides temporary storage space for the operation of said instructions during operation. The processor 32 may be a single-core processor, a multi-core processor, a computing cluster, or any other configuration. The memory 31 may include random access memory (RAM), read-only memory, flash memory, or any other suitable storage system.

[0058] Computer device 30 may also include a network adapter / interface and an input / output (I / O) interface. The I / O interface allows external devices that can be connected to the computer device to input and output data. The network adapter / interface provides communication between the computer device and a network, typically represented as a communication network.

[0059] 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 method of de-duplication of photolithography layout, characterized in that, include: Multiple target graphic regions are identified in the photolithography pattern; Based on the relative distance between each of the target graphic regions, multiple target graphic regions are clustered to obtain multiple candidate graphic clusters; Based on the graphic similarity of each candidate graphic cluster, multiple candidate graphic clusters are clustered to obtain multiple target graphic classes; One of the candidate graphic clusters in each target graphic class is determined as a representative graphic cluster, and the remaining candidate graphic clusters in the photolithography pattern, excluding each of the representative graphic clusters, are removed.

2. The deduplication method of claim 1, wherein, The photolithographic pattern is used for design rule checks, and the determination of multiple target graphic regions in the photolithographic pattern includes: The photolithography pattern was subjected to design rule checks, resulting in multiple hotspot regions; Each of the aforementioned hotspot areas is treated as a target graphic area; After the step of removing the remaining candidate pattern clusters in the photolithography pattern except for each of the representative pattern clusters, the method further includes: The new photolithography pattern is subjected to design rule checks.

3. The deduplication method according to claim 1, characterized in that, The clustering of multiple target graphic regions based on the relative distance between each target graphic region includes: Obtain the geometric center of each target graphic region and use each geometric center as an anchor point; Using a defined spatial indexing algorithm, multiple anchor points are clustered to obtain multiple anchor point clusters; The cluster center point of each anchor point cluster is obtained, and the region within a preset range centered on the cluster center point is taken as the candidate graphic cluster.

4. The deduplication method according to claim 3, characterized in that, The specified spatial indexing algorithm includes KD-tree indexing or quadtree indexing; The spatial indexing algorithm is configured with clustering parameters, including a query distance threshold and a cluster density threshold, and the preset range is greater than three times the query distance threshold.

5. The deduplication method according to claim 1, characterized in that, The clustering of multiple candidate graphic clusters based on the graphic similarity of each candidate graphic cluster includes: Obtain the graphic feature descriptors for each of the candidate graphic clusters; Using the aforementioned graphic feature descriptor and a defined graphic matching algorithm, multiple candidate graphic clusters are clustered to obtain multiple target graphic classes.

6. The deduplication method according to claim 5, characterized in that, The set graph matching algorithm is the Locality Sensitive Hash Algorithm, and the graph feature descriptor is the graph hash vector corresponding to the candidate graph cluster.

7. The deduplication method according to claim 1, characterized in that, The removal of the remaining candidate pattern clusters in the photolithography pattern, excluding each of the representative pattern clusters, includes: Obtain a clustering mapping table that represents the clustering results for each of the target graph classes; Delete the remaining candidate pattern clusters in the photolithography pattern, except for each of the representative pattern clusters. Each of the representative graphic clusters is treated as a new instance, and the hierarchical structure of the lithography pattern is updated in conjunction with the clustering mapping table.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the deduplication method for the photolithographic pattern as described in any one of claims 1 to 7.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the deduplication method for the photolithographic pattern as described in any one of claims 1 to 7.

10. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the deduplication method for photolithographic patterns according to any one of claims 1 to 7.

Images