Chip product comparison method, modeling method, device and storage medium

Abstract

The present application provides a chip product comparison method, modeling method, device, and storage medium. In this method, a three-dimensional model of a chip product is obtained by pre-modeling the chip product using a neural network based on the chip product's slice sequence. When comparing chip products, comparison features are obtained in response to user operations. For each chip product, the comparison results corresponding to the comparison features are obtained from its corresponding three-dimensional model, and then the comparison results corresponding to each chip product are displayed. This method eliminates the need for manual image comparison through multiple electronic files and the need for manual processing of large amounts of information, thereby improving the efficiency and accuracy of chip product comparison.

Description

Technical Field

[0001] The present application relates to the field of semiconductors, and in particular to a comparison method, modeling method, device and storage medium for chip products. Background Art

[0002] Chip is a general term for semiconductor component products. In the semiconductor field, chip, also known as integrated circuit (IC), microcircuit (microcircuit), or microchip (microchip), is a way to miniaturize circuits (mainly including semiconductor devices, but also passive components, etc.) in electronics, and is often manufactured on the surface of semiconductor wafers.

[0003] Currently, the comparative analysis method for chip products involves analyzing the chip product to obtain a series of discrete two-dimensional images. Professionals manually mark the images with specific information about relevant parameters. The chip product parameters and images are then manually organized and stored in electronic files. During the analysis and comparison process, the electronic files of the chip products to be compared must be selected and the various parameter values ​​searched within the electronic files for manual comparison.

[0004] However, when comparing multiple chip products, the above solution requires opening multiple electronic files at the same time to compare multiple pictures, and each picture is annotated with a large amount of information, resulting in low comparison efficiency and poor accuracy of manual comparison. Summary of the Invention

[0005] The present application provides a chip product comparison method, modeling method, device and storage medium to improve the efficiency and accuracy of chip product comparison.

[0006] In a first aspect, an embodiment of the present application provides a method for comparing chip products, comprising:

[0007] When a task of comparing at least two chip products is acquired, obtaining a comparison feature in response to a user operation;

[0008] For each chip product, a comparison result corresponding to the comparison feature is obtained from a three-dimensional model corresponding to the chip product; wherein the three-dimensional model corresponding to each chip product is a three-dimensional model reconstructed using a neural network based on a slice sequence of the chip product;

[0009] Displays the comparison results corresponding to each chip product.

[0010] In a specific embodiment, the comparison feature includes three-dimensional position coordinates or parameter information of the chip product to be compared; and obtaining the comparison feature in response to the user's operation includes:

[0011] In response to a position selected by a user in the three-dimensional stereoscopic model of any one of the at least two chip products, acquiring three-dimensional position coordinates corresponding to the selected position;

[0012] or,

[0013] Receive parameter information of chip products that need to be compared input by the user.

[0014] In another specific embodiment, the comparison feature includes the three-dimensional position coordinates of the chip product to be compared, and for each chip product, obtaining a comparison result corresponding to the comparison feature from the three-dimensional model corresponding to the chip product includes:

[0015] For each chip product, according to the three-dimensional position coordinates, locate a two-dimensional plane position in the three-dimensional model corresponding to the chip product;

[0016] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0017] Acquire parameter information at the three-dimensional position coordinates from the stored parameters of the chip product;

[0018] The comparison result includes the slice image and the parameter information.

[0019] In another specific embodiment, the comparison feature includes parameter information of the chip product to be compared, and obtaining, for each chip product, a comparison result corresponding to the comparison feature from a three-dimensional model corresponding to the chip product includes:

[0020] For each chip product, locating three-dimensional position coordinates corresponding to the parameter information in a three-dimensional model corresponding to the chip product;

[0021] Locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates;

[0022] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0023] The comparison result includes the three-dimensional position coordinates and the slice image.

[0024] In another specific embodiment, the comparison result further includes the parameter information and / or the two-dimensional plane position.

[0025] In another specific embodiment, locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates includes:

[0026] A neural network stereo vision positioning method is adopted to locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product.

[0027] In another specific embodiment, displaying the comparison result corresponding to each chip product includes:

[0028] Displaying the corresponding comparison results in the three-dimensional model corresponding to each chip product in the graphical user interface;

[0029] or,

[0030] The comparison result display area preset in the graphical user interface displays the comparison result corresponding to each chip product.

[0031] In a second aspect, an embodiment of the present application provides a modeling method for a chip product, including:

[0032] Acquire a slice sequence of a chip product to be modeled, wherein the slice sequence includes a plurality of consecutive slice images obtained by slicing the chip product;

[0033] According to the slice sequence, a neural network is used to reconstruct and construct a three-dimensional stereo model corresponding to the chip product; wherein, the three-dimensional stereo model is used to compare parameter information and slice images when comparing chip products.

[0034] In a specific embodiment, the neural network includes a residual network, a U-net network model, and a multi-granularity network model; the three-dimensional model corresponding to the chip product obtained by reconstructing the slice sequence using the neural network includes:

[0035] Inputting each slice image in the slice sequence into a residual network for feature extraction to obtain a first image feature of each slice image;

[0036] The slice sequence is segmented using a U-net network model, and each slice image in the segmented slice sequence is input into a residual network for feature extraction to obtain a second image feature of each slice image;

[0037] Merging the first image feature and the second image feature of each slice image to obtain the image feature of the slice image;

[0038] According to the slice sequence and the image features of each slice image, a multi-granularity network model is used to construct a three-dimensional stereo model corresponding to the chip product.

[0039] In another specific embodiment, the three-dimensional model corresponding to the chip product is constructed using a multi-granularity network model based on the slice sequence and the image features of each slice image, including:

[0040] Based on the plurality of slice images in the slice sequence, a multi-granularity network model is used, with the center of the chip product as the center of the three-dimensional coordinate system, to generate a grid-like three-dimensional geometric shape;

[0041] The image features of each slice image are embedded into the three-dimensional geometric shape and trained to obtain the three-dimensional stereo model.

[0042] In another specific embodiment, the image features of each slice image in the slice sequence include: structural features in the slice image and association information between the slice image and corresponding preceding and following images in the slice sequence.

[0043] In a third aspect, an embodiment of the present application provides a chip product comparison device, comprising:

[0044] an acquisition module, configured to acquire a comparison feature in response to a user operation when acquiring a task of comparing at least two chip products;

[0045] a processing module configured to obtain, for each chip product, a comparison result corresponding to the comparison feature from a three-dimensional model corresponding to the chip product; wherein the three-dimensional model corresponding to each chip product is a three-dimensional model reconstructed using a neural network based on a slice sequence of the chip product;

[0046] The display module is used to display the comparison results corresponding to each chip product.

[0047] Optionally, the comparison feature includes three-dimensional position coordinates or parameter information of the chip product to be compared; the acquisition module is specifically configured to:

[0048] In response to a position selected by a user in the three-dimensional stereoscopic model of any one of the at least two chip products, acquiring three-dimensional position coordinates corresponding to the selected position;

[0049] or,

[0050] Receive parameter information of chip products that need to be compared input by the user.

[0051] Optionally, the comparison feature includes the three-dimensional position coordinates of the chip product to be compared, and the processing module is specifically configured to:

[0052] For each chip product, according to the three-dimensional position coordinates, locate a two-dimensional plane position in the three-dimensional model corresponding to the chip product;

[0053] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0054] Acquire parameter information at the three-dimensional position coordinates from the stored parameters of the chip product;

[0055] The comparison result includes the slice image and the parameter information.

[0056] Optionally, the comparison feature includes parameter information of chip products to be compared, and the processing module is specifically configured to:

[0057] For each chip product, locating three-dimensional position coordinates corresponding to the parameter information in a three-dimensional model corresponding to the chip product;

[0058] Locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates;

[0059] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0060] The comparison result includes the three-dimensional position coordinates and the slice image.

[0061] Optionally, the comparison result also includes the parameter information and / or the two-dimensional plane position.

[0062] Optionally, the processing module is specifically configured to:

[0063] A neural network stereo vision positioning method is adopted to locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product.

[0064] Optionally, the display module is specifically used to:

[0065] Displaying the corresponding comparison results in the three-dimensional model corresponding to each chip product in the graphical user interface;

[0066] or,

[0067] The comparison result display area preset in the graphical user interface displays the comparison result corresponding to each chip product.

[0068] In a fourth aspect, an embodiment of the present application provides a modeling device for a chip product, comprising:

[0069] An acquisition module, configured to acquire a slice sequence of a chip product to be modeled, wherein the slice sequence includes a plurality of consecutive slice images acquired by slicing the chip product;

[0070] A processing module is used to reconstruct a three-dimensional stereo model corresponding to the chip product using a neural network based on the slice sequence; wherein the three-dimensional stereo model is used to compare parameter information and slice images when comparing chip products.

[0071] Optionally, the neural network includes a residual network, a U-net network model, and a multi-granularity network model; the processing module is specifically used to:

[0072] Inputting each slice image in the slice sequence into a residual network for feature extraction to obtain a first image feature of each slice image;

[0073] The slice sequence is segmented using a U-net network model, and each slice image in the segmented slice sequence is input into a residual network for feature extraction to obtain a second image feature of each slice image;

[0074] Merging the first image feature and the second image feature of each slice image to obtain the image feature of the slice image;

[0075] According to the slice sequence and the image features of each slice image, a multi-granularity network model is used to construct a three-dimensional stereo model corresponding to the chip product.

[0076] Optionally, the processing module is specifically configured to include:

[0077] Based on the plurality of slice images in the slice sequence, a multi-granularity network model is used, with the center of the chip product as the center of the three-dimensional coordinate system, to generate a grid-like three-dimensional geometric shape;

[0078] The image features of each slice image are embedded into the three-dimensional geometric shape and trained to obtain the three-dimensional stereo model.

[0079] Optionally, the image features of each slice image in the slice sequence include: structural features in the slice image and association information between the slice image and corresponding preceding and following images in the slice sequence.

[0080] In a fifth aspect, an embodiment of the present application provides an electronic device, including:

[0081] processor and memory;

[0082] The memory is used to store a computer program, and the processor calls the computer program stored in the memory to execute the method described in any one of the first aspect or any one of the second aspect.

[0083] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the processor is enabled to execute the method described in any one of the first aspect or any one of the second aspect.

[0084] The embodiments of the present application provide a chip product comparison method, modeling method, device, and storage medium. By pre-modeling the chip product using a neural network based on its slice sequence to obtain a three-dimensional model, the chip product is compared. When comparing chip products, comparison features are obtained in response to user operations. For each chip product, the comparison results corresponding to the comparison features are obtained from its corresponding three-dimensional model, and then the comparison results corresponding to each chip product are displayed. This eliminates the need for manual image comparisons through multiple electronic files and the need for manual processing of large amounts of information, thereby improving the efficiency and accuracy of chip product comparisons. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0086] Figure 1 Flowchart of Example 1 of the modeling method for a chip product provided in an embodiment of the present application;

[0087] Figure 2 A schematic diagram of constructing a three-dimensional model of a chip product provided in an embodiment of the present application;

[0088] Figure 3 A schematic diagram of constructing a three-dimensional model of a chip product using a neural network model provided in an embodiment of the present application;

[0089] Figure 4 A flowchart of an embodiment of a method for comparing chip products provided in an embodiment of the present application;

[0090] Figure 5 A schematic diagram of the modeling and comparison process of the chip products provided in the embodiments of the present application;

[0091] Figure 6 A comparative schematic diagram of a chip product provided in an embodiment of the present application;

[0092] Figure 7 A schematic structural diagram of a comparative device embodiment of the chip product provided in an embodiment of the present application;

[0093] Figure 8 A schematic diagram of the structure of an embodiment of a modeling device for a chip product provided in an embodiment of the present application;

[0094] Figure 9 A schematic diagram of the structure of an electronic device embodiment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0095] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0096] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the application described herein can, for example, be implemented in an order other than that illustrated or described herein.

[0097] In addition, the terms "comprises" and "having" and any variations thereof are intended to cover a non-exclusive inclusion. For example, a process, method, system, product or apparatus that includes a series of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, product or apparatus.

[0098] The chip (i.e., integrated circuit) industry is a strategic, basic, and leading industry for the national economy and social development, and plays a key role in several major fields such as computers, consumer electronics, network communications, and automotive electronics. During the production and application of chip products, it is necessary to compare products to clarify the characteristics and differences between different chip products. At present, the comparison of chip products is mainly carried out manually. The chip products are analyzed to obtain a two-dimensional image of the cross-section of the chip products. The two-dimensional images are manually marked with relevant parameters, and then organized into electronic documents. Different documents are used to store the parameters and images of different chip products. When chip products need to be compared, multiple documents need to be opened manually, and each parameter needs to be searched for manually. The comparison efficiency is low, and the accuracy of the manual correspondence and marking scheme is poor.

[0099] To address the above issues, this application provides a modeling method and a comparison method for chip products. The overall idea of ​​this solution is that, in the process of solving the above problems, the inventors found that for the same chip products, their internal structures are the same, and the parameter information of different locations or chips is also consistent. If the chip products can be digitally modeled and the model can be automatically compared by the system, the above problems can be solved.

[0100] Based on this, cross-sectional images of chip products can be continuously obtained, the images can be analyzed, and a three-dimensional model can be established based on the analysis results. By modeling the three-dimensional structure of the chip product, the internal structure of the chip product and related parameter information can be intuitively viewed to solve the problem of not being able to locate the specific position when comparing chip products. At the same time, when comparing chip products, the system can compare the three-dimensional models of different chip products, automatically detect the same parameter information or different parameter information between chip products, and directly view the structure of different positions and the corresponding interface pictures, effectively solving the problem of efficiency and accuracy of chip product comparison.

[0101] The solution can be applied to computer devices with computing capabilities, such as computers, personal computers, smart phones and other electronic devices.

[0102] The following describes in detail the chip product modeling method and the chip product comparison method provided in this application through several specific embodiments.

[0103] Figure 1 This is a flowchart of Example 1 of the modeling method for a chip product provided in an embodiment of the present application, as shown in FIG. Figure 1 As shown, the modeling method of the chip product specifically includes the following steps:

[0104] S101: Acquire a slice sequence of a chip product to be modeled, where the slice sequence includes a plurality of consecutive slice images obtained by slicing the chip product.

[0105] In this step, to model the chip product, you first need to obtain a slice sequence of the chip product. This slice sequence is actually a series of continuous images, which are slice images of the chip product. In practice, the chip product can be sliced ​​continuously according to a certain structure or slice interval, and each slice is captured or scanned to obtain an image of each section. This results in multiple continuous slice images, thus obtaining a slice sequence.

[0106] S102: Based on the slice sequence, a neural network is used to reconstruct and obtain a three-dimensional model corresponding to the chip product.

[0107] Among them, the three-dimensional model is used to compare parameter information and slice images when comparing chip products.

[0108] In this step, after obtaining the slice sequence, the slice sequence can be reconstructed through a neural network to obtain a three-dimensional model of the chip product. The specific modeling process can be implemented as follows: the neural networks involved in this scheme include residual networks, U-net network models and multi-granularity network models.

[0109] After obtaining the slice sequence, each slice image in the slice sequence is input into a residual network for feature extraction to obtain a first image feature of each slice image.

[0110] In another branch, the slice sequence is segmented using a U-net network model, and each slice image in the segmented slice sequence is input into a residual network for feature extraction, thereby obtaining a second image feature for each slice image. In other words, for the same slice image, the first image feature is the feature directly extracted using the residual network, while the second image feature is the feature extracted by first segmenting the slice image into small blocks and then inputting it into the residual network.

[0111] The first image feature and the second image feature of each slice image are merged to obtain the image feature of the slice image. The image feature of the slice image indicates the feature of the image itself. The merging of the first image feature and the second image feature can effectively improve the accuracy of the acquired slice image. Image features include but are not limited to color features, texture features, shape features, depth features, etc. In a specific implementation of this solution, the image features of each slice image in the slice sequence include: structural features in the slice image and association information between the slice image and the corresponding previous and next images in the slice sequence.

[0112] After obtaining the features of each slice image, a three-dimensional stereo model corresponding to the chip product is constructed using a multi-granularity network model according to the slice sequence and the image features of each slice image.

[0113] In a specific embodiment, a multi-granularity network model can be used based on multiple slice images in the slice sequence, with the center of the chip product as the center of the three-dimensional coordinate system, to generate a grid-like three-dimensional geometric shape, and then the image features of each slice image are embedded in the three-dimensional geometric shape and trained to obtain a three-dimensional stereo model.

[0114] The chip product modeling method provided in this embodiment obtains a three-dimensional model by pre-modeling the chip product using a neural network based on the chip product's slicing sequence. When comparing chip products, there is no need to manually compare images through multiple electronic files, nor is there any need to manually process large amounts of information, thereby improving the efficiency and accuracy of chip product comparison.

[0115] Figure 2 A schematic diagram of constructing a three-dimensional model of a chip product provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, based on the above embodiment, to more accurately reconstruct the 3D model of the chip product, multiple sub-networks are established to extract the correlation information between the upper and lower images, and the multiple sub-networks form an overall neural network model. After the chip product (i.e., the chip in the figure) is cut to obtain a slice sequence of the chip product, the slice sequence of the chip product is used as input. The neural network model extracts the image features of the slice images (including the characteristics of the image itself and the correlation information between the previous and next images) to reconstruct the 3D structure of the chip.

[0116] Based on any of the above embodiments, a chip product modeling method is described below through a specific implementation process.

[0117] Figure 3 The schematic diagram of using a neural network model to construct a three-dimensional model of a chip product according to an embodiment of the present application is as follows: Figure 3 As shown, the slice sequence is segmented by the U-Net network model, and the different segmented areas are subjected to feature extraction using various residual networks (Residual Network, ResNet), and the extracted features are merged to obtain the second image feature in the aforementioned scheme. At the same time, for each slice image in the slice sequence, its features can also be extracted without cutting. In the specific implementation, the residual network can also be used for feature extraction to obtain the aforementioned first image feature. In order to improve the accuracy of the feature value, the original image (or the extracted features of the original image) is merged with the segmented image features to obtain the final image feature. The image feature can also be represented as some feature data added to the image, that is, it can also be called a feature image.

[0118] After the above processing is performed, according to the feature map, a multi-granularity network (MGN) model is used, with the center of the chip product as the canonical coordinate system, and a three-dimensional geometric shape is generated by the grid. The image features of each slice image output by each sub-network are embedded in the three-dimensional geometric shape, and joint training and reasoning are performed to reconstruct (that is, the reorganization shown in the figure) a three-dimensional structure containing more features, and obtain a three-dimensional model of the chip product. For example, a 10×10×4 slice sequence can be input into a neural network, and after feature extraction and reorganization, it becomes a 10×10×16 three-dimensional structure. According to the 4-dimensional depth slice sequence, a 16-dimensional depth can be obtained after feature learning and reorganization, which can more accurately restore the accuracy of the three-dimensional model of the chip product entity. It should be understood that the dimensions of the slice sequence and the dimensions of the reorganized three-dimensional structure are only described as examples and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0119] For each type of chip product, a corresponding three-dimensional model can be obtained by modeling in the above manner. At the same time, the parameter information of the chip product can be associated with the three-dimensional model so that the required parameter information can be retrieved when comparing or querying.

[0120] The chip product modeling method provided in the above embodiment obtains the slice sequence of the chip product as the input of the neural network model, and can reconstruct a model of the chip product with a three-dimensional structure. During the application of this model, the image and parameter information of each position can be intuitively displayed according to the user's operation. Whether in the application process of chip product comparison or understanding the internal structure of the chip, it can effectively improve efficiency, avoid the possibility of errors in manual comparison, and improve accuracy.

[0121] Figure 4 This is a flowchart of a comparison method embodiment of a chip product provided in an embodiment of the present application, such as Figure 4 As shown, after obtaining the three-dimensional models corresponding to different chip products through the solutions provided in the above embodiments, the following steps are specifically included when comparing the chip products:

[0122] S201: When a task of comparing at least two chip products is acquired, a comparison feature is acquired in response to a user operation.

[0123] In its implementation, this solution will first integrate the chip product comparison solution into a software system, which includes a graphical user interface for user interaction and a database for backend data storage. During chip product application and production, when multiple chip products need to be compared, users can import 3D models corresponding to different chips through the software system, or directly select the chip product's logo to obtain the 3D model.

[0124] In this step, when chip products need to be compared, the user can input a comparison feature through operation. This comparison feature can be a certain location in the chip product. That is, the user can directly select a location in the three-dimensional model of a chip product. The system can obtain the three-dimensional position coordinates of the selected location and use the three-dimensional position coordinates as the comparison feature. Optionally, the comparison feature can also be any parameter of the chip product. The user can enter the parameter to be compared through the input box provided by the graphical user interface provided by the system, or the graphical user interface can directly display different parameters, and the user can click to select the parameter to be compared.

[0125] To sum up, obtaining comparison features includes at least the following two methods: Method 1, in response to a position selected by the user in the three-dimensional stereo model of any chip product among the at least two chip products, obtaining the three-dimensional position coordinates corresponding to the selected position; Method 2, receiving parameter information of the chip product to be compared input by the user.

[0126] S202: For each chip product, obtain a comparison result corresponding to the comparison feature from the three-dimensional model corresponding to the chip product.

[0127] The three-dimensional stereoscopic model corresponding to each chip product is a three-dimensional model reconstructed by using a neural network based on the slice sequence of the chip product. The specific modeling process can be referred to the above embodiment.

[0128] In this step, after obtaining the comparison features based on the user's operation, the system for comparing chip products obtains other parameter features corresponding to the comparison features from the three-dimensional model corresponding to each chip product, or position coordinates, or images of the corresponding positions as comparison results. The specific content included in the comparison results varies depending on the comparison features.

[0129] In one specific implementation, when the comparison features include the three-dimensional position coordinates of the chip product to be compared, the process of obtaining the comparison result based on the three-dimensional model is as follows: for each chip product, based on the three-dimensional position coordinates, locate the two-dimensional plane position in the three-dimensional model corresponding to the chip product. Then, from the slice sequence corresponding to the chip product, obtain the slice image corresponding to the two-dimensional plane position, and then obtain the parameter information at the three-dimensional position coordinates from the stored parameters of the chip product; wherein, the comparison result includes the slice image and parameter information. The parameter information here refers to the parameters of the chip product itself, such as some device, performance and other parameters.

[0130] Optionally, during the chip product comparison process, the two-dimensional plane position or three-dimensional position coordinates may also be displayed in a graphical user interface, and this solution does not impose any restrictions.

[0131] In another specific implementation, when the comparison feature includes parameter information of the chip product to be compared, the process of obtaining the comparison result based on the three-dimensional model corresponding to the chip product is specifically as follows: for each chip product, in the three-dimensional model corresponding to the chip product, the three-dimensional position coordinates corresponding to the parameter information are located; based on the three-dimensional position coordinates, the two-dimensional plane position is located in the three-dimensional model corresponding to the chip product; from the slice sequence corresponding to the chip product, a slice image corresponding to the two-dimensional plane position is obtained; wherein the comparison result includes the three-dimensional position coordinates and the slice image.

[0132] In this solution, the position of the chip product can be located according to a certain parameter of the chip product, and the specific image can be used as the comparison result. In actual applications, the parameter information of the chip product may also include other parameter information, which can also be displayed as the comparison result.

[0133] Optionally, during the chip product comparison process, the two-dimensional plane position or three-dimensional position coordinates may also be displayed in a graphical user interface, and this solution does not impose any restrictions.

[0134] That is to say, in the implementation of this solution, the comparison result may further include parameter information selected as a comparison feature, other parameter information, and / or the two-dimensional plane position, etc.

[0135] In any of the above implementation methods, the specific implementation method of locating the two-dimensional position plane in the three-dimensional model according to the three-dimensional position coordinates can be: using the neural network stereo vision positioning method to locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product.

[0136] The neural network stereo vision positioning method is mainly based on the principle of stereo vision. It uses neural networks to establish a nonlinear mapping relationship between the three-dimensional coordinates of spatial points and image coordinates. This allows the system to match two-dimensional coordinates with three-dimensional coordinates one-to-one without complex calibration of internal and external camera parameters. This simplifies the calibration and positioning calculations of the visual system, is more scientific than traditional methods, and achieves good results in positioning accuracy.

[0137] S203: Display the comparison result corresponding to each chip product.

[0138] In this step, after obtaining the chip product comparison results, the comparison results can be displayed in the comparison system's graphical user interface. Specifically, the graphical user interface can display three-dimensional models of different chip products. In this case, the comparison results can be displayed directly within the three-dimensional models of the chip products, or separately within a pre-set comparison result display area. Specifically, the corresponding comparison results can be displayed within the corresponding three-dimensional model of each chip product in the graphical user interface; alternatively, the comparison results for each chip product can be displayed within a pre-set comparison result display area within the graphical user interface.

[0139] The chip product comparison method provided in this embodiment pre-models the chip products using a neural network based on their slice sequences to generate three-dimensional models. When comparing chip products, the method obtains comparison features in response to user input. For each chip product, the method extracts the comparison results corresponding to the comparison features from its corresponding three-dimensional model, and then displays the comparison results for each chip product. This method eliminates the need for manual image comparisons from multiple electronic files or the processing of large amounts of information, improving the efficiency and accuracy of chip product comparisons.

[0140] Based on any of the above embodiments, the present solution is described below through a comparison process of a specific chip product.

[0141] The core idea of ​​the technical solution provided by this application is to use the chip product segmentation sequence as input and use the neural network method to reconstruct the chip three-dimensional structure, that is, the three-dimensional model, so that the internal overall structure of the chip and its related information can be intuitively viewed, solving the problem of being unable to locate the specific position of the physical chip. When comparing multiple chip products, any product is selected as the target group and the remaining products are used as the control group. When the three-dimensional model of the target group chip product is selected, the same target area is displayed in the control group through automatic target detection and positioning, and the corresponding two-dimensional sequence image and related parameter information are displayed, effectively solving the problem of parameter comparison efficiency and accuracy.

[0142] Figure 5 A schematic diagram of the modeling and comparison process of the chip product provided in the embodiment of the present application, such as Figure 5As shown, before comparing chip products, the chip products are cut in some way (for example, directly cutting at a certain interval to obtain slice images, or polishing the chip products and obtaining continuous images during the polishing process) to obtain a series of slice sequences. The slice sequence of the chip products is then reorganized using a neural network to construct a three-dimensional structure to obtain a three-dimensional model, which can be imported into the developed comparison system. Based on the aforementioned preparation process, when comparing various chip products, any position of the three-dimensional model of any chip product can be selected, and the remaining products can automatically detect and locate the same selected target area and display the corresponding two-dimensional image (i.e., slice image) and specific related parameters.

[0143] Figure 6 A comparative schematic diagram of a chip product provided in an embodiment of the present application is shown in FIG. Figure 6 The figure shows a specific method for comparing multiple chip products. Product 1 is the target group. The user manually selects the location to be compared within Product 1, represented by three-dimensional coordinates (x, y, z). The comparison system extracts features from this location and uses a neural network stereo vision method for target detection. It locates the corresponding two-dimensional plane position (x1, y1) for the selected location and then automatically locates the same target location in other products 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 111, 112, 1

[0144] Furthermore, the slice images and parameter information of different chip products at the position can be obtained through the two-dimensional plane position as the comparison result of this chip product comparison.

[0145] Alternatively, in another implementation, based on the target area selected by Product 1, other products automatically detect and locate the same target area and display the corresponding 2D images and parameter information for each product, enabling quick comparison. For example, you can select the 2D plane position (x1, y1) in Product 1. If the parameter information corresponding to the 2D plane position (x1, y1) is detected to contain a structure name, the 2D images of each product containing the structure name can be displayed.

[0146] The chip product modeling method and comparison method provided in the embodiments of the present application can intuitively view the internal parameters and structure of each chip product by reconstructing the three-dimensional structure of the chip product. At the same time, when comparing chip products, the target automatic detection and positioning method is used to make each chip product display the same target area at the same time, effectively improving the comparison efficiency and accuracy. Furthermore, during the comparison process, when any three-dimensional model area of ​​a chip product is selected, both the target group and the control group can display the corresponding two-dimensional image and related specific parameter information, which is convenient for intuitive comparison to view the differences in the same parameters between the products.

[0147] Figure 7A schematic diagram of a comparative device embodiment of the chip product provided in the embodiment of the present application is shown in FIG. Figure 7 As shown, the comparison device 10 of the chip product includes:

[0148] An acquisition module 11 is configured to acquire a comparison feature in response to a user operation when a task of comparing at least two chip products is acquired;

[0149] a processing module 12 configured to obtain, for each chip product, a comparison result corresponding to the comparison feature from a three-dimensional model corresponding to the chip product; wherein the three-dimensional model corresponding to each chip product is a three-dimensional model reconstructed using a neural network based on a slice sequence of the chip product;

[0150] The display module 13 is used to display the comparison result corresponding to each chip product.

[0151] Optionally, the comparison feature includes three-dimensional position coordinates or parameter information of the chip product to be compared; the acquisition module 11 is specifically configured to:

[0152] In response to a position selected by a user in the three-dimensional stereoscopic model of any one of the at least two chip products, acquiring three-dimensional position coordinates corresponding to the selected position;

[0153] or,

[0154] Receive parameter information of chip products that need to be compared input by the user.

[0155] Optionally, the comparison feature includes the three-dimensional position coordinates of the chip product to be compared, and the processing module 12 is specifically configured to:

[0156] For each chip product, according to the three-dimensional position coordinates, locate a two-dimensional plane position in the three-dimensional model corresponding to the chip product;

[0157] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0158] Acquire parameter information at the three-dimensional position coordinates from the stored parameters of the chip product;

[0159] The comparison result includes the slice image and the parameter information.

[0160] Optionally, the comparison feature includes parameter information of chip products to be compared, and the processing module 12 is specifically configured to:

[0161] For each chip product, locating three-dimensional position coordinates corresponding to the parameter information in a three-dimensional model corresponding to the chip product;

[0162] Locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates;

[0163] Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product;

[0164] The comparison result includes the three-dimensional position coordinates and the slice image.

[0165] Optionally, the comparison result also includes the parameter information and / or the two-dimensional plane position.

[0166] Optionally, the processing module 12 is specifically configured to:

[0167] A neural network stereo vision positioning method is adopted to locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product.

[0168] Optionally, the display module 13 is specifically used to:

[0169] Displaying the corresponding comparison results in the three-dimensional model corresponding to each chip product in the graphical user interface;

[0170] or,

[0171] The comparison result display area preset in the graphical user interface displays the comparison result corresponding to each chip product.

[0172] The chip product comparison device provided in any of the aforementioned embodiments is used to implement the technical solution of the chip product comparison method in the aforementioned embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0173] Figure 8 This is a schematic diagram of the structure of the chip product modeling device embodiment provided in the embodiment of the present application, as shown in FIG. Figure 8 As shown, the chip product modeling device 20 includes:

[0174] An acquisition module 21 is configured to acquire a slice sequence of a chip product to be modeled, wherein the slice sequence includes a plurality of consecutive slice images obtained by slicing the chip product;

[0175] The processing module 22 is used to reconstruct a three-dimensional stereo model corresponding to the chip product using a neural network according to the slice sequence; wherein the three-dimensional stereo model is used to compare parameter information and slice images when comparing chip products.

[0176] Optionally, the neural network includes a residual network, a U-net network model, and a multi-granularity network model; the processing module 22 is specifically used to:

[0177] Inputting each slice image in the slice sequence into a residual network for feature extraction to obtain a first image feature of each slice image;

[0178] The slice sequence is segmented using a U-net network model, and each slice image in the segmented slice sequence is input into a residual network for feature extraction to obtain a second image feature of each slice image;

[0179] Merging the first image feature and the second image feature of each slice image to obtain the image feature of the slice image;

[0180] According to the slice sequence and the image features of each slice image, a multi-granularity network model is used to construct a three-dimensional stereo model corresponding to the chip product.

[0181] Optionally, the processing module 22 is specifically configured to:

[0182] Based on the plurality of slice images in the slice sequence, a multi-granularity network model is used, with the center of the chip product as the center of the three-dimensional coordinate system, to generate a grid-like three-dimensional geometric shape;

[0183] The image features of each slice image are embedded into the three-dimensional geometric shape and trained to obtain the three-dimensional stereo model.

[0184] Optionally, the image features of each slice image in the slice sequence include: structural features in the slice image and association information between the slice image and corresponding preceding and following images in the slice sequence.

[0185] The chip product modeling device provided in any of the aforementioned embodiments is used to implement the technical solution of the chip product modeling method in the aforementioned embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0186] Figure 9 A schematic diagram of the structure of an electronic device embodiment provided in an embodiment of the present application is shown in FIG. Figure 9 As shown, the electronic device 30 includes:

[0187] Processor 31 and memory 32; optionally, it may also include a display 33 and a bus 34 for connecting various devices.

[0188] The memory 32 is used to store computer programs, and the processor 31 calls the computer program stored in the memory 32 to execute the technical solution in any of the above method embodiments.

[0189] Optionally, the memory 32 may be independent or integrated with the processor 31 .

[0190] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the processor is enabled to execute the technical solution in any of the aforementioned method embodiments.

[0191] It should be understood that the processor mentioned in the embodiments of the present application may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0192] It should also be understood that the memory mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0193] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, the memory (storage module) is integrated into the processor.

[0194] It should be noted that the memory described herein is intended to include, but not be limited to, these and any other suitable types of memory.

[0195] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0196] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A chip product comparison method, characterized in that: include: When a task of comparing at least two chip products is acquired, obtaining a comparison feature in response to a user operation; For each chip product, a comparison result corresponding to the comparison feature is obtained from a three-dimensional model corresponding to the chip product; wherein the three-dimensional model corresponding to each chip product is a three-dimensional model reconstructed using a neural network based on a slice sequence of the chip product; Display the comparison results corresponding to each chip product; The comparison feature includes the three-dimensional position coordinates of the chip product to be compared, and for each chip product, obtaining a comparison result corresponding to the comparison feature from the three-dimensional model corresponding to the chip product includes: For each chip product, according to the three-dimensional position coordinates, locate a two-dimensional plane position in the three-dimensional model corresponding to the chip product; Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product; Acquire parameter information at the three-dimensional position coordinates from the stored parameters of the chip product; Wherein, the comparison result includes the slice image and the parameter information; The step of locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates includes: A neural network stereo vision positioning method is adopted to locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product.

2. The method according to claim 1, characterized in that The acquiring of the comparison feature in response to the user's operation includes: In response to a position selected by a user in the three-dimensional stereoscopic model of any one of the at least two chip products, acquiring three-dimensional position coordinates corresponding to the selected position; or, Receive parameter information of chip products that need to be compared input by the user.

3. The method according to claim 2, characterized in that The comparison feature includes parameter information of the chip product to be compared. For each chip product, obtaining a comparison result corresponding to the comparison feature from a three-dimensional model corresponding to the chip product includes: For each chip product, locating three-dimensional position coordinates corresponding to the parameter information in a three-dimensional model corresponding to the chip product; Locating a two-dimensional plane position in a three-dimensional model corresponding to the chip product according to the three-dimensional position coordinates; Acquire a slice image corresponding to the two-dimensional plane position from a slice sequence corresponding to the chip product; The comparison result includes the three-dimensional position coordinates and the slice image.

4. The method according to claim 3, characterized in that The comparison result also includes the two-dimensional plane position.

5. The method according to any one of claims 1 to 4, characterized in that The display of the comparison results corresponding to each chip product includes: Displaying the corresponding comparison results in the three-dimensional model corresponding to each chip product in the graphical user interface; or, The comparison result display area preset in the graphical user interface displays the comparison result corresponding to each chip product.

6. A modeling method for a chip product, characterized in that: include: Acquire a slice sequence of a chip product to be modeled, wherein the slice sequence includes a plurality of consecutive slice images obtained by slicing the chip product; Based on the slice sequence, a neural network is used to reconstruct and obtain a three-dimensional model corresponding to the chip product; wherein the three-dimensional model is used to compare parameter information and slice images when comparing chip products; The neural network includes a residual network, a U-net network model, and a multi-granularity network model; the three-dimensional model corresponding to the chip product obtained by reconstructing the slice sequence using the neural network includes: Inputting each slice image in the slice sequence into a residual network for feature extraction to obtain a first image feature of each slice image; The slice sequence is segmented using a U-net network model, and each slice image in the segmented slice sequence is input into a residual network for feature extraction to obtain a second image feature of each slice image; Merging the first image feature and the second image feature of each slice image to obtain the image feature of the slice image; According to the slice sequence and the image features of each slice image, a multi-granularity network model is used to construct a three-dimensional stereo model corresponding to the chip product.

7. The method according to claim 6, characterized in that The method of constructing a three-dimensional model corresponding to the chip product using a multi-granularity network model based on the slice sequence and the image features of each slice image includes: Based on the plurality of slice images in the slice sequence, a multi-granularity network model is used, with the center of the chip product as the center of the three-dimensional coordinate system, to generate a grid-like three-dimensional geometric shape; The image features of each slice image are embedded into the three-dimensional geometric shape and trained to obtain the three-dimensional stereo model.

8. The method according to claim 6 or 7, characterized in that The image features of each slice image in the slice sequence include: structural features in the slice image and association information between the slice image and corresponding preceding and following images in the slice sequence.

9. A comparison device for chip products, characterized in that: include: an acquisition module, configured to acquire a comparison feature in response to a user operation when acquiring a task of comparing at least two chip products; A processing module is configured to obtain, for each chip product, a comparison result corresponding to the comparison feature from a three-dimensional model corresponding to the chip product; wherein the three-dimensional model corresponding to each chip product is a three-dimensional model reconstructed by using a neural network based on a slice sequence of the chip product; the comparison feature includes the three-dimensional position coordinates of the chip product to be compared, and the processing module is configured to: for each chip product, locate a two-dimensional plane position in the three-dimensional model corresponding to the chip product based on the three-dimensional position coordinates; obtain a slice image corresponding to the two-dimensional plane position from the slice sequence corresponding to the chip product; and obtain parameter information at the three-dimensional position coordinates from stored parameters of the chip product; wherein the comparison result includes the slice image and the parameter information; and the processing module is further configured to: locate the two-dimensional plane position corresponding to the three-dimensional position coordinates in the three-dimensional model corresponding to the chip product using a neural network stereo vision positioning method; The display module is used to display the comparison results corresponding to each chip product.

10. An electronic device, characterized in that: include: processor and memory; The memory is used to store a computer program, and the processor calls the computer program stored in the memory to execute the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor is enabled to perform the method according to any one of claims 1 to 8.

Images