Intelligent analysis method and system for automobile body metal material model

Abstract

The embodiment of the invention discloses an intelligent analysis method and system for the model of a metal material of an automobile body. The method comprises the following steps: acquiring a metal cutting sample corresponding to a target automobile body; performing multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data; based on a preset material sample library, determining a fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library, the fusion similarity is a similarity determined after obtaining a weight corresponding to the multi-dimensional feature data based on a pre-trained weight prediction model and carrying out weighted fusion processing on the multi-dimensional feature data according to the weight; determining a model matching probability corresponding to each candidate material according to the fusion similarity; and obtaining an optimal analysis result corresponding to the target automobile body based on the model matching probability corresponding to each candidate material. According to the embodiment of the invention, the analysis accuracy of the model of the automobile body metal material is improved.

Description

TECHNICAL FIELD

[0001] One or more embodiments of the present application relate to the field of automotive technology, in particular to an intelligent analysis method and system for a metal material type of an automobile body. BACKGROUND

[0002] In the development process of a new automobile model, it is a crucial link to disassemble and analyze the benchmark vehicle body. Disassembling the body-in-white refers to a technical process of systematically decomposing the uncoated body structure, which is mainly used for safety testing and verification, reverse engineering analysis, and manufacturing process optimization. Among them, accurately obtaining the metal material type of the benchmark vehicle body is the basis for cost analysis, lightweight design, performance evaluation, and process selection. Currently, the mainstream method for obtaining the metal material type of the benchmark vehicle body mainly relies on supply chain consultation or technical exchange and experience analogy. This method relies on the analysis experience of technical personnel and has strong subjectivity, making the accuracy of the finally determined metal material type of the benchmark vehicle body relatively low. Therefore, there is an urgent need for an intelligent analysis method that can improve the accuracy of the metal material type of the benchmark vehicle body. SUMMARY

[0003] The embodiments of the present application provide an intelligent analysis method and system for a metal material type of an automobile body, and the technical solutions are as follows: In a first aspect, the embodiments of the present application provide an intelligent analysis method and system for a metal material type of an automobile body, which includes: obtaining a metal cutting sample corresponding to a target automobile body; performing multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, the multi-dimensional feature data including material composition data, crystal phase organization images, tensile performance data, and hardness data; determining a fusion similarity between the multi-dimensional feature data and each candidate material in a preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data are weighted and fused based on a pre-trained weight prediction model to obtain a weight corresponding to the multi-dimensional feature data; determining a type matching probability corresponding to each candidate material according to the fusion similarity; and obtaining an optimal analysis result corresponding to the target automobile body based on the type matching probability corresponding to each candidate material.

[0004] In a second aspect, an embodiment of the present application provides an intelligent analysis system for a metal material type of an automobile body, comprising: a sample acquisition module configured to acquire a metal cutting sample corresponding to a target automobile body; a detection module configured to perform multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, the multi-dimensional feature data including material composition data, a crystal phase organization image, tensile property data, and hardness data; a similarity calculation module configured to determine a fusion similarity between the multi-dimensional feature data and each candidate material in a preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data is weighted and fused based on a pre-trained weight prediction model to obtain a weight corresponding to the multi-dimensional feature data; a probability determination module configured to determine a type matching probability of each candidate material based on the fusion similarity; and a type matching module configured to obtain an optimal analysis result corresponding to the target automobile body based on the type matching probability of each candidate material.

[0005] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor and a memory; The processor is connected to the memory, and the memory is configured to store executable program codes; the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute steps in the intelligent analysis method for a metal material type of an automobile body.

[0006] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement steps in the intelligent analysis method for a metal material type of an automobile body.

[0007] The technical scheme provided by some embodiments of the present application has at least the following beneficial effects: The embodiment of the present application adopts multi-dimensional feature data including material composition data, crystal phase organization image, tensile property data and hardness data as the analysis basis of the model, and determines the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library based on the preset material sample library, the fusion similarity being the similarity determined after the multi-dimensional feature data is weighted and fused based on the weight of the pre-trained weight prediction model, the embodiment of the present application applies the weighted fusion algorithm to the material reverse analysis process, dynamically determines the weight through the pre-trained weight prediction model, thereby accurately measures the contribution degree of different data types in determining different material models, realizes information complementation and fusion, and solves the misjudgment problem caused by a single data source; moreover, the embodiment of the present application can intelligently dynamically adjust the weight of each dimension feature based on the weight prediction model, automatically adapt the material model emphasis of different materials such as high-strength steel, aluminum alloy and cast iron, thereby improving the generalization ability of the system in processing various metal materials; the embodiment of the present application determines the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library to make intelligent comprehensive decision, so that the intelligent analysis result of the automobile body metal material model has scientific data support, and the accuracy and reliability of the analysis result are greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0009] Figure 1 is an application scenario diagram of the intelligent analysis method of the automobile body metal material model provided by the present application.

[0010] Figure 2 is a flowchart of the intelligent analysis method of the automobile body metal material model provided by the present application.

[0011] Figure 3 is a flowchart of determining the fusion similarity provided by the present application.

[0012] Figure 4 is a structure diagram of the intelligent analysis system of the automobile body metal material model provided by the present application.

[0013] Figure 5 is a structure diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application.

[0015] The terms "first", "second", etc. in the present application and the claims and the above drawings are used to distinguish different objects, rather than to describe a specific sequence. In addition, the term "comprising" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0016] The intelligent analysis method for the type of metal material of the automobile body provided in the embodiments of the present application can be an intelligent analysis system for the type of metal material of the automobile body provided in the embodiments of the present application.

[0017] Before the intelligent analysis method for the type of metal material of the automobile body is described in detail in combination with one or more embodiments, the scene in which the intelligent analysis method for the type of metal material of the automobile body is applied is introduced.

[0018] Please refer to Figure 1 , Figure 1 The application scene of the intelligent analysis method for the type of metal material of the automobile body provided in the embodiments of the present application is shown in the figure. In this embodiment, the intelligent analysis system 100 for the type of metal material of the automobile body can include an X-ray fluorescence spectrometer, a scanning electron microscope, a universal testing machine, a hardness tester, an electronic device, etc. The X-ray fluorescence spectrometer is used to obtain material composition data corresponding to the metal cutting sample; the scanning electron microscope is used to obtain a crystal phase organization image corresponding to the metal cutting sample; the universal testing machine is used to obtain tensile property data corresponding to the metal cutting sample; and the hardness tester is used to obtain hardness data corresponding to the hardness tester.

[0019] In some embodiments, the electronic device of the intelligent analysis system 100 can be a terminal, a server, etc. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a notebook computer, or a personal computer (PC), etc. The server can be a single server, or a server cluster composed of multiple servers.

[0020] In some embodiments, the intelligent analysis system 100 can be integrated in multiple servers, and the multiple servers can be used to implement the intelligent analysis method for the type of metal material of the automobile body of the present application.

[0021] In some embodiments, the server can also be implemented in the form of a terminal. The terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a notebook computer, or a personal computer (PC), and the like. The terminal includes a central processing unit (CPU), a graphics processing unit (GPU), a memory, a storage device, a network communication module, a sensor, a display screen, a battery and a power management module, and the like.

[0022] For example, referring to Figure 1 The electronic device can include a server 110 and a storage terminal 120, and the storage terminal 120 stores material composition data, crystal phase image data, tensile performance data, hardness data, and the like, and the server 110 and the storage terminal 120 are in communication connection with each other, which will not be described here.

[0023] The server 110 can include a processor and a memory, and the like. The intelligent analysis system 100 can obtain a metal cutting sample corresponding to the target automobile body based on the cutting machine; then the metal cutting sample is detected by an X-ray fluorescence spectrometer, a scanning electron microscope, a universal testing machine, and a hardness tester, respectively, to obtain multi-dimensional feature data, including material composition data, crystal phase image data, tensile performance data, and hardness data; then the intelligent analysis system 100 can execute the following through the server 110: obtaining a metal cutting sample corresponding to the target automobile body; performing multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, including material composition data, crystal phase image data, tensile performance data, and hardness data; determining the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data is weighted and fused based on a pre-trained weight prediction model to obtain the weight corresponding to the multi-dimensional feature data; determining the model matching probability corresponding to each candidate material according to the fusion similarity; and obtaining the optimal analysis result corresponding to the target automobile body based on the model matching probability corresponding to each candidate material.

[0024] It should be noted that Figure 1 The scene diagram of the intelligent analysis system of the automobile body metal material model shown is only an example, and the automobile body metal material model intelligent analysis system and the scene described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. It is known to those skilled in the art that, as the automobile body metal material model intelligent analysis system evolves and new scenes appear, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0025] Referring to Figure 2 , Figure 2 is a flowchart of an intelligent analysis method for a vehicle body metal material type provided by an embodiment of the present application. The intelligent analysis method for a vehicle body metal material type can be executed by an intelligent analysis system 100 as shown in the figure. The intelligent analysis method for a vehicle body metal material type can at least include the following steps: Figure 1 200, obtaining a metal cutting sample corresponding to a target vehicle body; 210, performing multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, the multi-dimensional feature data including material composition data, crystal phase organization images, tensile property data, and hardness data; 220, determining a fusion similarity between the multi-dimensional feature data and each candidate material in a preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data is weighted and fused based on a pre-trained weight prediction model to obtain a weight corresponding to the multi-dimensional feature data; 230, determining a type matching probability corresponding to each candidate material according to the fusion similarity; 240, obtaining an optimal analysis result corresponding to the target vehicle body based on the type matching probability corresponding to each candidate material.

[0026] In this embodiment, the multi-dimensional similarity can be a similarity between the multi-dimensional feature data calculated by a material type reverse modeling model and a material sample in the preset material sample library. The similarity can be a cosine similarity, etc. The preset material sample library can be a sample database corresponding to a plurality of candidate materials. The sample data corresponding to the candidate materials can include material composition sample data, crystal phase organization feature vectors, tensile property data, and hardness data, etc. corresponding to the candidate materials. The material composition data can be alloy elements measured by an X-ray fluorescence spectrometer, such as C, Mn, Si, Cr, Ni, Al, Mg, etc. The crystal phase organization feature vector can be a crystal phase organization image measured by a scanning electron microscope and a feature vector extracted by a convolutional neural network model. The tensile property data can be the tensile strength, yield strength, elastic modulus, etc. measured by a universal testing machine. The tensile strength is the maximum stress that the material can withstand before being pulled apart. The yield strength is the stress at which the material begins to exhibit significant plastic deformation. The elastic modulus is the ratio of stress to strain.

[0027] ​The embodiment of the present application adopts multi-dimensional feature data including material composition data, crystal phase organization image, tensile property data and hardness data as the analysis basis of the model, and calculates the multi-dimensional similarity between the multi-dimensional feature data and the material samples in the preset material sample library based on the preset material sample library and the pre-trained material model reverse calculation model algorithm, so as to make intelligent comprehensive decision, so that the intelligent analysis result of the automobile body metal material model has scientific data support, and the accuracy and reliability of the analysis result are greatly improved.

[0028] In some embodiments, before the multi-dimensional feature data is input into the pre-trained material model reverse calculation model, the method further includes: preprocessing the crystal phase organization image to obtain a preprocessed crystal phase organization image; and extracting a crystal phase feature vector corresponding to the preprocessed crystal phase organization image based on the pre-trained convolutional neural network model.

[0029] In some embodiments, please refer to Figure 3 , Figure 3 is a flowchart for determining the fusion similarity provided by the embodiment of the present application. The preset material sample library includes a plurality of candidate materials; the preset material sample library includes a plurality of candidate materials; based on the preset material sample library, the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library is determined, including: 300, determining any one candidate material in the preset material sample library; 310, determining the composition similarity between the material composition data and any one candidate material based on the material composition data; 320, determining the crystal phase similarity between the crystal phase feature vector and any one candidate material based on the crystal phase feature vector corresponding to the preprocessed crystal phase organization image; 330, determining the tensile property similarity between the tensile property data and any one candidate material based on the tensile property data; 340, determining the hardness similarity between the hardness data and any one candidate material based on the hardness data; 350, performing weighted fusion processing on the composition similarity, the crystal phase similarity, the tensile property similarity and the hardness similarity to obtain the fusion similarity between the multi-dimensional feature data and any one candidate material.

[0030] In this embodiment, any one candidate material can include corresponding material composition sample data, crystal phase organization feature vector, tensile property data, hardness data, etc. The composition similarity can be the similarity between the material composition data corresponding to the metal cutting sample and the material composition data corresponding to any one candidate material; the crystal phase similarity can be the similarity between the crystal phase feature vector corresponding to the pre-processed crystal phase organization image and the crystal phase feature vector corresponding to any one candidate material; the tensile property similarity can be the similarity between the tensile property data corresponding to the metal cutting sample and the tensile property data corresponding to any one candidate material; and the hardness similarity can be the similarity between the hardness data corresponding to the metal cutting sample and the hardness data corresponding to any one candidate material.

[0031] In this embodiment, the model matching probability corresponding to the candidate material can be the fusion similarity between the multi-dimensional feature data and any one candidate material.

[0032] In some embodiments, obtaining the weight corresponding to the multi-dimensional feature data based on the pre-trained weight prediction model includes: based on the pre-trained weight prediction model, detecting the weights of the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data to obtain the weights corresponding to the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data, respectively.

[0033] This embodiment applies the weighted fusion algorithm to the material reverse analysis process, dynamically determines the weights through the pre-trained weight prediction model, thereby accurately measures the contribution degrees of different data types in determining different material models, realizes information complementation and fusion, and solves the misjudgment problem that may be caused by a single data source.

[0034] In some embodiments, detecting the weights of the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data based on the pre-trained weight prediction model includes: obtaining a weight training set, the weight training set including a plurality of training samples, each training sample including training data and a preset weight corresponding to the training data, the training data including material composition training data, crystal phase training feature vector, tensile property training data, and hardness training data; and performing supervised training on the weight prediction model through the weight training set to obtain the pre-trained weight prediction model.

[0035] In some embodiments, the weight prediction model is a machine learning model trained through supervised learning, and the machine learning model can include an XGBoost model, etc.

[0036] The embodiment can intelligently dynamically adjust the weight of each dimension feature according to the preliminary features (such as high aluminum content or high hardness) of the metal cutting sample based on the weight prediction model, automatically adapt the discrimination emphasis of different materials such as high-strength steel, aluminum alloy, cast iron, and the like, and thus improve the generalization ability of the system in processing various metal materials.

[0037] In some embodiments, the metal cutting sample is subjected to multi-modal detection to obtain multi-dimensional feature data, including: obtaining material composition data corresponding to the metal cutting sample based on an X-ray fluorescence spectrometer; obtaining a crystal phase organization image corresponding to the metal cutting sample based on a scanning electron microscope; obtaining tensile performance data corresponding to the metal cutting sample based on a universal testing machine; and obtaining hardness data corresponding to the hardness tester based on a hardness tester.

[0038] The embodiment of the present application adopts multi-dimensional feature data including material composition data, crystal phase organization image, tensile performance data and hardness data as the basis for model analysis and resolution, and calculates the multi-dimensional similarity between the multi-dimensional feature data and the material samples in the preset material sample library based on the pre-trained material model reverse calculation model algorithm, so as to make intelligent comprehensive decision, so that the intelligent analysis result of the automobile body metal material model has scientific data support, and the accuracy and reliability of the analysis result are greatly improved.

[0039] Moreover, the embodiment of the present application can apply a weighted fusion algorithm to the material reverse analysis process, dynamically determine the weight through the pre-trained weight prediction model, accurately measure the contribution degree of different data types in determining different material models, realize information complementation and fusion, and solve the misjudgment problem caused by a single data source.

[0040] Moreover, the embodiment of the present application can intelligently dynamically adjust the weight of each dimension feature according to the preliminary features (such as high aluminum content or high hardness) of the metal cutting sample based on the weight prediction model, automatically adapt the discrimination emphasis of different materials such as high-strength steel, aluminum alloy, cast iron, and the like, and thus improve the generalization ability of the system in processing various metal materials.

[0041] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order other than that described in the embodiments and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing can be advantageous or possible.

[0042] Please refer to Figure 4 , Figure 4A structural schematic diagram of an intelligent analysis system for a vehicle body metal material type is provided for an embodiment of the present application.

[0043] As shown in Figure 4 The intelligent analysis system for the vehicle body metal material type can at least include a sample acquisition module 400, a detection module 410, a similarity calculation module 420, a probability determination module 430, and a type matching module 440, etc., wherein: The sample acquisition module 400 is configured to acquire a metal cutting sample corresponding to a target vehicle body; The detection module 410 is configured to perform multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, which includes material composition data, crystal phase organization image, tensile property data, and hardness data; The similarity calculation module 420 is configured to determine a fusion similarity between the multi-dimensional feature data and each candidate material in a preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data is weighted and fused based on a pre-trained weight prediction model to obtain a weight corresponding to the multi-dimensional feature data; The probability determination module 430 is configured to determine a type matching probability corresponding to each candidate material according to the fusion similarity; The type matching module 440 is configured to obtain an optimal analysis result corresponding to the target vehicle body based on the type matching probabilities corresponding to the candidate materials.

[0044] In some embodiments, the intelligent analysis system can include an image processing module configured to pre-process the crystal phase organization image to obtain a pre-processed crystal phase organization image, and extract a crystal phase feature vector corresponding to the pre-processed crystal phase organization image based on a pre-trained convolutional neural network model.

[0045] In some embodiments, the similarity calculation module 420 includes a similarity calculation sub-module configured to: determine any one candidate material in the preset material sample library; determine a composition similarity between the material composition data and the any one candidate material based on the material composition data; determine a crystal phase similarity between the crystal phase feature vector corresponding to the pre-processed crystal phase organization image and the any one candidate material based on the crystal phase feature vector; determine a tensile property similarity between the tensile property data and the any one candidate material based on the tensile property data; determine a hardness similarity between the hardness data and the any one candidate material based on the hardness data; and perform weighted fusion processing on the composition similarity, the crystal phase similarity, the tensile property similarity, and the hardness similarity to obtain the fusion similarity between the multi-dimensional feature data and the any one candidate material.

[0046] In some embodiments, the similarity calculation submodule further comprises a weight acquisition module, which is configured to: based on a pre-trained weight prediction model, perform weight detection on the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data, to obtain weights corresponding to the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data respectively.

[0047] In some embodiments, the weight acquisition module comprises a training module, which is configured to: acquire a weight training set, the weight training set comprising a plurality of training samples, each training sample comprising training data and a preset weight corresponding to the training data, the training data comprising material composition training data, crystal phase training feature vector, tensile property training data, and hardness training data; and perform supervised training on the weight prediction model through the weight training set, to obtain the pre-trained weight prediction model.

[0048] In some embodiments, the weight prediction model is a machine learning model trained through supervised learning, and the machine learning model comprises an XGBoost model.

[0049] In some embodiments, the detection module 410 comprises a detection submodule, which is configured to: based on an X-ray fluorescence spectrometer, acquire material composition data corresponding to the metal cutting sample; based on a scanning electron microscope, acquire a crystal phase microstructure image corresponding to the metal cutting sample; based on a universal testing machine, acquire tensile property data corresponding to the metal cutting sample; and based on a hardness tester, acquire hardness data corresponding to the hardness tester.

[0050] Based on the content of the intelligent analysis system for the type of the automobile body metal material in the embodiments of the present application, it can be known that the embodiments of the present application can use multi-dimensional feature data including material composition data, crystal phase microstructure images, tensile property data, and hardness data as the basis for type analysis and analysis, and calculate the multi-dimensional similarity between the multi-dimensional feature data and the material samples in the preset material sample library based on the preset material sample library and the pre-trained material type reverse modeling algorithm, so as to make intelligent comprehensive decisions, so that the intelligent analysis result of the type of the automobile body metal material has scientific data support, and the accuracy and reliability of the analysis result are greatly improved.

[0051] Moreover, the embodiments of the present application can apply a weighted fusion algorithm to the material reverse analysis process, dynamically determine the weight through the pre-trained weight prediction model, so as to accurately measure the contribution degree of different data types in determining different material types, realize information complementation and fusion, and solve the misjudgment problem possibly caused by a single data source.

[0052] Moreover, the embodiment of the present application can intelligently and dynamically adjust the weight of each dimension feature according to the preliminary characteristics (such as high aluminum content or high hardness) of the metal cutting sample based on the weight prediction model, automatically adapt the distinguishing emphasis of different materials such as high-strength steel, aluminum alloy and cast iron, and thus improve the generalization ability of the system in processing various metal materials.

[0053] Each embodiment in the present application is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly explains the difference from other embodiments. Especially, for the embodiment of the intelligent analysis system of the metal material type of the automobile body, since it is basically similar to the embodiment of the intelligent analysis method of the metal material type of the automobile body, the description is relatively simple, and the related parts can be referred to the part of the method embodiment.

[0054] Please refer to Figure 5 The embodiment of the present application shown provides a structure schematic diagram of an electronic device based on a start control device.

[0055] As Figure 5 The electronic device 500 can include at least one processor 510, at least one network interface 540, a user interface 530, a memory 550 and at least one communication bus 520.

[0056] The communication bus 520 can be used to realize the connection and communication of the above-mentioned components.

[0057] The user interface 530 can include a key, and the optional user interface can further include a standard wired interface, a wireless interface.

[0058] The network interface 540 can include but is not limited to a Bluetooth module, an NFC module, a ZigBee module and a UWB module.

[0059] The processor 510 can include one or more processing cores. The processor 510 connects various parts in the electronic device 500 through various interfaces and lines, executes various functions of the electronic device 500 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 550, and calling data stored in the memory 550. Optionally, the processor 510 can be realized by at least one of a DSP, a FPGA and a PLA. The processor 510 can be integrated with one or a combination of a CPU and a GPU. The CPU mainly processes an operating system, a user interface and an application program; and the GPU is responsible for rendering and drawing the content to be displayed on the display screen.

[0060] The memory 550 can include a RAM and can also include a ROM. Optionally, the memory 550 includes a non-transitory computer-readable medium. The memory 550 can be used to store instructions, programs, codes, code sets, or instruction sets. The memory 550 can include a program storage area and a data storage area, wherein the program storage area can store the instructions for implementing the operating system, the instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), the instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store the data involved in the above-mentioned various method embodiments, etc. The memory 550 can also be at least one storage device located away from the aforementioned processor 510. The memory 550, as a computer storage medium, can include an operating system, a communication module, a user interface module, and an intelligent analysis application program of the automobile body metal material type. The processor 510 can be used to invoke the intelligent analysis application program of the automobile body metal material type stored in the memory 550, and execute the steps of the intelligent analysis of the automobile body metal material type mentioned in the foregoing embodiments.

[0061] The embodiments of the present application also provide a computer readable storage medium, which stores instructions, and when the instructions are run on a computer or a processor, the computer or the processor executes the steps of one or more of the above-mentioned embodiments. Figures 2-4 The above-mentioned various constituent modules of the electronic device can be stored in the computer readable storage medium if they are realized in the form of software function units and sold or used as independent products.

[0062] In the above embodiments, all or part of the methods can be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the methods can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the embodiments of the present application are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in or transmitted from a computer readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center through a wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes one or more available media sets. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a digital versatile disc (DVD)), or a semiconductor medium (for example, a solid state disk (SSD)), etc.

[0063] A person of ordinary skill in the art can understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware, which can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above embodiments. The storage medium includes ROM, RAM, magnetic or optical disc, and other program code storage media. In the case of no conflict, the technical features in the embodiments and the implementations can be combined arbitrarily.

[0064] The above embodiments are merely preferred embodiments of the present application and do not limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements of the technical solutions of the present application made by a person of ordinary skill in the art should fall within the protection scope of the claims of the present application.

Claims

1. An intelligent analysis method of a car body metal material type, characterized by, The method comprises the following steps: acquiring a metal cutting sample corresponding to a target automobile body; performing multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, the multi-dimensional feature data including material composition data, crystal phase organization image, tensile property data, and hardness data; based on a preset material sample library, determining a fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library, the fusion similarity being a similarity determined after the multi-dimensional feature data is weighted and fused based on a pre-trained weight prediction model and the weight corresponding to the multi-dimensional feature data is obtained; determining a model matching probability corresponding to each candidate material according to the fusion similarity; based on the model matching probability corresponding to each candidate material, obtaining an optimal analytical result corresponding to the target automobile body.

2. The method of claim 1, wherein, Before the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library is determined based on the preset material sample library, the method further comprises the following steps: preprocessing the crystal phase organization image to obtain a preprocessed crystal phase organization image; based on a pre-trained convolutional neural network model, extracting a crystal phase feature vector corresponding to the preprocessed crystal phase organization image.

3. The method of claim 2, wherein, The preset material sample library includes a plurality of candidate materials; determining the fusion similarity between the multi-dimensional feature data and each candidate material in the preset material sample library based on the preset material sample library comprises the following steps: determining any one candidate material in the preset material sample library; based on the material composition data, determining a composition similarity between the material composition data and the any one candidate material; based on the crystal phase feature vector corresponding to the preprocessed crystal phase organization image, determining a crystal phase similarity between the crystal phase feature vector and the any one candidate material; based on the tensile property data, determining a tensile property similarity between the tensile property data and the any one candidate material; based on the hardness data, determining a hardness similarity between the hardness data and the any one candidate material; performing weighted fusion processing on the composition similarity, the crystal phase similarity, the tensile property similarity, and the hardness similarity to obtain the fusion similarity between the multi-dimensional feature data and the any one candidate material.

4. The method of claim 3, wherein, The method comprises the following steps: based on the pre-trained weight prediction model, performing weight detection on the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data to obtain the weight corresponding to each of the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data.

5. The method of claim 4, wherein, The method comprises the following steps: based on the pre-trained weight prediction model, performing weight detection on the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data to obtain the weight corresponding to each of the material composition data, the crystal phase feature vector, the tensile property data, and the hardness data. obtain a weight training set, the weight training set comprising a plurality of training samples, each training sample comprising training data and a preset weight corresponding to the training data, the training data comprising material composition training data, crystal phase training feature vectors, tensile property training data, and hardness training data; perform supervised training on the weight prediction model based on the weight training set to obtain a pre-trained weight prediction model.

6. The method of claim 5, wherein, The weight prediction model is a machine learning model trained through supervised learning, and the machine learning model comprises an XGBoost model.

7. The method of claim 1, wherein, The multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data comprises: obtaining material composition data corresponding to the metal cutting sample based on an X-ray fluorescence spectrometer; obtaining a crystal phase organization image corresponding to the metal cutting sample based on a scanning electron microscope; obtaining tensile property data corresponding to the metal cutting sample based on a universal testing machine; obtaining hardness data corresponding to the metal cutting sample based on a hardness tester.

8. An intelligent analysis system for a car body metal material type, characterized by, comprise: a sample obtaining module configured to obtain a metal cutting sample corresponding to a target automobile body; a detection module configured to perform multi-modal detection on the metal cutting sample to obtain multi-dimensional feature data, the multi-dimensional feature data comprising material composition data, a crystal phase organization image, tensile property data, and hardness data; a similarity calculation module configured to determine a fusion similarity between the multi-dimensional feature data and each candidate material in a preset material sample library based on the preset material sample library, the fusion similarity being a similarity determined after weighted fusion processing of the multi-dimensional feature data based on a pre-trained weight prediction model obtaining a weight corresponding to the multi-dimensional feature data and according to the weight; a probability determination module configured to determine a model matching probability corresponding to each candidate material according to the fusion similarity; a model matching module configured to obtain an optimal analytical result corresponding to the target automobile body based on the model matching probabilities corresponding to the plurality of candidate materials.

9. An electronic device, comprising: comprise a processor and a memory; the processor is connected to the memory; the memory is configured to store executable program codes; the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the method according to any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method according to any one of claims 1-7.

Images