Vehicle part identification method, electronic device, storage medium and program product

Abstract

The invention provides a vehicle part identification method, electronic equipment, a storage medium and a program product. The method comprises the steps that a first identification result is determined according to part geometric parameters of a vehicle model and a first identification system; determining a second identification result according to the first identification result and a second identification system; and outputting an identification result of the part according to the second identification result. According to the invention, preliminary recognition and classification of the parts are realized by determining the first recognition result, then precise recognition and classification of the parts are realized by determining the second recognition result in combination with the second recognition system, and finally the recognition result of the parts is output according to the second recognition result. The parts in the informal model of the vehicle are efficiently and accurately identified and classified in the research and development stage of the vehicle, the research and development efficiency of the vehicle is improved, and the research and development cost of the vehicle is saved.

Description

Technical Field

[0001] The present invention relates to the field of vehicles, and in particular to a vehicle parts identification method, electronic equipment, computer-readable storage medium and computer program product. Background Art

[0002] The identification and classification of vehicle parts are very important for vehicle research and development. In related technologies, the identification and classification of parts are achieved by determining the target CAE (Computer Aided Engineering) model, extracting all parts and their names, creating multiple target classification sets, and automatically adding parts to the corresponding sets. Alternatively, samples of sample parts are used as input parameters for learning and training to obtain a part recognition model, and the features of the parts to be recognized are input into the part recognition model to achieve part recognition and classification.

[0003] However, the above method for identifying and classifying vehicle parts is only applicable to parts with naming specifications, cannot identify parts without naming specifications, cannot be applied to process models before the official version of the vehicle model, or requires labeling and training of a large number of images, which is costly. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0005] To this end, an object of the present invention is to propose a vehicle parts identification method, which can efficiently and accurately identify and classify parts in an informal version model of a vehicle during the vehicle's research and development stage, thereby improving the vehicle's research and development efficiency and saving the vehicle's research and development costs.

[0006] To this end, a second objective of the present invention is to provide an electronic device.

[0007] To this end, a third object of the present invention is to provide a computer-readable storage medium.

[0008] To this end, a fourth object of the present invention is to provide a computer program product.

[0009] In order to achieve the above-mentioned purpose, an embodiment of the first aspect of the present invention proposes a vehicle parts identification method, the method comprising: determining a first recognition result based on the component geometric parameters of the vehicle model and a first recognition system; determining a second recognition result based on the first recognition result and a second recognition system; and outputting the recognition result of the component based on the second recognition result.

[0010] According to the vehicle parts identification method of the embodiment of the present invention, a first identification result is determined by the component geometric parameters of the vehicle model and a first identification system to achieve preliminary identification and classification of the components, a second identification result is determined based on the first identification result and the second identification system to achieve accurate identification and classification of the components, and the identification result of the components is output based on the second identification result to achieve efficient and accurate identification and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development cost of the vehicle.

[0011] In some embodiments, determining a first recognition result based on component geometric parameters of a vehicle model and a first recognition system includes: determining a first input feature set based on the component geometric parameters; and determining the first recognition result based on the first input feature set and the first recognition system.

[0012] In some embodiments, determining a second recognition result based on the first recognition result and the second recognition system includes: obtaining the number of first components of the first recognition type in the first recognition result and the number of second components of the second recognition type in the first recognition result; determining a first component recognition rate based on the first component number and the second component number; when the first component recognition rate is greater than or equal to a preset recognition rate threshold, determining the second recognition result based on the first recognition result and the second recognition system.

[0013] In some embodiments, determining the second recognition result based on the first recognition result and the second recognition system includes: obtaining second component geometric parameters of the second recognition type in the first recognition result and third component geometric parameters of the third recognition type in the first recognition result; determining a second input feature set based on the second component geometric parameters and the third component geometric parameters; and determining the second recognition result based on the second input feature set and the second recognition system.

[0014] In some embodiments, after determining the component recognition rate based on the first component quantity and the second component quantity, it also includes: when the first component recognition rate is less than a preset recognition rate threshold, updating the first recognition system according to the component geometric parameters of the first recognition type in the first recognition result.

[0015] In some embodiments, outputting the recognition result of the component according to the second recognition result includes: obtaining the updated number of the first component of the first recognition type in the second recognition result and the updated number of the third component of the third recognition type in the second recognition result; determining the first component error rate according to the updated number of the first component and the updated number of the third component; when the error rate of the first component is less than or equal to a preset error rate threshold, outputting the recognition result of the component.

[0016] In some embodiments, after determining the component error recognition rate based on the second component update quantity and the third component update quantity, it also includes: when the first component error recognition rate is greater than a preset error recognition rate threshold, updating the second recognition system according to the component geometric parameters of the second recognition type in the second recognition result.

[0017] In some embodiments, before determining the first recognition result based on the geometric parameters of the components of the vehicle model and the first recognition system, it also includes: obtaining the geometric parameters of the components to be identified; determining the first cluster center based on the geometric parameters of the components to be identified; and determining the first recognition system based on the first cluster center and the vehicle model.

[0018] In some embodiments, determining a first recognition system based on the first cluster center and the vehicle model includes: acquiring discrete geometric parameters of the vehicle model; determining a third input feature set based on the discrete geometric parameters; and determining the first recognition system based on the third input feature set and the first cluster center.

[0019] In some embodiments, after determining the first recognition system according to the third input feature set and the first cluster center, it includes: determining a first distance according to the third input feature set and the first cluster center; obtaining a third recognition result in which the first distance is less than a preset distance threshold; determining a first component recognition rate according to the third recognition result; when the first component recognition rate is less than a preset recognition rate threshold, updating the first cluster center according to the third recognition result.

[0020] In some embodiments, after determining the first cluster center according to the geometric parameters of the parts to be identified, the method further includes: acquiring additional geometric parameters of the parts to be identified; and determining the second cluster center according to the additional geometric parameters.

[0021] In some embodiments, after determining the first component recognition rate according to the third recognition result, it also includes: when the first component recognition rate is greater than or equal to a preset recognition rate threshold, determining a fourth input feature set according to the third recognition result; determining a second distance according to the fourth input feature set and the second cluster center; obtaining a fourth recognition result in which the second distance is less than the preset distance threshold; determining the first component error recognition rate according to the fourth recognition result; when the first component error recognition rate is less than or equal to the preset error recognition rate threshold, determining a second recognition system according to the second cluster center.

[0022] In some embodiments, after determining the first component error recognition rate according to the fourth recognition result, it also includes: when the first component error recognition rate is greater than a preset error recognition rate threshold, updating the second cluster center according to the fourth recognition result.

[0023] In order to achieve the above-mentioned purpose, an embodiment of the second aspect of the present invention proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a vehicle parts identification program that can be executed by the at least one processor, and when the vehicle parts identification program is executed by the at least one processor, the at least one processor executes the vehicle parts identification method as described in the above-mentioned embodiment.

[0024] According to the electronic device of the embodiment of the present invention, a first recognition result is determined by the geometric parameters of the components of the vehicle model and the first recognition system to achieve preliminary recognition and classification of the components, a second recognition result is determined based on the first recognition result and the second recognition system to achieve accurate recognition and classification of the components, and the recognition result of the components is output based on the second recognition result to achieve efficient and accurate recognition and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0025] In order to achieve the above-mentioned purpose, an embodiment of the third aspect of the present invention proposes a computer-readable storage medium, on which a vehicle parts identification program is stored. When the vehicle parts identification program is executed by a processor, a device equipped with the vehicle parts identification program implements the vehicle parts identification method described in the above-mentioned embodiment.

[0026] In order to achieve the above-mentioned purpose, an embodiment of the fourth aspect of the present invention proposes a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the vehicle part identification method as described in the above-mentioned embodiment.

[0027] According to the computer program product of the embodiment of the present invention, a first recognition result is determined by the geometric parameters of the components of the vehicle model and a first recognition system to achieve preliminary recognition and classification of the components, a second recognition result is determined based on the first recognition result and the second recognition system to achieve accurate recognition and classification of the components, and the recognition result of the components is output based on the second recognition result to achieve efficient and accurate recognition and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0028] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 is a block diagram of a CAE simulation process according to an embodiment of the present invention; Figure 2 is a block diagram of a computer device carrier according to an identification system and a visualization interface according to an embodiment of the present invention; Figure 3 is a flow chart of a vehicle parts identification method according to an embodiment of the present invention; Figure 4 is a flow chart of an identification system and online database update according to an embodiment of the present invention; Figure 5 is a schematic diagram of a visual interface of an identification system according to an embodiment of the present invention; Figure 6 is a flow chart of constructing an identification system according to an embodiment of the present invention; Figure 7 is a flowchart of database construction and updating of an identification system according to an embodiment of the present invention; Figure 8 is a block diagram of an electronic device according to an embodiment of the present invention.

[0030] Reference numerals: Model pre-processing 230; simulation calculation 231; result output 232; Model collection 233; Component assembly 234; Component naming 235; Geometry cleanup 236; Geometry closure 237; Mesh generation 238; Calculation settings 239; Calculation submission 240; Result analysis 241; Report writing 242; Computer equipment 220; User memory 221; display 222; processor 227; online memory 226; communication interface 228; Identification system 224; online management system 226; more functions 210; Project navigation bar 200; function selection bar 206; model input bar 211; recognition result bar 212; parameter setting 219; Model import 201; download database 202; execute 203; verify recognition result 204; upload database 205; delete unnecessary parts 207; combine and name parts 208; display specific parts 209; prompt 213; model dissection 214; part one 215; part two 216; part three 217; Processor 101; Memory 102; Electronic device 100. DETAILED DESCRIPTION

[0031] Embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. Embodiments of the present invention are described in detail below.

[0032] In the early stages of vehicle development, in order to evaluate the potential of new models, engineers need to disassemble the digital model of existing vehicles, assemble the disassembled parts into the model to be analyzed, and use CAE (Computer Aided Engineering) tools to perform finite element analysis and CFD (Computational Fluid Dynamics) simulation on the assembled and unnamed car model. Model pre-processing is a necessary step in simulation analysis. It is crucial to remove unnecessary parts and accurately name parts. The pre-processing step can effectively improve the mesh quality and simulation accuracy of the model. The use of pre-processing automation methods can improve modeling efficiency and simulation accuracy, among which automatic identification and classification of parts technology is the key to the automation of pre-processing work.

[0033] In the related art, for example, automation of component classification in a CAE model can be achieved by determining a target CAE model, extracting all components and their names, creating multiple target classification sets, and automatically adding components to corresponding sets.

[0034] However, the above method is only applicable to parts with naming specifications and cannot identify parts without naming specifications. Because the pace of vehicle model development is fast and the model is replaced frequently, it is difficult to standardize the naming of parts before the official version of the styling is confirmed. Therefore, the above method cannot be applied to the process model before the official version of the vehicle, that is, it cannot be applied to the identification and classification of parts in the informal version of the vehicle during the vehicle development stage.

[0035] For another example, by obtaining label data samples, two-dimensional image feature samples, two-dimensional drawing feature samples and three-dimensional model feature samples of sample parts, these samples are used as input parameters for learning and training to obtain a part recognition model, and the two-dimensional image features in the two-dimensional image data of the parts to be recognized, the two-dimensional drawing features in the two-dimensional drawing data, and the three-dimensional model features in the three-dimensional model data are collected; these features are input into the part recognition model to obtain the label data recognition results of the parts to be recognized, thereby realizing the recognition and classification of parts.

[0036] However, the above method requires labeling and training of a large number of images, which is costly.

[0037] Therefore, the vehicle parts identification method of the embodiment of the present invention is adopted to determine the first identification result through the geometric parameters of the parts of the vehicle model and the first identification system to achieve preliminary identification and classification of the parts, and determine the second identification result according to the first identification result and the second identification system to achieve accurate identification and classification of the parts, and output the identification result of the parts according to the second identification result to achieve efficient and accurate identification and classification of the parts in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development cost of the vehicle.

[0038] The above-mentioned vehicle parts identification method is implemented based on an identification system, and the modules related to the identification system are illustrated below.

[0039] like Figure 1 FIG. 2 is a flowchart of a CAE simulation process according to an embodiment of the present invention. The flowchart of the CAE simulation process includes: model pre-processing 230 , simulation calculation 231 and result output 232 .

[0040] Among them, model pre-processing 230 accounts for 80% of the work in CAE simulation analysis, including: digital model collection 233, component assembly 234, component naming 235, geometry cleaning 236, geometry closure 237 and mesh generation 238; component assembly 234, component naming 235 and geometry cleaning 236 account for 30% of the work in model pre-processing 230. The use of the recognition system in these three places is expected to improve the efficiency of model pre-processing 230 by 20%; simulation calculation includes calculation setting 239 and calculation submission 240, accounting for 15% of the work in CAE simulation analysis; result output 232 includes: result analysis 241 and report writing 242, accounting for 5% of the work in CAE simulation analysis. It can be seen that the use of the recognition system in the model pre-processing 230 can greatly improve the efficiency of CAE simulation analysis and optimize the entire simulation analysis process.

[0041] like Figure 2, which is a block diagram of a computer device carrier of an identification system and a visualization interface of an embodiment of the present invention. The identification system 224 is based on a dual memory architecture, and realizes data transmission between the user memory 221 and the online memory 226 through the communication interface 228. After the user inputs the model data of the vehicle into the identification system 224, the identification system 224 will display the identification result in real time on the display 222, which is convenient for the user to quickly verify and check. After passing the verification, the database can be updated. After the user confirms that updating the database can improve the performance of the identification system, the updated database can be submitted to the online management system 225.

[0042] Based on the above recognition system, the following Figure 1-Figure 7 A vehicle parts identification method according to an embodiment of the present invention is described.

[0043] like Figure 3 FIG. 1 is a flow chart of a vehicle parts identification method according to an embodiment of the present invention. The vehicle parts identification method according to the embodiment of the present invention at least includes steps S1 to S3.

[0044] Step S1, determining a first recognition result according to geometric parameters of components of a vehicle model and a first recognition system.

[0045] In an embodiment, geometric parameters are quantities of component shapes, sizes, and positional relationships, such as surface area and volume parameters of components; the first recognition system is a system for preliminary recognition and classification of components; the first recognition result is a preliminary recognition and classification result of components; the preliminary recognition and classification result of components is determined based on the surface area and volume parameters of the vehicle model and the system for preliminary recognition and classification of components, for example, based on the continuity and closure of the geometric body, the components are discretized, and the surface area and volume parameters of all components are extracted to form an input feature set for the first recognition system. The first recognition system uses a real-time updated database to process the input feature set by fast clustering and controlling the clustering radius to obtain preliminary recognition and classification results, so that the recognition and classification of components do not rely on the standardized naming of components, and prepares for the efficient and accurate recognition and classification of components in the informal version model of the vehicle during the vehicle R&D stage.

[0046] Step S2, determining a second recognition result according to the first recognition result and the second recognition system.

[0047] In an embodiment, the second recognition system is a system for accurately identifying and classifying parts; the second recognition result is determined based on the preliminary recognition and classification results of the parts and the second recognition system, for example, the number of parts of the ideal recognition type and the number of parts of the missed recognition type in the preliminary recognition and classification results are obtained, and a recognition rate is calculated, and the recognition rate of parts of the ideal recognition type is quantitatively evaluated by the recognition rate. When the recognition rate reaches a critical value for accurately identifying and classifying parts, the geometric parameters of parts of the ideal recognition type and the geometric parameters of parts of the misjudged recognition type in the preliminary recognition and classification results are obtained, and an input feature set is determined based on the geometric parameters of parts of the ideal recognition type and the geometric parameters of parts of the misjudged recognition type. The second recognition system uses a real-time updated database to process the feature set by fast clustering and controlling the clustering radius to obtain accurate recognition and classification results of parts, i.e., the second recognition result, so as to achieve efficient and accurate recognition and classification of parts.

[0048] Step S3: outputting the recognition result of the component according to the second recognition result.

[0049] Output the recognition results of the parts according to the accurate recognition and classification results of the parts, for example, obtain the number of parts of the ideal recognition type and the number of parts of the misjudged recognition type in the accurate recognition and classification results of the parts, calculate the error rate, the error rate is the proportion of the number of parts of the misjudged recognition type, and when the error rate is less than or equal to a preset error rate threshold, for example, less than or equal to 0.01%, output the recognition results of the parts, for example, display the results on a display, so as to achieve the output of the results of accurate recognition and classification of the parts, and the above process does not require the use of a large number of pictures for marking and training, thereby achieving efficient and accurate recognition and classification of parts in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0050] According to the vehicle parts identification method of the embodiment of the present invention, a first identification result is determined by the component geometric parameters of the vehicle model and a first identification system to achieve preliminary identification and classification of the components, a second identification result is determined based on the first identification result and the second identification system to achieve accurate identification and classification of the components, and the identification result of the components is output based on the second identification result to achieve efficient and accurate identification and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development cost of the vehicle.

[0051] In some embodiments, determining a first recognition result based on component geometric parameters of the vehicle model and a first recognition system includes: determining a first input feature set based on the component geometric parameters; and determining the first recognition result based on the first input feature set and the first recognition system.

[0052] In the embodiment, the geometric parameters are quantities of the shape, size and positional relationship of the parts, such as the surface area and volume parameters of the parts; the first recognition system is a system for preliminary recognition and classification of the parts; the first input feature set is a feature set of the parts extracted from the geometric parameters of the parts; the first recognition result is a preliminary recognition and classification result of the parts; according to the continuity and closure of the geometric body, the parts are discretized, and the surface area and volume parameters of all the parts are extracted to form the input feature set X={x1, x2,...,x n}, where the characteristic variable x i is a two-dimensional vector { x i 0 , x i 1}, n It is the number of types of discrete parts, which provides data basis for preliminary identification and classification of parts, so that the identification and classification of parts do not rely on the naming of parts, and prepares for the efficient and accurate identification and classification of parts in the informal version model of the vehicle during the vehicle's R&D stage.

[0053] The first recognition system uses a real-time updated database to process the input feature set X by fast clustering and controlling the cluster radius to obtain preliminary recognition and classification results. However, the incompleteness of the database and the use of surface area and volume parameters for clustering may lead to differences between the recognition results and actual needs. The preliminary recognition and classification results, namely the first recognition results, are divided into three types: missed recognition is set to R1, ideal recognition is set to R2, and misjudgment recognition is set to R3. The first recognition result is determined by the first input feature set and the first recognition system to achieve preliminary recognition and classification of parts.

[0054] In some embodiments, determining a second recognition result based on a first recognition result and a second recognition system includes: obtaining the number of first components of a first recognition type in the first recognition result and the number of second components of a second recognition type in the first recognition result; determining a first component recognition rate based on the first component number and the second component number; when the first component recognition rate is greater than or equal to a preset recognition rate threshold, determining a second recognition result based on the first recognition result and the second recognition system.

[0055] In the embodiment, the first component is a component of an ideal recognition type; the second recognition system is a system for accurately recognizing and classifying components; the first recognition type is ideal recognition (i.e., R2); the second recognition type is missed recognition (i.e., R1); the second component is a component of a missed recognition type; the second recognition result is a result of accurate recognition and classification of the component; the preset recognition rate threshold is a critical value for accurately recognizing and classifying the component; the number of components of the ideal recognition type and the number of components of the missed recognition type in the preliminary recognition and classification results are obtained to provide a data basis for determining the recognition rate of components of the ideal recognition type; the recognition rate is set to , through the recognition rate To quantitatively evaluate the recognition rate of parts of the ideal recognition type, where N1 and N2 are the number of parts in the R1 and R2 types, and provide a theoretical basis for realizing a system for accurate recognition and classification of parts; the recognition rate of parts of the ideal recognition type is greater than or equal to the preset recognition rate threshold. For example, the preset recognition rate threshold is 98%, and it is believed that the accuracy of the parts is high, and the parts can be more accurately identified and classified. The second recognition result is determined based on the preliminary recognition and classification results and the second recognition system to achieve efficient and accurate recognition and classification of parts.

[0056] In some embodiments, determining a second recognition result based on the first recognition result and the second recognition system includes: obtaining second component geometric parameters of the second recognition type in the first recognition result and third component geometric parameters of the third recognition type in the first recognition result; determining a second input feature set based on the second component geometric parameters and the third component geometric parameters; and determining the second recognition result based on the second input feature set and the second recognition system.

[0057] In the embodiment, the third recognition type is a misjudged recognition (i.e., R3), and the third component is a component of a misjudged recognition type; geometric parameters of components of an ideal recognition type and geometric parameters of components of a misjudged recognition type in the preliminary recognition and classification results are obtained to prepare data for determining a second input feature set; the second input feature set Y={y1, y2,...,y m}, each eigenvector y j Depend on k The multidimensional vector corresponding to the categories is { y j 1 , y j 2 , ......, y j k}constitute, mThe number of types of parts in the preliminary recognition result is used to prepare data for determining the second recognition result; the second recognition result is determined according to the second input feature set and the second recognition system. The second recognition system uses a real-time updated database to process the feature set Y by fast clustering and controlling the clustering radius to obtain accurate recognition and classification results of parts, that is, the second recognition result. There are two situations for this result: (Ideal Identification), (Misjudgment identification), the number of parts corresponding to the type is , , providing data basis for determining the error recognition rate; In some embodiments, after determining the component recognition rate according to the first component quantity and the second component quantity, it also includes: when the first component recognition rate is less than a preset recognition rate threshold, updating the first recognition system according to the component geometric parameters of the first recognition type in the first recognition result.

[0058] In an embodiment, the recognition rate of parts of an ideal recognition type is less than a preset recognition rate threshold. For example, the preset recognition rate threshold is 98%. It is considered that the accuracy of the parts is low and the parts cannot be more accurately identified and classified. It is necessary to update the geometric parameters of the parts of the ideal recognition type to the database in the first recognition system to update the geometric parameter data of the parts of the ideal recognition type in the first recognition system, so as to provide reliable data support for efficient and accurate identification and classification of parts.

[0059] In some embodiments, outputting the recognition result of the component based on the second recognition result includes: obtaining the first component update number of the first recognition type in the second recognition result and the third component update number of the third recognition type in the second recognition result; determining the first component error rate based on the first component update number and the third component update number; when the first component error rate is less than or equal to a preset error rate threshold, outputting the recognition result of the component.

[0060] In the embodiment, the preset error recognition rate threshold is a critical value for determining whether to output the recognition result of the component; the first component error recognition rate is the proportion of the number of components of the misjudged recognition type; the number of components of the ideal recognition type in the accurate recognition and classification results of the components, that is, the first component update number, and the number of components of the misjudged recognition type, that is, the third component update number, are obtained to provide a data basis for determining the first component error recognition rate; the first component error recognition rate is determined according to the number of components of the ideal recognition type and the number of components of the misjudged recognition type, and the error recognition rate is set to , ,in, and They are and The number of part types in the system is used to achieve quantitative evaluation of the recognition and classification results; when the first component error recognition rate is less than or equal to a preset error recognition rate threshold, for example, less than or equal to 0.01%, the component recognition result is output, for example, the result is displayed on a display, so as to achieve the output of the result of accurate recognition and classification of the component. For users, the recognition and classification of components can be completed by operating the visual interface of the recognition system, without the need for manual recognition and classification of components, making the recognition and classification of components more convenient and intelligent. Specifically, Figure 5 As shown, a schematic diagram of a visual interface of a recognition system according to an embodiment of the present invention. The user can click on a function button in the function selection bar 206, which includes: deleting non-essential parts 207, combining and naming parts 208, and displaying specific parts 209, etc., and will switch to the parameter setting 219 page of the function; after the user completes the parameter setting, click the model import 201 button, select the model file to be processed, and the model will be displayed in the model input bar 211. The user can immediately check whether the imported model is accurate, and continue to the next step after ensuring that it is correct; the user clicks the execute 203 button to start the recognition function, and after the recognition is completed, the result will be immediately displayed on the right side of the recognition result bar 212, including: model dissection 214, display of parts such as part one 215 and part two 216, wherein the model dissection 214 is a diagram generated after dissecting the parts in the vehicle model, part one 215 is a bolt part, and part two 216 is a buckle part, and the user can quickly view the recognition result; click the recognition result verification 204 button, and the system guides the user to confirm twice. First, enter the "Missed Recognition" option, and the user selects the part that has not been correctly recognized in the recognition result column 212, and the system automatically calculates its first component recognition rate. When the first component recognition rate reaches 98%, the system switches to the "Misrecognition" option. After the user selects the incorrectly recognized part, the first component error recognition rate is automatically calculated. If the first component error recognition rate is lower than 0.01%, the result verification is completed and output to the left side of the recognition result column 212 for display, including the database storage status, database data volume, number of components, recognition rate and error recognition rate, etc.; after user verification and verification, click the Upload Database 205 button to upload the recognition results to the online database to update the system. After completing the output of the results of accurate recognition and classification of components, the user can choose to upload the updated database online to improve the recognition and classification effects of the recognition system. Specifically, if Figure 4 FIG. 1 is a flowchart of an identification system and online database update process according to an embodiment of the present invention. After receiving the update request, the administrator evaluates the impact of the submitted data on the performance of the identification system. If the verification is passed, the data is stored in the online database; otherwise, if the standard is not met, the user is notified of the reason and the original database is maintained. The process includes at least steps S50-S58.

[0061] Step S50: model input.

[0062] Step S55, system identification.

[0063] Step S52, output the recognition result.

[0064] Step S53, determine whether the user uploads data. If yes, execute step S55; otherwise, execute step S54.

[0065] Step S54: do not update the online database.

[0066] Step S55, generating a local database.

[0067] Step S56, online management system.

[0068] Step S57, determine whether the uploaded data is reasonable. If so, execute step S58; otherwise, execute step S54.

[0069] Step S58, storing the data in the hard disk, updating the online database, and providing data to the recognition system.

[0070] The first recognition system and the second recognition system can download the parameters of the parts to the online database, and the corresponding users can choose to update and store the recognition results of the first recognition system and the second recognition system to the online database to improve the data in the recognition system database and ensure the accuracy of recognition and classification.

[0071] In some embodiments, after determining the component error rate based on the second component update quantity and the third component update quantity, it also includes: when the first component error rate is greater than a preset error rate threshold, updating the second recognition system according to the component geometric parameters of the second recognition type in the second recognition result.

[0072] In an embodiment, when the error recognition rate of the first component is greater than a preset error recognition rate threshold, for example, greater than 0.01%, the geometric parameters of the components of the ideal recognition type in the accurate recognition and classification results of the components are updated to the database in the second recognition system, so as to update the geometric parameter data of the components of the ideal recognition type in the second recognition system, and provide reliable data support for efficient and accurate recognition and classification of components.

[0073] In some embodiments, before determining the first recognition result based on the geometric parameters of the components of the vehicle model and the first recognition system, it also includes: obtaining the geometric parameters of the components to be identified; determining the first cluster center based on the geometric parameters of the components to be identified; and determining the first recognition system based on the first cluster center and the vehicle model.

[0074] In the embodiment, the cluster center refers to a special sample point representing a certain class in cluster analysis, and other samples are determined to belong to the class by calculating the distance from it; before determining the first recognition result based on the geometric parameters of the components of the vehicle model and the first recognition system, the geometric parameters of the components to be recognized are obtained, for example, the surface area and volume parameters of the input components are extracted to provide a data basis for determining the first cluster center; the first cluster center is determined based on the surface area and volume parameters of the components to be recognized , where the characteristic variable is a two-dimensional vector , l It is the number of input parts, which is used to prepare data for determining the first recognition system; the first recognition system is determined according to the first cluster center and the input vehicle model to achieve preliminary recognition and classification of parts through the first recognition system.

[0075] In some embodiments, determining a first recognition system based on a first cluster center and a vehicle model includes: acquiring discrete geometric parameters of the vehicle model; determining a third input feature set based on the discrete geometric parameters; and determining the first recognition system based on the third input feature set and the first cluster center.

[0076] In the embodiment, a vehicle model is input, discrete parts are formed by continuum segmentation, and surface area and volume parameters of the discrete parts are extracted to prepare for output for determining a third input feature set; based on the extracted surface area and volume parameters of the discrete parts, an input feature set, i.e., the third input feature set X={ x 1, x 2,..., x n},in, n is the number of discrete parts, which is used to prepare data for determining the first recognition system and enabling the first recognition system to quickly complete the preliminary recognition and classification of parts through a small number of features; according to the third input feature set X and the first cluster center A first identification system is determined to achieve preliminary identification and classification of parts through the first identification system, and it is determined that the first identification system does not rely on standardized naming of parts, thereby achieving preliminary identification and classification of parts in an informal version model of the vehicle during the research and development stage of the vehicle.

[0077] In some embodiments, after determining the first recognition system based on the third input feature set and the first cluster center, it includes: determining the first distance based on the third input feature set and the first cluster center; obtaining a third recognition result in which the first distance is less than a preset distance threshold; determining a first component recognition rate based on the third recognition result; when the first component recognition rate is less than a preset recognition rate threshold, updating the first cluster center based on the third recognition result.

[0078] In the embodiment, the distance threshold is a critical value for determining whether to use it as a preliminary recognition result or an accurate recognition result, and is set to 0.01; the third recognition result is the preliminary recognition result of the first recognition system, including R 2 and R 3; According to the third input feature set X and the first cluster center Determine the first distance D={ d 1, d 2,..., d n}, to provide data basis for determining preliminary recognition and classification results; obtaining recognition results where the first distance D is less than a preset distance threshold of 0.01, including R 2 and R 3 is used as the preliminary recognition result, i.e., the third recognition result, to serve as the data basis for determining the recognition rate of the first component; R 2. R 3 and manually completed R 1. Determine the first component recognition rate by classification, which is used as a basis for determining whether to update the first cluster center; when the first component recognition rate is less than a preset recognition rate threshold, such as 98%, update the first cluster center according to the third recognition result, for example, R The feature of 1 is added to the database of the first recognition system, and the new cluster center is also recorded as , its dimension is updated to , the iteration is stable Update to the online database, complete database construction and update, so as to supplement the accurate data in the first recognition system, and improve the accuracy of the data in the second recognition system through the iteration of the first cluster center.

[0079] In some embodiments, after determining the first cluster center according to the geometric parameters of the parts to be identified, the method further includes: obtaining additional geometric parameters of the parts to be identified; and determining the second cluster center according to the additional geometric parameters.

[0080] In an embodiment, the second cluster center is the initial cluster center of the second recognition system; extracting additional geometric parameters of the input parts, that is, additional feature parameters ( k ), prepare data for determining the second cluster center; form the initial cluster center set of the second recognition system, i.e., the second cluster center, according to the additional feature parameters , where the characteristic variable for k Dimensional vector , l It is the number of input parts to prepare data for determining the second recognition system.

[0081] In some embodiments, after determining the recognition rate of the first component according to the third recognition result, it also includes: when the recognition rate of the first component is greater than or equal to a preset recognition rate threshold, determining a fourth input feature set according to the third recognition result; determining a second distance according to the fourth input feature set and the second cluster center; obtaining a fourth recognition result in which the second distance is less than the preset distance threshold; determining the first component error recognition rate according to the fourth recognition result; when the first component error recognition rate is less than or equal to the preset error recognition rate threshold, determining a second recognition system according to the second cluster center.

[0082] In the embodiment, the fourth input feature set is the second recognition system input feature set; the second distance is the distance between the fourth input feature set and the parameter in the second cluster center; when the first component recognition rate is greater than or equal to the preset recognition rate threshold, for example, 98%, according to the third recognition result R 2 and R 3. Determine the fourth input feature set, for example, extract the minimum circumscribed sphere volume, maximum side length, minimum side length and other parameters of the parts in the recognition results to form the second recognition system input feature set, that is, the fourth input feature set Y={y1, y2,...,y m},in, m is the number of types of parts in the recognition result, in order to determine the second distance, and to enable the second recognition system to complete the accurate feature recognition task based on the multiple features of the preliminary recognition and classification results, so as to prepare data for improving the recognition effect; according to the fourth input feature set Y and the second cluster center Determine the second distance , prepare data for determining the fourth recognition result; obtain the second distance Less than the preset distance threshold of 0.01 As the accurate recognition result, i.e., the fourth recognition result, data is prepared for determining the error recognition rate of the first component; according to the fourth recognition result and manual completion The first component error rate is determined by classification to prepare data for determining the second recognition system and outputting the recognition result; when the first component error rate is less than or equal to the preset error rate threshold, for example, 0.01, the second recognition system is determined according to the second cluster center to achieve accurate component recognition and classification results output by the second recognition system.

[0083] In some embodiments, after determining the first component error recognition rate according to the fourth recognition result, it also includes: when the first component error recognition rate is greater than a preset error recognition rate threshold, updating the second cluster center according to the fourth recognition result.

[0084] In an embodiment, when the first component error recognition rate is greater than a preset error recognition rate threshold, such as 0.01, the second cluster center is updated according to the fourth recognition result, for example, The features of are added to the database of the second recognition system, and the new cluster center is also recorded as , its dimension is updated to , the iteration is stable Update to the online database, complete database construction and update, and iterate the second cluster center to improve the accuracy of the data in the second recognition system.

[0085] Reference below Figure 6 and Figure 7 The vehicle parts identification method according to the embodiment of the present invention is described in detail.

[0086] like Figure 6 FIG. 1 is a flow chart of an identification system construction according to an embodiment of the present invention. The vehicle parts identification method according to the embodiment of the present invention at least includes steps S10-S23.

[0087] Step S10, determining a first input feature set according to component geometric parameters.

[0088] Step S11, determining a first recognition result according to a first input feature set and a first recognition system.

[0089] Step S12, obtaining geometric parameters of a second component of a second recognition type in the first recognition result and geometric parameters of a third component of a third recognition type in the first recognition result.

[0090] Step S13, determining a second input feature set according to the second component geometric parameter and the third component geometric parameter.

[0091] Step S14, obtaining the quantity of first components of the first identification type in the first identification result and the quantity of second components of the second identification type in the first identification result.

[0092] Step S15, determining the first component recognition rate according to the first component quantity and the second component quantity.

[0093] Step S16, determining whether the first component recognition rate is greater than or equal to a preset recognition rate threshold. If so, executing step S18; otherwise, executing step S17.

[0094] Step S17, updating the first recognition system according to the geometric parameters of the components of the first recognition type in the first recognition result.

[0095] Step S18: determining a second recognition result according to the second input feature set and the second recognition system.

[0096] Step S19, obtaining the updated number of the second components of the second identification type in the second identification result and the updated number of the third components of the third identification type in the second identification result.

[0097] Step S20, determining the first component misrecognition rate according to the second component update quantity and the third component update quantity.

[0098] Step S21, determining whether the first component error recognition rate is less than or equal to a preset error recognition rate threshold. If so, executing step S23; otherwise, executing step S22.

[0099] Step S22: updating the second recognition system according to the geometric parameters of the components of the second recognition type in the second recognition result.

[0100] Step S23, outputting the recognition result of the parts.

[0101] like Figure 7 FIG. 1 is a flowchart of database construction and updating of an identification system according to an embodiment of the present invention. The vehicle parts identification method according to the embodiment of the present invention at least includes steps S30-S48.

[0102] Step S30, obtaining geometric parameters of the parts to be identified.

[0103] Step S31, determining the first cluster center according to the geometric parameters of the parts to be identified.

[0104] Step S32, obtaining discrete geometric parameters of the vehicle model.

[0105] Step S33: determining a third input feature set according to the discrete geometric parameters.

[0106] Step S34, obtaining additional geometric parameters of the parts to be identified.

[0107] Step S35, determining a second cluster center according to the additional geometric parameters.

[0108] Step S36: determining a first recognition system according to the third input feature set and the first cluster center.

[0109] Step S37: determining a first distance according to the third input feature set and the first cluster center.

[0110] Step S38, obtaining a third recognition result in which the first distance is less than a preset distance threshold.

[0111] Step S39: determining the first component recognition rate according to the third recognition result.

[0112] Step S40, determining whether the first component recognition rate is less than a preset recognition rate threshold. If so, executing step S41; otherwise, executing step S42.

[0113] Step S41: update the first cluster center according to the third recognition result.

[0114] Step S42: determining a fourth input feature set according to the third recognition result.

[0115] Step S43: determining a second distance according to the fourth input feature set and the second cluster center.

[0116] Step S44, obtaining a fourth recognition result in which the second distance is less than a preset distance threshold.

[0117] Step S45, determining the first component misrecognition rate according to the fourth recognition result.

[0118] Step S46, determining whether the first component error recognition rate is less than or equal to a preset error recognition rate threshold. If so, executing step S48; otherwise, executing step S47.

[0119] Step S47: determining to update the second cluster center according to the fourth recognition result.

[0120] Step S48: determining a second recognition system according to the second cluster center.

[0121] According to the vehicle parts identification method of the embodiment of the present invention, a first identification result is determined by the component geometric parameters of the vehicle model and a first identification system to achieve preliminary identification and classification of the components, a second identification result is determined based on the first identification result and the second identification system to achieve accurate identification and classification of the components, and the identification result of the components is output based on the second identification result to achieve efficient and accurate identification and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development cost of the vehicle.

[0122] Reference below Figure 8 An electronic device according to an embodiment of the present invention is described.

[0123] The electronic device 100 of the embodiment of the present invention includes: at least one processor 101; and a memory 102 communicatively connected to the at least one processor 101; wherein the memory 102 stores a vehicle parts identification program that can be executed by the at least one processor 101, and when the vehicle parts identification program is executed by the at least one processor 101, the at least one processor 101 executes the vehicle parts identification method as described in the above embodiment.

[0124] According to the electronic device 100 of the embodiment of the present invention, a first recognition result is determined by the geometric parameters of the components of the vehicle model and the first recognition system to achieve preliminary recognition and classification of the components, a second recognition result is determined based on the first recognition result and the second recognition system to achieve accurate recognition and classification of the components, and the recognition result of the components is output based on the second recognition result to achieve efficient and accurate recognition and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0125] In some embodiments, the processor 101 determines a first recognition result based on component geometric parameters of the vehicle model and a first recognition system, including: determining a first input feature set based on the first component geometric parameters; determining the first recognition result based on the first input feature set and the first recognition system.

[0126] In some embodiments, the processor 101 determines the second recognition result based on the first recognition result and the second recognition system, including: obtaining the number of first components of the first recognition type in the first recognition result and the number of second components of the second recognition type in the first recognition result; determining the first component recognition rate based on the first component number and the second component number; when the first component recognition rate is greater than or equal to a preset recognition rate threshold, determining the second recognition result based on the first recognition result and the second recognition system.

[0127] In some embodiments, the processor 101 determines a second recognition result based on the first recognition result and the second recognition system, including: obtaining second component geometric parameters of the second recognition type in the first recognition result and third component geometric parameters of the third recognition type in the first recognition result; determining a second input feature set based on the second component geometric parameters and the third component geometric parameters; determining the second recognition result based on the second input feature set and the second recognition system.

[0128] In some embodiments, after the processor 101 determines the component recognition rate based on the first component quantity and the second component quantity, it also includes: when the first component recognition rate is less than a preset recognition rate threshold, updating the first recognition system according to the component geometric parameters of the first recognition type in the first recognition result.

[0129] In some embodiments, the processor 101 outputs the recognition result of the component based on the second recognition result, including: obtaining the update number of the first component of the first recognition type in the second recognition result and the update number of the third component of the third recognition type in the second recognition result; determining the first component error rate based on the first component update number and the third component update number; when the first component error rate is less than or equal to a preset error rate threshold, outputting the recognition result of the component.

[0130] In some embodiments, after the processor 101 determines the component error recognition rate based on the second component update quantity and the third component update quantity, it also includes: when the first component error recognition rate is greater than a preset error recognition rate threshold, updating the second recognition system according to the component geometric parameters of the second recognition type in the second recognition result.

[0131] In some embodiments, before the processor 101 determines the first recognition result based on the geometric parameters of the components of the vehicle model and the first recognition system, it also includes: obtaining the geometric parameters of the components to be identified; determining the first cluster center based on the geometric parameters of the components to be identified; and determining the first recognition system based on the first cluster center and the vehicle model.

[0132] In some embodiments, the processor 101 determines a first recognition system based on the first cluster center and the vehicle model, including: obtaining discrete geometric parameters of the vehicle model; determining a third input feature set based on the discrete geometric parameters; and determining the first recognition system based on the third input feature set and the first cluster center.

[0133] In some embodiments, after the processor 101 determines the first recognition system based on the third input feature set and the first cluster center, it includes: determining the first distance based on the third input feature set and the first cluster center; obtaining a third recognition result in which the first distance is less than a preset distance threshold; determining a first component recognition rate based on the third recognition result; when the first component recognition rate is less than a preset recognition rate threshold, updating the first cluster center based on the third recognition result.

[0134] In some embodiments, after the processor 101 determines the first cluster center according to the geometric parameters of the parts to be identified, the process further includes: acquiring additional geometric parameters of the parts to be identified; and determining the second cluster center according to the additional geometric parameters.

[0135] In some embodiments, after the processor 101 determines the recognition rate of the first component according to the third recognition result, it also includes: when the recognition rate of the first component is greater than or equal to a preset recognition rate threshold, determining a fourth input feature set according to the third recognition result; determining a second distance according to the fourth input feature set and the second cluster center; obtaining a fourth recognition result in which the second distance is less than the preset distance threshold; determining the first component error recognition rate according to the fourth recognition result; when the first component error recognition rate is less than or equal to the preset error recognition rate threshold, determining a second recognition system according to the second cluster center.

[0136] In some embodiments, after the processor 101 determines the first component error recognition rate according to the fourth recognition result, the further step includes: when the first component error recognition rate is greater than a preset error recognition rate threshold, updating the second cluster center according to the fourth recognition result.

[0137] According to the electronic device 100 of the embodiment of the present invention, a first recognition result is determined by the geometric parameters of the components of the vehicle model and the first recognition system to achieve preliminary recognition and classification of the components, a second recognition result is determined based on the first recognition result and the second recognition system to achieve accurate recognition and classification of the components, and the recognition result of the components is output based on the second recognition result to achieve efficient and accurate recognition and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0138] The following describes a computer-readable storage medium according to an embodiment of the present invention.

[0139] The computer-readable storage medium of an embodiment of the present invention stores a vehicle parts identification program. When the vehicle parts identification program is executed by a processor, a device installed with the vehicle parts identification program implements a vehicle parts identification method as described in the above embodiment.

[0140] The following describes a computer program product according to an embodiment of the present invention.

[0141] The computer program product of the embodiment of the present invention includes a computer program. When the computer program is executed by a processor, the vehicle parts identification method as described in the above embodiment is implemented.

[0142] According to the computer program product of the embodiment of the present invention, a first recognition result is determined by the geometric parameters of the components of the vehicle model and a first recognition system to achieve preliminary recognition and classification of the components, a second recognition result is determined based on the first recognition result and the second recognition system to achieve accurate recognition and classification of the components, and the recognition result of the components is output based on the second recognition result to achieve efficient and accurate recognition and classification of the components in the informal version model of the vehicle during the research and development stage of the vehicle, thereby improving the research and development efficiency of the vehicle and saving the research and development costs of the vehicle.

[0143] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example.

[0144] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A vehicle parts identification method, characterized in that: include: Determining a first recognition result according to geometric parameters of components of the vehicle model and a first recognition system; Determining a second recognition result based on the first recognition result and the second recognition system; The recognition result of the component is output according to the second recognition result.

2. The vehicle parts identification method according to claim 1, characterized in that: Determining a first recognition result according to the component geometric parameters of the vehicle model and the first recognition system includes: Determine a first input feature set based on the component geometric parameters; The first recognition result is determined according to the first input feature set and the first recognition system.

3. The vehicle parts identification method according to claim 1, characterized in that: Determining a second recognition result according to the first recognition result and the second recognition system includes: Obtaining the quantity of first components of a first identification type in the first identification result and the quantity of second components of a second identification type in the first identification result; Determining a first component recognition rate according to the first component quantity and the second component quantity; When the first component recognition rate is greater than or equal to a preset recognition rate threshold, the second recognition result is determined according to the first recognition result and the second recognition system.

4. The vehicle parts identification method according to claim 3, characterized in that: Determining the second recognition result according to the first recognition result and the second recognition system includes: Acquire geometric parameters of a second component of a second recognition type in the first recognition result and geometric parameters of a third component of a third recognition type in the first recognition result; Determine a second input feature set based on the second component geometric parameter and the third component geometric parameter; The second recognition result is determined according to the second input feature set and the second recognition system.

5. The vehicle parts identification method according to claim 3, characterized in that: After determining the component recognition rate according to the first component quantity and the second component quantity, the method further includes: When the first component recognition rate is less than a preset recognition rate threshold, the first recognition system is updated according to the geometric parameters of the component of the first recognition type in the first recognition result.

6. The vehicle parts identification method according to claim 1, characterized in that: Outputting the recognition result of the component according to the second recognition result includes: Obtaining an updated number of first components of a first identification type in the second identification result and an updated number of third components of a third identification type in the second identification result; Determine a first component misrecognition rate according to the first component update quantity and the third component update quantity; When the first component error recognition rate is less than or equal to a preset error recognition rate threshold, the recognition result of the component is output.

7. The vehicle parts identification method according to claim 6, characterized in that: After determining the component misrecognition rate according to the second component update quantity and the third component update quantity, the method further includes: When the first component error recognition rate is greater than a preset error recognition rate threshold, the second recognition system is updated according to the component geometric parameters of the second recognition type in the second recognition result.

8. The vehicle parts identification method according to claim 1, characterized in that: Before determining the first recognition result according to the component geometric parameters of the vehicle model and the first recognition system, the method further includes: Obtain the geometric parameters of the parts to be identified; Determining a first cluster center according to the geometric parameters of the parts to be identified; A first identification system is determined according to the first cluster center and the vehicle model.

9. The vehicle parts identification method according to claim 8, characterized in that: Determining a first identification system according to the first cluster center and the vehicle model includes: Obtaining discrete geometric parameters of the vehicle model; determining a third input feature set based on the discrete geometric parameters; The first recognition system is determined according to the third input feature set and the first cluster center.

10. The vehicle parts identification method according to claim 9, characterized in that: After determining the first recognition system according to the third input feature set and the first cluster center, the method includes: Determining a first distance according to the third input feature set and the first cluster center; Obtaining a third recognition result in which the first distance is less than a preset distance threshold; Determining a first component recognition rate according to the third recognition result; When the first component recognition rate is less than a preset recognition rate threshold, the first cluster center is updated according to the third recognition result.

11. The vehicle parts identification method according to claim 8, characterized in that: After determining the first cluster center according to the geometric parameters of the parts to be identified, the method further includes: Obtain additional geometric parameters of the parts to be identified; A second cluster center is determined based on the additional geometric parameter.

12. The vehicle parts identification method according to claim 10, characterized in that: After determining the first component recognition rate according to the third recognition result, the method further includes: When the first component recognition rate is greater than or equal to a preset recognition rate threshold, determining a fourth input feature set according to the third recognition result; Determining a second distance based on the fourth input feature set and the second cluster center; Acquire a fourth recognition result in which the second distance is less than a preset distance threshold; Determining a first component misrecognition rate according to the fourth recognition result; When the first component error recognition rate is less than or equal to a preset error recognition rate threshold, a second recognition system is determined according to the second cluster center.

13. The vehicle parts identification method according to claim 12, characterized in that: After determining the first component misrecognition rate according to the fourth recognition result, the method further includes: When the first component error recognition rate is greater than a preset error recognition rate threshold, the second cluster center is updated according to the fourth recognition result.

14. An electronic device, characterized in that: include: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a vehicle parts identification program that can be executed by the at least one processor. When the vehicle parts identification program is executed by the at least one processor, the at least one processor executes the vehicle parts identification method as described in any one of claims 1-13.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a vehicle parts identification program, and when the vehicle parts identification program is executed by a processor, a device installed with the vehicle parts identification program implements the vehicle parts identification method as described in any one of claims 1-13.

16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the vehicle parts identification method according to any one of claims 1 to 13 is implemented.

Images