Automobile part identification method, device and equipment and storage medium

Abstract

The invention belongs to the field of artificial intelligence, and relates to an automobile part identification method, which comprises the following steps: acquiring image data acquired by a mobile terminal camera, and acquiring multiple frames of to-be-identified images from the image data according to a preset time interval; performing component detection on each frame of the to-be-recognized image by adopting a pre-trained automobile component recognition model to obtain a component detection result corresponding to each frame of the to-be-recognized image; the component detection result corresponding to each frame of the to-be-recognized image is sequentially stored in a cache queue according to the timestamp of the to-be-recognized image, and the cache queue only reserves the component detection result of a preset duration; and outputting a component identification result of the image data according to the component detection result in the cache queue and a judgment result of a preset rule. In addition, the invention also relates to a block chain technology, and the recognition result of the automobile part and the like can be stored in a block chain. According to the invention, the accuracy of automobile part identification can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a method for identifying automobile parts. Background Art

[0002] With the rapid development of artificial intelligence, the application of artificial intelligence in various industries is becoming more and more extensive, and the application of artificial intelligence has also been added to the auto insurance claims industry. In the auto insurance claims process, photographing and identifying damaged car parts is an indispensable part. In the traditional way, the driver photographs the damaged car parts and uploads them to the claims platform or sends them to the claims personnel, who then manually identify the damaged car parts. This process is not only time-consuming and labor-intensive, but also has low efficiency and accuracy.

[0003] With the introduction of artificial intelligence technology, when the driver takes photos of damaged car parts, artificial intelligence can be used to automatically identify the damaged car parts, reducing the workload of claims adjusters and improving the efficiency of identifying damaged car parts. However, in the process of automatic identification of damaged car parts, since only a single photo is used for identification, and the light and clarity of the photos cause uneven picture quality, automatic identification may result in misidentification. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, equipment and storage medium for identifying automobile parts, the main purpose of which is to improve the accuracy of identifying automobile parts.

[0005] In order to solve the above technical problems, the present application provides an automobile parts identification method, which adopts the following technical solution:

[0006] Acquire image data collected by a mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals;

[0007] Using a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized, and obtaining a component detection result corresponding to each frame of the image to be recognized;

[0008] The component detection results corresponding to each frame of the image to be identified are sequentially stored in a cache queue according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length;

[0009] The component recognition result of the image data is output according to the component detection result in the cache queue and the judgment result of the preset rule.

[0010] Furthermore, collecting multiple frames of to-be-recognized images from the image data at preset time intervals includes:

[0011] According to the preset time interval, a preset timer is used to capture multiple frames of initial images from the image data;

[0012] Performing clarity verification on multiple frames of the initial image according to a preset clarity threshold;

[0013] An image satisfying the clarity check is acquired to obtain multiple frames of the image to be recognized.

[0014] Furthermore, performing clarity check on multiple frames of the initial image according to a preset clarity threshold includes:

[0015] For each of the multiple frames of the initial images, a convolution operation is performed on the initial image to extract a feature vector of the initial image to obtain an image feature vector;

[0016] The image feature vector is transformed by a single neuron to obtain the image clarity score;

[0017] When the image clarity score is greater than or equal to the preset clarity threshold, retaining the initial image corresponding to the image clarity score;

[0018] When the image clarity score is less than the preset clarity threshold, the initial image corresponding to the image clarity score is deleted.

[0019] Furthermore, the pre-trained automobile component recognition model is used to perform component detection on each frame of the image to be recognized, and the component detection result corresponding to each frame of the image to be recognized is obtained, including:

[0020] Using a pre-trained automobile part recognition model to perform part detection on each frame of the image to be recognized, and obtain the name and quantity of automobile parts corresponding to each frame of the image to be recognized;

[0021] Determine the region to which the image to be identified belongs according to the component region mapping table and the automobile component name;

[0022] The automobile component names, the automobile component data and the regions are collected to obtain component detection results for each frame of the image to be identified.

[0023] Furthermore, the component detection results corresponding to each frame of the image to be identified are sequentially stored in the cache queue according to the timestamp of the image to be identified, including:

[0024] According to the sequence of the timestamps of the images to be identified, the component detection results corresponding to each frame of the images to be identified are stored in the cache queue;

[0025] Determining, based on the timestamp of the image to be identified, whether the component detection results in the cache queue exceed a preset time length;

[0026] When the component detection result in the cache queue exceeds the preset time length, the component detection result first stored in the cache queue is deleted until the component detection result in the cache queue does not exceed the preset time length.

[0027] Furthermore, outputting the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule includes:

[0028] Randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result;

[0029] If the number of automobile components in the current component detection result is one, obtaining component detection results adjacent to the current component detection result;

[0030] Determine whether the adjacent component detection result includes the automobile component name of the current component detection result;

[0031] If not included, the component recognition result outputted from the image data is a misrecognition.

[0032] Furthermore, the randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result, includes:

[0033] If there are multiple automobile parts in the current part detection result, obtaining part detection results adjacent to the current part detection result;

[0034] Determine whether the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, wherein the front region and the rear region are mutually exclusive, and the left region and the right region are mutually exclusive;

[0035] If the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, then it is determined that there is misidentification in the component detection result of the image data;

[0036] If the regions to which the current component detection result and the adjacent component detection result belong are not mutually exclusive, it is determined that there is no misidentification in the component detection result of the image data.

[0037] In order to solve the above technical problems, the embodiment of the present application further provides an identification device for automobile parts, which adopts the following technical solution:

[0038] A framing module is used to obtain image data collected by a mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals;

[0039] A detection module is used to perform component detection on each frame of the image to be identified using a pre-trained automobile component recognition model to obtain a component detection result corresponding to each frame of the image to be identified;

[0040] A cache module, used for storing the component detection results corresponding to each frame of the image to be identified into a cache queue in sequence according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length;

[0041] The output module is used to output the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule.

[0042] In order to solve the above technical problems, an embodiment of the present application also provides a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the automobile parts identification method as described above.

[0043] In order to solve the above technical problems, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned method for identifying automobile parts when executed by a processor.

[0044] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0045] Acquire image data collected by a mobile terminal camera, and collect multiple frames of images to be identified from the image data at preset time intervals. By collecting multiple frames of images to be identified from the image data, misidentification caused by a single image is avoided, the accuracy of subsequent automobile parts identification is improved, and the probability of rework by claims personnel is reduced, thereby improving the efficiency of automobile parts identification;

[0046] By identifying the detection results corresponding to each frame of the images to be identified in multiple frames and combining the detection results of multiple images to be identified, it can avoid misjudgment caused by a single photo, thereby improving the accuracy of subsequent automobile parts identification;

[0047] Only the component detection results of the preset time length are retained to ensure the timeliness of cached data and avoid interference from outdated data; by setting the preset time length of the cache queue, memory overflow is prevented and the stability of the system is improved;

[0048] By comparing the component detection results with preset rules, the components in the image can be identified more accurately; by analyzing the component detection results of multiple frames of images, misidentification of a single image can be avoided, thereby improving the accuracy of image data detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0051] Figure 2 A flowchart of an embodiment of a method for identifying automobile parts according to the present application;

[0052] Figure 3 is a schematic structural diagram of an embodiment of an automobile component identification device according to the present application;

[0053] Figure 4 It is a structural schematic diagram of an embodiment of a device according to the present application. DETAILED DESCRIPTION

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0055] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0056] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0057] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0058] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0059] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0060] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0061] It should be noted that the automobile parts identification method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the automobile parts identification device is generally set in the server / terminal device.

[0062] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0063] Continue to refer Figure 2, shows a flow chart of an embodiment of the method for identifying automobile parts according to the present application. According to different requirements, the order of the steps in the flow chart can be changed, and some steps can be omitted. The method for identifying automobile parts provided in the embodiment of the present application can be applied to any scenario where the identification of automobile parts is required, and the method for identifying automobile parts can be applied to products in these scenarios. The method for identifying automobile parts includes the following steps:

[0064] Step S201: Acquire image data collected by a mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals.

[0065] In this embodiment, the method for identifying automobile parts is executed on an electronic device (eg Figure 1 The server / terminal device shown in the figure) can obtain the image data collected by the mobile camera through a wired connection or a wireless connection. It should be pointed out that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future. The executor of the present application may be an automobile part recognition model, or simply referred to as a model. The above-mentioned image data is usually of the same size as the image training data used in the training of the automobile part recognition model, but the training data contains other noise, depending on the specific situation. Among them, in the business scenario of identifying automobile parts, the image data may include automobile image videos taken by the camera of the mobile terminal.

[0066] In this embodiment, the above-mentioned mobile terminal includes but is not limited to handheld terminals, such as mobile phones, iPads, etc., and the above-mentioned image data refers to video data captured by the camera of the mobile terminal; the image data collected by the user through the mobile terminal camera is obtained, and the image data is framed according to a preset time interval to obtain multiple frames of images to be identified.

[0067] In this embodiment, after obtaining multiple frames of images to be recognized, it is necessary to perform preprocessing operations on each frame of the image. The above preprocessing operations include image denoising, grayscale processing and normalization processing on each frame of the image to be recognized, so as to obtain multiple frames of preprocessed images to be recognized; wherein, the image denoising method is used to reduce the noise to be recognized and improve the image quality; the grayscale processing method is used to process the image to be recognized, which can reduce the calculation amount of subsequent processing and improve the efficiency and accuracy of subsequent recognition steps; the normalization processing method can convert the image to be recognized into a numerical range that can be recognized by the pre-trained automobile part recognition model, so as to avoid the problem of subsequent unrecognition.

[0068] Specifically, the image to be identified is denoised by using Gaussian filtering, each pixel in the image to be identified is scanned, and the value of the central pixel of the template is replaced by the weighted average gray value of the pixels in the area determined by the template to obtain the denoised image to be identified; each pixel of the color image to be identified is converted from the RGB color space to the grayscale space to obtain the image to be identified after grayscale processing; the image to be identified is normalized by using image normalization to complete the preprocessing steps.

[0069] In one embodiment, collecting multiple frames of to-be-recognized images from the image data at preset time intervals includes:

[0070] According to the preset time interval, a preset timer is used to capture multiple frames of initial images from the image data;

[0071] Performing clarity verification on multiple frames of the initial image according to a preset clarity threshold;

[0072] An image satisfying the clarity check is acquired to obtain multiple frames of the image to be recognized.

[0073] In this embodiment, a preset time interval is set in advance, for example, 30ms, and the image data is positioned using a preset timer according to the preset time interval. Multiple frames of initial images are captured from the image data according to the positioning, and the multiple frames of initial images are input into a pre-trained clarity verification model to obtain a clarity score corresponding to each frame of the initial image, and the clarity score corresponding to each frame of the initial image is compared with a preset clarity threshold. The initial images with a clarity score greater than or equal to the preset clarity threshold are retained, and the initial images with a clarity score less than the preset clarity threshold are deleted, and finally multiple frames of images to be identified are obtained.

[0074] In this embodiment, image data is collected at preset time intervals to ensure the stability of the image data and reduce the waste of computing resources caused by directly processing the image data. The extracted multiple frames of images are screened by pre-setting the clarity threshold, which reduces invalid data processing and improves the accuracy of subsequent image recognition.

[0075] In another embodiment, the capturing of multiple frames of initial images from the image data using a preset timer according to the preset time interval includes:

[0076] When an image capture action is detected in the image data;

[0077] A timestamp corresponding to the image capture action is generated according to the image capture action and the current time, and a plurality of frames of the initial image are obtained, wherein each frame of the initial image has a corresponding timestamp.

[0078] In this embodiment, when an image capture action is detected, a timestamp is automatically generated according to the current time when the image capture action occurs, and an initial image with a timestamp is obtained. All initial images with timestamps are collected to obtain multiple frames of the initial image.

[0079] In another embodiment, performing clarity check on multiple frames of the initial image according to a preset clarity threshold comprises:

[0080] For each of the multiple frames of the initial images, a convolution operation is performed on the initial image to extract a feature vector of the initial image to obtain an image feature vector;

[0081] The image feature vector is transformed by a single neuron to obtain the image clarity score;

[0082] When the image clarity score is greater than or equal to the preset clarity threshold, retaining the initial image corresponding to the image clarity score;

[0083] When the image clarity score is less than the preset clarity threshold, the initial image corresponding to the image clarity score is deleted.

[0084] In this embodiment, a pre-trained clarity model is used to perform clarity check on multiple frames of initial images, and the clarity model refers to a convolutional neural network; each frame of the initial image is output to the clarity model, and a convolution operation and a dimensionality reduction operation are performed on the initial image to extract an image feature vector of the initial image, and the image feature vector is transformed by a single neuron of the fully connected layer to finally obtain an image clarity score. Specifically, the fully connected layer first performs a linear transformation on the image feature vector to obtain a linear transformation vector, and the linear transformation is achieved by multiplying the feature vector by a weight matrix and adding a bias vector. The formula of the linear transformation is: y=Wx+b;

[0085] Among them, y is the linear transformation vector, W is the weight matrix, x is the input feature vector, and b is the bias vector.

[0086] The linear vector is processed using an activation function to finally obtain a clarity score. The activation function includes but is not limited to a ReLU activation function, a Sigmoid activation function, and the like.

[0087] In this embodiment, the clarity of multiple frames of initial images may be identified by using a Tenengrad gradient method, a Laplacian gradient method, a variance method, and the like.

[0088] In this embodiment, by performing clarity recognition on images, high-quality images are screened out to avoid the situation where subsequent images cannot be recognized, thereby reducing the waste of computing resources.

[0089] In an implementable example A, a user uses a mobile phone camera to shoot a video of a damaged car to obtain a video of the damaged car. The video of the damaged car shot by the user is obtained, and the image of the damaged car video is located using a timer according to a time interval of 30 ms. The damaged car video is framed according to the above positioning to obtain multiple frames of initial damaged car images, and clarity verification is performed on each frame of the initial damaged car image to screen out initial damaged car images with qualified clarity to obtain multiple frames of damaged car images.

[0090] In this embodiment, image data captured by a mobile camera is obtained, and multiple frames of images to be identified are collected from the image data at preset time intervals. By collecting multiple frames of image data to be identified, misidentification caused by a single image is avoided, the accuracy of subsequent automobile parts identification is improved, and the probability of rework by claims personnel is reduced, thereby improving the efficiency of automobile parts identification.

[0091] Step S202: Use a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized, and obtain a component detection result corresponding to each frame of the image to be recognized.

[0092] In this embodiment, the pre-trained automobile component recognition model refers to a convolutional neural network (CNN), and the component detection result includes the recognized automobile component name and the recognized automobile component quantity.

[0093] In this embodiment, the multiple frames of images to be identified acquired in step S201 are input into a pre-trained automobile parts recognition model, and the pre-trained automobile parts recognition model is used to perform parts detection on each frame of the image to be identified to obtain the area to which the automobile parts, the name of the automobile parts, and the number of automobile parts corresponding to each image are obtained.

[0094] Pre-set component area mapping table:

[0095] Front area: front bumper, hood, center grille, front windshield.

[0096] Rear area: rear bumper, trunk lid, rear windshield, and rear lip.

[0097] Left area: left front door, left rear door, left front door glass, left rear door glass, left front fender, left rear fender, left triangular glass, left wheel rim, left tire.

[0098] Right area: right front door, right rear door, right front door glass, right rear door glass, right front fender, right rear fender, right triangular glass, right wheel rim, right tire.

[0099] In one embodiment, the use of a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized to obtain a component detection result corresponding to each frame of the image to be recognized includes:

[0100] Using a pre-trained automobile part recognition model to perform part detection on each frame of the image to be recognized, and obtain the name and quantity of automobile parts corresponding to each frame of the image to be recognized;

[0101] Determine the region to which the image to be identified belongs according to the component region mapping table and the automobile component name;

[0102] The automobile component names, the automobile component data and the regions are collected to obtain component detection results for each frame of the image to be identified.

[0103] In this embodiment, a pre-trained automobile part recognition model is used to perform part detection on each frame of the image to be recognized. By alternately stacking multiple convolutional layers and pooling layers, the automobile part recognition model extracts features from the input image layer by layer to obtain a part feature vector. The part feature vector is converted into a probability distribution using a preset activation function, and the one with the largest probability distribution is selected as the output result to obtain the name and number of automobile parts of the image to be recognized. According to the above-mentioned part area mapping table and the detected automobile part names, the area corresponding to the image to be recognized is determined, and the automobile part names, the number of automobile parts and the area are collected to obtain the part detection result corresponding to each frame of the image to be recognized.

[0104] In this embodiment, each frame of the image to be identified is detected by a pre-trained component recognition model, which avoids manual recognition and improves the accuracy and efficiency of component recognition; the area to which the car belongs is located based on the component detection results, which can provide a basis for subsequent rule judgments, thereby improving the accuracy of misidentification judgments.

[0105] In another embodiment, before using the pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized to obtain the component detection result corresponding to each frame of the image to be recognized, the method further includes:

[0106] Obtaining a labeled image dataset, and dividing the labeled image dataset into a training set and a test set according to a preset ratio;

[0107] Constructing an initial component recognition model, and training the initial component recognition model using the training set to obtain an intermediate component recognition model;

[0108] Testing the intermediate component recognition model using a test set to obtain a loss value;

[0109] When the loss value is less than or equal to a preset loss threshold, obtaining the pre-trained component recognition model;

[0110] When the loss value is greater than the preset loss threshold, the initial component recognition model is continuously trained using the training set until the loss value is less than or equal to the preset loss threshold, and the training is completed to obtain the pre-trained component recognition model.

[0111] In this embodiment, a large number of annotated image data sets are obtained, for example, the component names, the number of components and the regions to which they belong are annotated. The large number of annotated image data sets are split into a training set and a test set according to a preset ratio; the initial component recognition model is trained with the training set to obtain an intermediate component recognition model; the intermediate component recognition model is tested with the test set to obtain a loss value, and it is determined whether the loss value is less than or equal to a preset loss threshold. When the loss value is less than or equal to the preset loss threshold, the training is completed to obtain a pre-trained component recognition model. When the loss value is greater than the preset loss threshold, the initial component recognition model is continuously trained with the training set until the loss value is less than or equal to the preset loss threshold, and the training is terminated to obtain a pre-trained component recognition model.

[0112] Continuing with the above-mentioned feasible example A, taking an image of a damaged car as an example, the pre-trained car recognition model recognizes that the image to be recognized contains two car parts, namely the front bumper and the middle grille. According to the front bumper and the middle grille, the area corresponding to the image to be recognized is determined to be the front area. All the detection results are collected to obtain the detection result of the image to be recognized:

[0113] Part name: front bumper, center grille;

[0114] Number of parts: 2;

[0115] Area: front area.

[0116] In this embodiment, by identifying the detection result corresponding to each frame of the image to be identified in multiple frames of images to be identified and combining the detection results of multiple images to be identified, misjudgment caused by a single photo can be avoided, thereby improving the accuracy of subsequent automobile parts identification.

[0117] Step S203: storing the component detection results corresponding to each frame of the image to be identified in a cache queue in sequence according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length.

[0118] In this embodiment, it is mentioned in step S201 that each image to be identified will generate a corresponding timestamp when it is captured. In step S203, the image to be identified and the corresponding component detection results are stored in the cache queue in sequence according to the timestamp of the image to be identified, wherein the cache queue can only retain the component detection results of a preset time length, for example, the component detection results with a timestamp within 3 seconds.

[0119] In one embodiment, storing the component detection results corresponding to each frame of the image to be identified in the cache queue in sequence according to the timestamp of the image to be identified includes:

[0120] According to the sequence of the timestamps of the images to be identified, the component detection results corresponding to each frame of the images to be identified are stored in the cache queue;

[0121] Determining, based on the timestamp of the image to be identified, whether the component detection results in the cache queue exceed a preset time length;

[0122] When the component detection result in the cache queue exceeds the preset time length, the component detection result first stored in the cache queue is deleted until the component detection result in the cache queue does not exceed the preset time length.

[0123] In this embodiment, the component detection results of each frame of image are stored in the cache queue in sequence according to the order of timestamps generated when the image is captured, and the cache queue is dynamically updated to prevent the cached data from being too large and causing memory overflow.

[0124] Specifically, the method of dynamically updating the cache queue includes: determining whether the component detection result in the cache queue exceeds the preset time length, such as the preset time length is 3s, and each component detection result is 3ms, then there are no more than 1000 component detection results in the cache queue; when the component detection result in the cache queue exceeds the preset time length, deleting the first stored component detection result until the component detection result in the cache queue does not exceed the preset time length. For example, when the component detection result in the cache queue exceeds the preset time length, the first stored component detection result is deleted, and the 1001st component detection result is stored in the cache queue. If the component detection result in the cache queue still exceeds the preset time length, the second stored component detection result is deleted, and the 1002nd component detection result is stored in the cache queue until the component detection result in the cache queue does not exceed the preset time length.

[0125] In this embodiment, by setting the preset duration and dynamic update strategy of the cache queue, memory overflow can be prevented, the earliest stored data can be dynamically deleted, and the queue can be ensured to always operate within a manageable range, thereby improving the stability of the system and ensuring the freshness of the component detection results.

[0126] Continuing with the above-mentioned feasible example A, after obtaining the component detection results of all the images to be identified, the component detection results are stored in the cache queue according to the timestamp of the images to be identified, and a dynamic update strategy is implemented for the cache images to ensure that only data within 3s is retained in the cache queue. In other words, the cache queue can only store 1000 component detection results. When the cache queue exceeds 1000 component detection results, the first one is deleted and re-judged. If it still exceeds, the second one is deleted and judged again until the component detection results in the cache queue do not exceed 3s.

[0127] In this embodiment, only component detection results of a preset time length are retained to ensure the timeliness of cached data and avoid interference from outdated data; by setting a preset time length for the cache queue, memory overflow is prevented and system stability is improved.

[0128] Step S204: output the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule.

[0129] In this embodiment, a preset rule is set, and the preset rule is used to determine whether there is misidentification in component detection. Specifically, the preset rule includes: determining whether the image detection result is a real vehicle, if not, outputting a misidentification judgment result, if so, determining whether there are mutually exclusive results in the area of ​​the component detection result, if so, outputting a misidentification judgment result, and if not, outputting a judgment result that there is no misidentification.

[0130] When the component detection results in the cache queue meet the preset rules, it is determined that there is no misidentification in the component detection results. If the component detection results in the cache queue do not meet the preset rules, it is determined that there is misidentification in the component detection results.

[0131] In one embodiment, outputting the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule includes:

[0132] Randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result;

[0133] If the number of automobile components in the current component detection result is one, obtaining component detection results adjacent to the current component detection result;

[0134] Determine whether the adjacent component detection result includes the automobile component name of the current component detection result;

[0135] If not included, the component recognition result outputted from the image data is a misrecognition.

[0136] In this embodiment, a component detection result is randomly selected from the cache queue as the current component detection result for analysis to determine whether the number of automobile components in the current component detection result is 1. If the number of automobile components in the current component detection result is not 1, the next step of judgment is performed, and the next step of judgment refers to determining whether there is a mutual exclusion situation among multiple numbers of automobile components, which is specifically reflected below and will not be elaborated here. If the number of automobile components in the current component detection result is 1, a preset number of component detection results adjacent to the current component detection result are obtained, and the names of automobile components of a preset number of adjacent component detection results are obtained respectively. When the name of an automobile component of any adjacent component detection result in the preset number of adjacent component detection results does not contain the name of an automobile component in the current component detection result, it is determined that there is misrecognition in the component recognition result of the image data.

[0137] In another embodiment, randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result includes:

[0138] If there are multiple automobile parts in the current part detection result, obtaining part detection results adjacent to the current part detection result;

[0139] Determine whether the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, wherein the front region and the rear region are mutually exclusive, and the left region and the right region are mutually exclusive;

[0140] If the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, then it is determined that there is misidentification in the component detection result of the image data;

[0141] If the regions to which the current component detection result and the adjacent component detection result belong are not mutually exclusive, it is determined that there is no misidentification in the component detection result of the image data.

[0142] In this embodiment, when there are multiple automobile parts in the current part detection result, a part detection result adjacent to the current part detection result is obtained, and it is determined whether there is a mutual exclusion between the regions to which the current part detection result and the adjacent part detection result belong, wherein the front region and the rear region are mutually exclusive, and the left region and the right region are mutually exclusive. If there is a mutual exclusion between the regions to which the current part detection result and the adjacent part detection result belong, it is determined that there is misidentification in the part detection result of the image data. If there is no mutual exclusion between the regions to which the current part detection result and the adjacent part detection result belong, it is determined that there is no misidentification in the part detection result of the image data.

[0143] In this embodiment, by randomly selecting component detection results in the cache queue for analysis one by one, each component in the image data can be checked more comprehensively, reducing errors caused by omissions or misrecognition; when a single automobile component is detected, its correctness is verified by checking adjacent component detection results, which helps to avoid misrecognition; when multiple automobile components are detected, the accuracy of the recognition can be further verified by checking the spatial relationship between these components (such as whether the areas they belong to are mutually exclusive). Since in actual situations, the combination or positional relationship of some components may be unreasonable, the accuracy of image data recognition is improved.

[0144] Continuing with the above-mentioned implementable example A, a component detection result is randomly selected from 1000 component detection results as the current component detection result, and the information of the current component detection result is obtained: component name: front bumper, center grille; number of components: 2; region: front region; the current component detection result is judged for its uniqueness, and if it is judged that the current component detection result passes the uniqueness judgment, the next step of judgment is performed to obtain the adjacent component detection result 1 adjacent to the current component detection result: component name: front bumper, center grille; number of components: 2; region: front region; adjacent component detection result 2: component name: rear bumper; number of components: 1; region: rear region; the region: front region to which the current component detection result belongs is judged with the region: front region to which the adjacent component detection result 1 belongs, and the region: rear region to which the adjacent component detection result 2 belongs, and it is judged that the current component detection result and the regions of the adjacent component detection result 2 are mutually exclusive, and the output image data has a result of misrecognition.

[0145] In this embodiment, by comparing the component detection results with preset rules, each component in the image can be identified more accurately; by analyzing the component detection results of multiple frames of images, misidentification of a single image can be avoided, thereby improving the accuracy of image data detection.

[0146] It should be emphasized that in order to further ensure the privacy and security of the identification results of the above-mentioned automobile parts, etc., the identification results of the above-mentioned automobile parts, etc. can also be stored in a node of a blockchain.

[0147] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0148] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0149] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0150] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0151] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0152] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a device for identifying automobile parts, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the apparatus can be specifically applied to various computer devices.

[0153] like Figure 3As shown, the automobile parts recognition device 300 described in this embodiment includes: a frame segmentation module 301, a detection module 302, a buffer module 303, and an output module 304. Among them:

[0154] The frame segmentation module 301 is used to obtain image data collected by the mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals;

[0155] In one embodiment, collecting multiple frames of to-be-recognized images from the image data at preset time intervals includes:

[0156] According to the preset time interval, a preset timer is used to capture multiple frames of initial images from the image data;

[0157] Performing clarity verification on multiple frames of the initial image according to a preset clarity threshold;

[0158] An image satisfying the clarity check is acquired to obtain multiple frames of the image to be recognized.

[0159] In one embodiment, the clarity check of the multiple frames of the initial images according to the preset clarity threshold comprises:

[0160] For each of the multiple frames of the initial images, a convolution operation is performed on the initial image to extract a feature vector of the initial image to obtain an image feature vector;

[0161] The image feature vector is transformed by a single neuron to obtain the image clarity score;

[0162] When the image clarity score is greater than or equal to the preset clarity threshold, retaining the initial image corresponding to the image clarity score;

[0163] When the image clarity score is less than the preset clarity threshold, the initial image corresponding to the image clarity score is deleted.

[0164] The detection module 302 is used to perform component detection on each frame of the image to be identified using a pre-trained automobile component recognition model to obtain a component detection result corresponding to each frame of the image to be identified;

[0165] In one embodiment, the use of a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized to obtain a component detection result corresponding to each frame of the image to be recognized includes:

[0166] Using a pre-trained automobile part recognition model to perform part detection on each frame of the image to be recognized, and obtain the name and quantity of automobile parts corresponding to each frame of the image to be recognized;

[0167] Determine the region to which the image to be identified belongs according to the component region mapping table and the automobile component name;

[0168] The automobile component names, the automobile component data and the regions are collected to obtain component detection results for each frame of the image to be identified.

[0169] The cache module 303 is used to store the component detection results corresponding to each frame of the image to be identified into a cache queue in sequence according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length;

[0170] In one embodiment, storing the component detection results corresponding to each frame of the image to be identified in the cache queue in sequence according to the timestamp of the image to be identified includes:

[0171] According to the sequence of the timestamps of the images to be identified, the component detection results corresponding to each frame of the images to be identified are stored in the cache queue;

[0172] Determining, based on the timestamp of the image to be identified, whether the component detection results in the cache queue exceed a preset time length;

[0173] When the component detection result in the cache queue exceeds the preset time length, the component detection result first stored in the cache queue is deleted until the component detection result in the cache queue does not exceed the preset time length.

[0174] The output module 304 is used to output the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule.

[0175] In one embodiment, outputting the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule includes:

[0176] Randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result;

[0177] If the number of automobile components in the current component detection result is one, obtaining component detection results adjacent to the current component detection result;

[0178] Determine whether the adjacent component detection result includes the automobile component name of the current component detection result;

[0179] If not included, the component recognition result outputted from the image data is a misrecognition.

[0180] In one embodiment, randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result includes:

[0181] If there are multiple automobile parts in the current part detection result, obtaining part detection results adjacent to the current part detection result;

[0182] Determine whether the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, wherein the front region and the rear region are mutually exclusive, and the left region and the right region are mutually exclusive;

[0183] If the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, then it is determined that there is misidentification in the component detection result of the image data;

[0184] If the regions to which the current component detection result and the adjacent component detection result belong are not mutually exclusive, it is determined that there is no misidentification in the component detection result of the image data.

[0185] In this embodiment, image data collected by a mobile terminal camera is obtained, and multiple frames of images to be identified are collected from the image data at preset time intervals. By collecting multiple frames of images to be identified from the image data, misidentification caused by a single image is avoided, the accuracy of subsequent automobile parts identification is improved, and the probability of rework by claims personnel is reduced, thereby improving the efficiency of automobile parts identification;

[0186] By identifying the detection results corresponding to each frame of the images to be identified in multiple frames and combining the detection results of multiple images to be identified, it can avoid misjudgment caused by a single photo, thereby improving the accuracy of subsequent automobile parts identification;

[0187] Only the component detection results of the preset time length are retained to ensure the timeliness of cached data and avoid interference from outdated data; by setting the preset time length of the cache queue, memory overflow is prevented and the stability of the system is improved;

[0188] By comparing the component detection results with preset rules, the components in the image can be identified more accurately; by analyzing the component detection results of multiple frames of images, misidentification of a single image can be avoided, thereby improving the accuracy of image data detection.

[0189] In order to solve the above technical problems, the present application embodiment also provides a device (computer device). Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0190] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 4 with a memory 41, a processor 42, and a network interface 43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0191] The computer device may be a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with the user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0192] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the method for identifying automobile parts, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0193] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the method for identifying automobile parts.

[0194] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0195] During the implementation of the electronic device of the present application, by acquiring image data collected by a mobile terminal camera, multiple frames of images to be identified are collected from the image data at preset time intervals. By collecting multiple frames of images to be identified from the image data, misidentification caused by a single image is avoided, the accuracy of subsequent automobile parts identification is improved, and the probability of rework by claims personnel is reduced, thereby improving the efficiency of automobile parts identification;

[0196] By identifying the detection results corresponding to each frame of the images to be identified in multiple frames and combining the detection results of multiple images to be identified, it can avoid misjudgment caused by a single photo, thereby improving the accuracy of subsequent automobile parts identification;

[0197] Only the component detection results of the preset time length are retained to ensure the timeliness of cached data and avoid interference from outdated data; by setting the preset time length of the cache queue, memory overflow is prevented and the stability of the system is improved;

[0198] By comparing the component detection results with preset rules, the components in the image can be identified more accurately; by analyzing the component detection results of multiple frames of images, misidentification of a single image can be avoided, thereby improving the accuracy of image data detection.

[0199] The present application also provides another embodiment, namely, providing a storage medium (computer-readable storage medium), wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned automobile parts identification method.

[0200] During the implementation of the computer-readable storage medium of the present application, by acquiring image data collected by a mobile terminal camera, multiple frames of images to be identified are collected from the image data at preset time intervals. By collecting multiple frames of images to be identified from the image data, misidentification caused by a single image is avoided, the accuracy of subsequent automobile parts identification is improved, and the probability of rework by claims personnel is reduced, thereby improving the efficiency of automobile parts identification;

[0201] By identifying the detection results corresponding to each frame of the images to be identified in multiple frames and combining the detection results of multiple images to be identified, it can avoid misjudgment caused by a single photo, thereby improving the accuracy of subsequent automobile parts identification;

[0202] Only the component detection results of the preset time length are retained to ensure the timeliness of cached data and avoid interference from outdated data; by setting the preset time length of the cache queue, memory overflow is prevented and the stability of the system is improved;

[0203] By comparing the component detection results with preset rules, the components in the image can be identified more accurately; by analyzing the component detection results of multiple frames of images, misidentification of a single image can be avoided, thereby improving the accuracy of image data detection.

[0204] The non-Company software tools or components appearing in the embodiments of this application are merely examples and do not represent actual use.

[0205] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0206] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for identifying automobile parts, characterized in that: The steps include: Acquire image data collected by a mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals; Using a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized, and obtaining a component detection result corresponding to each frame of the image to be recognized; The component detection results corresponding to each frame of the image to be identified are sequentially stored in a cache queue according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length; The component recognition result of the image data is output according to the component detection result in the cache queue and the judgment result of the preset rule.

2. The method for identifying automobile parts according to claim 1, characterized in that: The step of collecting a plurality of frames of images to be identified from the image data at preset time intervals includes: According to the preset time interval, a preset timer is used to capture multiple frames of initial images from the image data; Performing clarity verification on multiple frames of the initial image according to a preset clarity threshold; An image satisfying the clarity check is acquired to obtain multiple frames of the image to be recognized.

3. The method for identifying automobile parts according to claim 2, characterized in that: The performing clarity check on the multiple frames of the initial images according to the preset clarity threshold comprises: For each of the multiple frames of the initial images, a convolution operation is performed on the initial image to extract a feature vector of the initial image to obtain an image feature vector; The image feature vector is transformed by a single neuron to obtain an image clarity score; When the image clarity score is greater than or equal to the preset clarity threshold, retaining the initial image corresponding to the image clarity score; When the image clarity score is less than the preset clarity threshold, the initial image corresponding to the image clarity score is deleted.

4. The method for identifying automobile parts according to claim 1, characterized in that: The method of using a pre-trained automobile component recognition model to perform component detection on each frame of the image to be recognized to obtain a component detection result corresponding to each frame of the image to be recognized includes: Using a pre-trained automobile part recognition model to perform part detection on each frame of the image to be recognized, and obtain the name and quantity of automobile parts corresponding to each frame of the image to be recognized; Determine the region to which the image to be identified belongs according to the component region mapping table and the automobile component name; The automobile component names, the automobile component quantities, and the regions are collected to obtain component detection results for each frame of the image to be identified.

5. The method for identifying automobile parts according to claim 1, characterized in that: The step of storing the component detection results corresponding to each frame of the image to be identified in the cache queue in sequence according to the timestamp of the image to be identified includes: According to the sequence of timestamps of the images to be identified, the component detection results corresponding to each frame of the images to be identified are stored in the cache queue; Determining, based on the timestamp of the image to be identified, whether the component detection results in the cache queue exceed a preset time length; When the component detection result in the cache queue exceeds the preset time length, the component detection result first stored in the cache queue is deleted until the component detection result in the cache queue does not exceed the preset time length.

6. The method for identifying automobile parts according to claim 1, characterized in that: The step of outputting the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule includes: Randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result; If the number of automobile components in the current component detection result is one, obtaining component detection results adjacent to the current component detection result; Determine whether the adjacent component detection result includes the automobile component name of the current component detection result; If not included, the component recognition result outputted from the image data is a misrecognition.

7. The method for identifying automobile parts according to claim 6, characterized in that: The randomly selecting a component detection result from the cache queue as the current component detection result, and performing a single component judgment on the current component detection result, includes: If there are multiple automobile parts in the current part detection result, obtaining part detection results adjacent to the current part detection result; Determine whether the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, wherein the front region and the rear region are mutually exclusive, and the left region and the right region are mutually exclusive; If the regions to which the current component detection result and the adjacent component detection result belong are mutually exclusive, then it is determined that there is misidentification in the component detection result of the image data; If the regions to which the current component detection result and the adjacent component detection result belong are not mutually exclusive, it is determined that there is no misidentification in the component detection result of the image data.

8. An identification device for automobile parts, characterized in that: The device comprises: A framing module is used to obtain image data collected by a mobile terminal camera, and collect multiple frames of images to be recognized from the image data at preset time intervals; A detection module is used to perform component detection on each frame of the image to be identified using a pre-trained automobile component recognition model to obtain a component detection result corresponding to each frame of the image to be identified; A cache module, used for storing the component detection results corresponding to each frame of the image to be identified into a cache queue in sequence according to the timestamp of the image to be identified, wherein the cache queue only retains the component detection results of a preset time length; The output module is used to output the component recognition result of the image data according to the component detection result in the cache queue and the judgment result of the preset rule.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the automobile component identification method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for identifying automobile parts according to any one of claims 1 to 7 is implemented.

Images