Vehicle model intelligent recognition methods, systems, devices and media

Abstract

This invention discloses a vehicle model intelligent recognition method, system, device, and medium. The method includes creating a vehicle dataset; the vehicle dataset includes multiple vehicle images, each labeled with vehicle model information; training a vehicle model recognition network based on the vehicle dataset; acquiring a target image corresponding to a vehicle to be recognized; extracting features from the target image to obtain a vehicle feature matrix; and inputting the vehicle feature matrix into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the target vehicle. This invention enables intelligent vehicle model recognition without considering interference factors such as the vehicle's shooting angle, achieving high recognition accuracy.

Description

Technical Field

[0001] This invention relates to the field of vehicle inspection, specifically to a method, system, device, and medium for intelligent vehicle model recognition. Background Technology

[0002] Vehicle model recognition is an important function of modern intelligent transportation systems, with a wide range of applications. For example, in cases of speeding, running red lights, and other traffic violations, it is necessary to automatically identify the brand and model of the offending vehicle for better traffic supervision.

[0003] Currently, traditional vehicle model recognition technology typically analyzes and identifies car images captured from the front. While this technology can identify vehicle models, the accuracy of model identification is not high due to the complexity of images in the monitoring environment, the variable shooting angle of the vehicle, and numerous environmental interference factors. Summary of the Invention

[0004] In view of this, the present invention provides a vehicle model intelligent recognition method, system, device and medium to solve the above-mentioned technical problems.

[0005] This invention provides a vehicle model intelligent recognition method, the method comprising: Create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; A vehicle model recognition network is obtained by training the vehicle dataset. Obtain the image of the vehicle to be identified, and extract features from the image to obtain the vehicle feature matrix; The vehicle feature matrix is ​​input into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be identified.

[0006] Optionally, the vehicle feature matrix includes external contour features and vehicle geometric features.

[0007] Optionally, feature extraction is performed on the image to be identified to obtain a vehicle feature matrix, including: The background subtraction method is used to perform contour tracking on the image to be identified, thereby obtaining the external contour features and the vehicle geometric features.

[0008] Optionally, the vehicle geometric features include vehicle length, vehicle height, vehicle length-to-height ratio, image perimeter, image area, image dispersion, image axle spacing, and axle spacing-to-vehicle length ratio.

[0009] Optionally, before performing feature extraction on the image to be identified to obtain the vehicle feature matrix, the method further includes: The image to be identified is preprocessed.

[0010] Optionally, the image to be identified is preprocessed, including: Edge detection is performed on the image to be identified to obtain a vehicle region image; The vehicle area image is resized.

[0011] Optionally, before inputting the vehicle feature matrix into the vehicle model recognition network, the following steps are included: The vehicle feature matrix is ​​subjected to preset operations and fusion operations to obtain a feature map.

[0012] Furthermore, the present invention also provides a vehicle model intelligent recognition system, applied in the aforementioned vehicle model intelligent recognition method, comprising: A dataset creation module is used to create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; The network training module is used to train the vehicle model recognition network based on the vehicle dataset. The feature extraction module is used to acquire the image to be identified corresponding to the vehicle to be identified, and to extract features from the image to be identified to obtain the vehicle feature matrix; The vehicle model recognition module is used to input the vehicle feature matrix into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be recognized.

[0013] In addition, the present invention also provides a vehicle model intelligent recognition device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed, it implements the method steps in the aforementioned vehicle model intelligent recognition method.

[0014] In addition, the present invention also provides a computer storage medium comprising: at least one instruction that, when executed, implements the method steps of the aforementioned vehicle model intelligent recognition method.

[0015] The beneficial effects of this invention are as follows: a vehicle dataset is created by using multiple vehicle images labeled with vehicle model information, which facilitates the training of a vehicle model recognition network that can accurately identify vehicle models; through feature extraction of the image to be identified, a vehicle feature matrix containing multiple vehicle features is obtained, which can accurately reflect the vehicle model information of the vehicle to be identified; finally, the vehicle feature matrix is ​​input into the vehicle model recognition network to achieve vehicle model recognition of the vehicle to be identified. The vehicle model intelligent recognition method, system, device, and medium of the present invention can realize intelligent vehicle model recognition without considering interference factors such as the vehicle's shooting angle, and the recognition accuracy is high. Attached Figure Description

[0016] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings: Figure 1 A flowchart of a vehicle model intelligent recognition method according to Embodiment 1 of the present invention is shown; Figure 2 The flowchart of the vehicle model recognition network trained in Embodiment 1 of the present invention is shown; Figure 3 A flowchart of another vehicle model intelligent recognition method according to Embodiment 1 of the present invention is shown; Figure 4 A schematic diagram of a vehicle model intelligent recognition system according to Embodiment 2 of the present invention is shown. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 like Figure 1 As shown, a vehicle model intelligent recognition method includes: S1: Create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; S2: Train the vehicle model recognition network based on the vehicle dataset; S3: Obtain the image of the vehicle to be identified, and extract features from the image to obtain the vehicle feature matrix; S4: Input the vehicle feature matrix into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be identified.

[0019] In this embodiment, a vehicle dataset is created using multiple vehicle images labeled with vehicle model information, which facilitates the training of a vehicle model recognition network that can accurately identify vehicle models. By extracting features from the images to be identified, a vehicle feature matrix containing multiple vehicle features is obtained. These vehicle features can accurately reflect the vehicle model information of the vehicle to be identified. Finally, the vehicle feature matrix is ​​input into the vehicle model recognition network to achieve vehicle model recognition.

[0020] The vehicle model intelligent recognition method in this embodiment can achieve intelligent vehicle model recognition without considering interference factors such as the vehicle's shooting angle, and has a high recognition accuracy.

[0021] Specifically, in this embodiment S1, each vehicle image is manually annotated to achieve the corresponding vehicle model information.

[0022] Preferably, such as Figure 2 As shown, S2 includes: S21: Randomly select multiple vehicle images from the vehicle dataset, and preprocess each selected vehicle image to create a training set; S22: Extract features from each vehicle image in the training set to obtain a training feature matrix corresponding to each vehicle image; S23: Using the EfficientNet composite network of the YOLOv3 framework, the vehicle model recognition network is trained based on the training feature matrix of each vehicle image and the vehicle model information to obtain the vehicle model recognition network.

[0023] Training the network using the above method can facilitate accurate identification of vehicle models in the future, and can also directly and quickly identify vehicle models based on the vehicle feature matrix of the vehicle to be identified, thereby improving identification efficiency.

[0024] Specifically, the training feature matrix for each vehicle image includes external contour features and vehicle geometric features.

[0025] Specifically, the vehicle feature matrix of the image to be identified includes external contour features and vehicle geometric features.

[0026] The vehicle geometric features include vehicle length, vehicle height, vehicle length-to-height ratio, image perimeter, image area, image dispersion, image axle spacing, and axle spacing-to-vehicle length ratio.

[0027] The vehicle characteristics described above can accurately reflect the vehicle model, making it easy to identify the model.

[0028] Preferably, in step S3, before performing feature extraction on the image to be identified to obtain the vehicle feature matrix, the method further includes: The image to be identified is preprocessed.

[0029] Preferably, the image to be identified is preprocessed, including: Edge detection is performed on the image to be identified to obtain a vehicle region image; The vehicle area image is resized.

[0030] Edge detection extracts the vehicle region image from the image to be identified, removing interference from irrelevant information such as background and environment, which helps improve the accuracy of vehicle model recognition. Size processing unifies the vehicle region image to a fixed size, making it easy to accurately identify vehicle models for vehicle region images taken from different shooting angles.

[0031] Specifically, size processing refers to resizing, such as resizing the vehicle area image to 224 pixels * 224 pixels.

[0032] Preferably, such as Figure 3 As shown, prior to S4, the method further includes: S5: Perform preset operations and fusion operations on the vehicle feature matrix to obtain a feature map.

[0033] Through the aforementioned preset and fusion operations, fast and accurate vehicle model recognition can be achieved, as well as multi-size recognition, thus improving the versatility of vehicle model recognition.

[0034] Specifically, the preset operations include convolutional layer operations, convolutional block operations, and convolution operations; the fusion operations include block operations, max pooling operations, and summation operations.

[0035] Example 2 like Figure 4 As shown, a vehicle model intelligent recognition system, applied in the vehicle model intelligent recognition method of Embodiment 1, includes: A dataset creation module is used to create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; The network training module is used to train the vehicle model recognition network based on the vehicle dataset. The feature extraction module is used to acquire the image to be identified corresponding to the vehicle to be identified, and to extract features from the image to be identified to obtain the vehicle feature matrix; The vehicle model recognition module is used to input the vehicle feature matrix into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be recognized.

[0036] A dataset creation module is used to create a vehicle dataset using multiple vehicle images labeled with vehicle model information. This facilitates the training of a vehicle model recognition network that can accurately identify vehicle models using a network training module. A feature extraction module extracts features from the images to be identified, resulting in a vehicle feature matrix containing multiple vehicle features that accurately reflect the vehicle model information. Finally, the vehicle model recognition module inputs the vehicle feature matrix into the vehicle model recognition network to achieve vehicle model recognition.

[0037] The vehicle model intelligent recognition system in this embodiment can achieve intelligent vehicle model recognition without considering interference factors such as the vehicle's shooting angle, and has a high recognition accuracy.

[0038] The vehicle model intelligent recognition function described in this embodiment corresponds to the steps of the vehicle model intelligent recognition method in Embodiment 1. For details not covered in this embodiment, please refer to Embodiment 1 and... Figures 1 to 3 The specific details will not be repeated here.

[0039] Example 3 A vehicle model intelligent recognition device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed, it implements the method steps of the vehicle model intelligent recognition method of Embodiment 1.

[0040] The computer program, stored in the memory and running on the processor, can intelligently recognize vehicle models without considering interference factors such as the shooting angle of the vehicle, and has a high recognition accuracy.

[0041] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the computer device, connecting all parts of the computer device through various interfaces and lines.

[0042] Memory can be used to store computer programs and / or models. The processor performs various functions of the computer device by running or executing the computer programs and / or models stored in the memory, and by accessing data stored in the memory. Memory can primarily include a program storage area and a data storage area. The program storage area can store the operating system and at least one application program required for a function (e.g., sound playback, image playback, etc.); the data storage area can store data created based on the use of the mobile phone (e.g., audio data, video data, etc.). Furthermore, memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, SmartMedia Cards (SMC), Secure Digital (SD) cards, Flash Cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0043] It should be understood that each block of a flowchart and / or block diagram, and combinations of blocks in a flowchart and / or block diagram, can be implemented by a computer program. These computer programs can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that instructions executable by the processor of the computer or other programmable data processing device generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0044] These computer programs may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0045] These computer programs may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0046] This embodiment also provides a computer storage medium, which includes at least one instruction that, when executed, implements the method steps of the vehicle model intelligent recognition method of Embodiment 1.

[0047] By executing a computer storage medium containing at least one instruction, intelligent vehicle model recognition can be achieved without considering interference factors such as the vehicle's shooting angle, resulting in high recognition accuracy.

[0048] Similarly, for details not covered in Embodiment 3, please refer to Embodiments 1 and 2. Figures 1 to 4 The specific details will not be repeated here.

[0049] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for intelligent vehicle model recognition, characterized in that, The method includes: Create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; A vehicle model recognition network is obtained by training the vehicle dataset. Obtain the image of the vehicle to be identified, and extract features from the image to obtain the vehicle feature matrix; The vehicle feature matrix is ​​input into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be identified.

2. The method according to claim 1, characterized in that, The vehicle feature matrix includes external contour features and vehicle geometric features.

3. The method according to claim 2, characterized in that, Feature extraction is performed on the image to be identified to obtain a vehicle feature matrix, including: The background subtraction method is used to perform contour tracking on the image to be identified, thereby obtaining the external contour features and the vehicle geometric features.

4. The method according to claim 2, characterized in that, The vehicle geometric features include vehicle length, vehicle height, vehicle length-to-height ratio, image perimeter, image area, image dispersion, image axle spacing, and axle spacing-to-vehicle length ratio.

5. The method according to claim 1, characterized in that, Before performing feature extraction on the image to be identified to obtain the vehicle feature matrix, the method further includes: The image to be identified is preprocessed.

6. The method according to claim 5, characterized in that, Preprocessing the image to be identified includes: Edge detection is performed on the image to be identified to obtain a vehicle region image; The vehicle area image is resized.

7. The method according to any one of claims 1 to 6, characterized in that, Before inputting the vehicle feature matrix into the vehicle model recognition network, the process includes: The vehicle feature matrix is ​​subjected to preset operations and fusion operations to obtain a feature map.

8. A vehicle model intelligent recognition system, characterized in that, Applied to the vehicle model intelligent recognition method as described in any one of claims 1 to 7, comprising: A dataset creation module is used to create a vehicle dataset; the vehicle dataset includes multiple vehicle images, and each vehicle image is labeled with vehicle model information; The network training module is used to train the vehicle model recognition network based on the vehicle dataset. The feature extraction module is used to acquire the image to be identified corresponding to the vehicle to be identified, and to extract features from the image to be identified to obtain the vehicle feature matrix; The vehicle model recognition module is used to input the vehicle feature matrix into the vehicle model recognition network to obtain the vehicle model recognition result corresponding to the vehicle to be recognized.

9. A vehicle model intelligent recognition device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed, implements the steps of the method as claimed in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer storage medium includes at least one instruction that, when executed, implements the steps of the method as described in any one of claims 1 to 7.

Images