Dangerous goods vehicle monitoring method and system

Abstract

The method is suitable for dangerous goods vehicle monitoring in a special area, and comprises the following steps: obtaining target image information of a target vehicle; inputting the target image information into a dangerous goods identification model to obtain an identification result; when the identification result is a dangerous goods vehicle, generating a virtual image corresponding to the target image information; acquiring real-time position information of the target vehicle; displaying the virtual image in a map interface corresponding to a target area according to the real-time position information; by means of the method, the dangerous goods vehicles in the target area can be monitored in real time, the real-time positions of all the dangerous goods vehicles in the target area are displayed on the map interface in real time, and the real-time positions are displayed on the map in the mode of the virtual image corresponding to the target vehicle. A user can conveniently and intuitively obtain the information of each dangerous goods vehicle on the map interface, more key information can be displayed in a limited display area as much as possible in a virtual image display mode, and the display effect is better.

Description

Technical Field

[0001] This application belongs to the technical field of operation safety, and particularly relates to a method and system for monitoring dangerous goods vehicles. Background Art

[0002] In industries such as industry, construction, transportation, environment, energy, coal mines, oil and gas, on-site operations face the current situation of numerous points, wide coverage, small, miscellaneous and scattered, many new types of operations, high production risks, and great difficulty in quality control. For example, with the booming development of port operations, the transportation volume of dangerous goods shows a continuous growth trend, and the safety supervision work during its transportation process has become a key link to ensure the stable operation of the port. However, there is currently a lack of a solution for monitoring dangerous goods vehicles on specific routes, and managers cannot conveniently obtain various information of dangerous goods vehicles on specific routes. Summary of the Invention

[0003] An embodiment of this application provides a method for monitoring dangerous goods vehicles, which is used for monitoring dangerous goods vehicles in a target area.

[0004] In a first aspect, this application provides a method for monitoring dangerous goods vehicles, including:

[0005] Obtain target image information of a target vehicle;

[0006] Input the target image information into a dangerous goods recognition model to obtain a recognition result;

[0007] When the recognition result is a dangerous goods vehicle, generate a virtual image corresponding to the target image information;

[0008] Obtain the real-time position information of the target vehicle;

[0009] Display the virtual image in a map interface corresponding to the target area according to the real-time position information.

[0010] Further,

[0011] The obtaining of the target image information of the target vehicle includes:

[0012] Receive target image information at at least two angles uploaded by a camera at the entrance of the target area. The target image information at at least two angles includes a first target image at a first angle for displaying the license plate number and a second target image at a second angle for displaying the goods loaded on the vehicle.

[0013] Further, the recognition result includes at least one of the following:

[0014] First result information for characterizing whether the target vehicle belongs to a dangerous goods vehicle;

[0015] The second result information used to characterize the dangerous goods category on the target vehicle;

[0016] The third result information used to characterize the license plate number information of the target vehicle;

[0017] The fourth result information used to characterize the vehicle type category of the target vehicle;

[0018] The fifth result information used to characterize the body color of the target vehicle.

[0019] Further, when the recognition result is a dangerous goods vehicle, a virtual image corresponding to the target vehicle is generated; the method includes the following steps:

[0020] Send the second result information, the third result information, the fourth result information, and the fifth result information to the virtual image generation model to obtain the virtual image corresponding to the target vehicle.

[0021] Further, obtaining the real-time position information of the target vehicle includes the following steps:

[0022] Obtain the position information of the target vehicle through the GPS module on the target vehicle.

[0023] Further, the dangerous goods recognition model is trained in the following manner:

[0024] Obtain the first sample data, where the first sample data includes the first target image, the second target image corresponding to a number of sample vehicles, the first result information, the second result information, the third result information, the fourth result information, and the fifth result information corresponding to the sample vehicles;

[0025] Use the first target image and the second target image in the first sample data as input data, and the first result information, the second result information, the third result information, the first result information, and the fifth result information as output data to train the first initial model to obtain the trained dangerous goods recognition model.

[0026] Further, the virtual image generation model is trained in the following manner:

[0027] Obtain the second sample data, where the second sample data includes the first target image, the second target image corresponding to a number of sample vehicles, the second result information, the third result information, the fourth result information, and the fifth result information corresponding to the sample vehicles;

[0028] Use the first target image, the second target image, the second result information, the third result information, the fourth result information, and the fifth result information in the second sample data as input data, and use the virtual image as the output data to train the second initial model to obtain the virtual image generation model.

[0029] In a second aspect, the present application provides a dangerous goods vehicle monitoring system, including:

[0030] An image acquisition module for acquiring image information of a target vehicle in a target area;

[0031] A dangerous goods identification module for determining whether the target vehicle is a dangerous goods vehicle according to the image information;

[0032] A virtual image generation module for generating a virtual image corresponding to the target vehicle when it is determined to be a dangerous goods vehicle;

[0033] A position acquisition module for acquiring real-time position information of the target vehicle;

[0034] A result display module for displaying the virtual image in a map interface corresponding to the target area according to the real-time position information.

[0035] In a third aspect, the present application further provides an electronic device, including a memory and a processor, the memory is used to store a computer application program, and the processor is used to execute the method in any of the above aspects when executing the computer application program.

[0036] The beneficial effects of the embodiments of the present application compared with the prior art are as follows: It is possible to realize real-time monitoring of dangerous goods vehicles in the target area, display the real-time positions of all dangerous goods vehicles in the target area in real time on the map interface, and display them on the map in the form of a virtual image corresponding to the target vehicle, which is convenient for users to intuitively obtain information about each dangerous goods vehicle on the map interface. The display method of the virtual image can display more key information within a limited display area as much as possible, and the display effect is better. Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0038] Figure 1 It is a flowchart of the method provided by an embodiment of the present application;

[0039] Figure 2 It is a schematic diagram of a target area in an embodiment of the present application;

[0040] Figure 3 It is a system block diagram provided by an embodiment of the present application;

[0041] Figure 4 It is a schematic diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0042] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system architectures, technologies, etc. are proposed to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0043] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0044] It should also be understood that the term "and / or" used in the specification and the appended claims of the present application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0045] The references to "an embodiment" or "some embodiments" or the like described in the specification of the present application mean that specific features, structures, or characteristics described in conjunction with the embodiment are included in one or more embodiments of the present application. Thus, the phrases "in an embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0046] The technical solutions provided by the embodiments of the present application will be introduced below through specific embodiments.

[0047] Embodiment 1

[0048] This embodiment provides a method for monitoring dangerous goods vehicles, which can be used for monitoring dangerous goods vehicles in specific areas, such as monitoring the area within a container yard at a port terminal, etc., and can greatly improve the supervision efficiency and build a solid defense line for transportation safety. Specifically, as Figure 1 shown, it includes the following steps:

[0049] S1: Obtain the target image information of the target vehicle;

[0050] S2: Input the target image information into the dangerous goods recognition model to obtain the recognition result;

[0051] S3: When the recognition result is a dangerous goods vehicle, generate a virtual image corresponding to the target image information;

[0052] S4: Obtain the real-time position information of the target vehicle;

[0053] S5: Display the virtual image in the map interface corresponding to the target area according to the real-time position information.

[0054] Specifically, the embodiment of the present application can obtain the image information of the target vehicle at the entrance of a specific area and send the image information of the target vehicle to the dangerous goods recognition model for recognition. The recognition result can include being a dangerous goods vehicle or not being a dangerous goods vehicle. When the recognition result is a dangerous goods vehicle, a virtual image corresponding to the target vehicle can be generated, and the virtual image can be displayed in real time in the map interface according to the real-time position information of the target vehicle. In this embodiment, a dangerous goods vehicle refers to a vehicle transporting dangerous goods. The same vehicle can transport non-dangerous goods or dangerous goods. When the vehicle transports dangerous goods, the technical solution provided by this embodiment can identify the dangerous goods. Specifically, the method provided by this embodiment includes a map interface of the target area. The above map interface can be drawn by the development team itself or directly obtained through third-party channels (such as Gaode Map, Baidu Map, etc.). The method provided by the present application can display a virtual image corresponding to a dangerous goods vehicle (i.e., the target vehicle) in the map interface of the target area, and the position of the virtual image in the map interface can be updated in real time according to the actual position of the dangerous goods vehicle, enabling managers to more intuitively observe the implementation situation of dangerous goods vehicles in the target area and providing a more efficient and convenient management solution.

[0055] In an alternative embodiment, the above step S1 specifically includes:

[0056] S11. Receive target image information at at least two angles uploaded by a camera at the entrance of the target area. The target image information at the at least two angles includes a first target image at a first angle for displaying a license plate number and a second target image at a second angle for displaying goods loaded on the vehicle.

[0057] Specifically, when a target vehicle appears at the entrance of the target area, in order to more clearly display the cargo image on the target vehicle and the license plate information of the target vehicle, in the embodiment of the present application, the target vehicle is image-captured by cameras at at least two angles. The angle of one camera can be directly facing the entrance direction, and the shooting angle of the other camera is perpendicular to the angle of the first camera, so as to simultaneously capture the front and side images of the target vehicle. This facilitates further processing of the information of the target vehicle later. For dangerous goods containers, there are usually signs of dangerous goods on their sides. Therefore, in the present application, by capturing the side image of the vehicle, it is convenient to identify the dangerous goods signs on the side of the container, and at the same time, it is necessary to ensure that the captured side image is complete to avoid the situation where the side image is not captured completely and the dangerous goods sign is not recognized.

[0058] In an alternative embodiment, the recognition result includes at least one of the following:

[0059] First result information for characterizing whether the target vehicle belongs to a dangerous goods vehicle;

[0060] Second result information for characterizing the category of dangerous goods on the target vehicle;

[0061] Third result information for characterizing the license plate number information of the target vehicle;

[0062] Fourth result information for characterizing the vehicle type category of the target vehicle;

[0063] Fifth result information for characterizing the body color of the target vehicle.

[0064] Specifically, the dangerous goods identification model can not only identify whether the target vehicle is a dangerous goods vehicle, but also simultaneously identify other basic information of the target vehicle, such as license plate number, body color, vehicle type category, and dangerous goods category, etc. The license plate number information and the body color information are the information directly reflected in the target image information. The dangerous goods identification model can extract specific pixel regions in the target image information and use the RGB values in the pixel regions as the fifth result information representing the color. The license plate information can be obtained by performing character recognition on the pixel image of the specific region. Whether it belongs to a dangerous goods vehicle, the vehicle type category, and the dangerous goods category can be obtained by training the first initial model with a large number of sample images. The first initial model can adopt the Transformer large model, which is a deep neural network architecture based on the self-attention mechanism and is designed specifically for processing sequence data and has become the core basic model in the fields of natural language processing and generative AI. In another alternative embodiment, a computer vision (CV) large model can also be adopted. Such models are mainly used for processing images and videos, such as object detection, semantic segmentation, image generation, etc. For example, Inception, ResNet, DenseNet, etc.

[0065] In an alternative embodiment, step S3 above specifically includes the following steps:

[0066] Send the second result information, the third result information, the fourth result information, and the fifth result information to the virtual image generation model to obtain the virtual image corresponding to the target vehicle.

[0067] Specifically, the method provided in this embodiment requires pre-training a virtual image generation model, which is used to generate a virtual image corresponding to the target vehicle according to the second result information, the third result information, and the fourth result information of the target vehicle. The virtual image is used for display on the map interface to provide a more intuitive and convenient monitoring system for the management personnel. Since it is difficult to intuitively and quickly display information such as license plate numbers, cargo types, and vehicle types simultaneously in the real photo of the target vehicle, and the real photo occupies too much area in the limited space of the map interface, while the virtual image can concisely and intuitively highlight the display of vehicle types, colors, dangerous goods categories, and license plate numbers. The vehicle type and color can be directly displayed through the appearance of the virtual image, and the license plate number and dangerous goods category can also be enlarged at will according to needs to highlight the display, thus achieving an excellent monitoring effect. Therefore, the virtual image is the optimal choice. The virtual image generation model can integrate the above information on a virtual image according to information such as vehicle type, color, dangerous goods category, and license plate number. Specifically, during the training process of the virtual image generation model, all common types of vehicles can be converted into virtual elements corresponding to their shapes in advance, and dangerous goods categories, colors, etc. can be converted into corresponding virtual elements. During the generation stage, the model can generate a virtual image corresponding to the target vehicle by calling various preset virtual elements and combining them adaptively.

[0068] In an optional embodiment, the above step S4 may specifically include obtaining the position information of the target vehicle through the GPS module on the target vehicle. The position information of the target vehicle can be accurately obtained through the GPS module on the target vehicle, and the position information collected by the GPS module can be transmitted to the background server through a wireless network (such as 4G, 5G, etc.), so as to display the virtual image in real time in the area corresponding to the position information on the map interface.

[0069] In an optional embodiment, the dangerous goods recognition model is trained in the following manner:

[0070] Obtain first sample data, where the first sample data includes first target images, second target images corresponding to a number of sample vehicles, first result information, second result information, third result information, fourth result information, and fifth result information corresponding to the sample vehicles;

[0071] Use the first target images and second target images in the first sample data as input data, and the first result information, second result information, third result information, first result information, and fifth result information as output data to train the first initial model to obtain a trained dangerous goods recognition model.

[0072] Specifically, the first sample data is used to train the dangerous goods identification model. The first target image and the second target image in the first sample data can be captured by a camera during the daily operation of the target area. The first result information, the second result information, and the fourth result information can be manually annotated on the corresponding images. The third result information can be directly identified by the image recognition module in the dangerous goods identification model through the image recognition function. Specifically, obtaining license plate number information through image recognition technology is an existing technology in the art and will not be elaborated in this embodiment. Similarly, obtaining body color information through image recognition technology also belongs to the existing technology in the art and will not be elaborated in this application embodiment. After obtaining a sufficient number of first target images, second target images, and the corresponding first result information, second result information, and fourth result information, the first target images and the second target images are used as input data, and the first result information, the second result information, the third result information, the first result information, and the fifth result information are used as output data to train the first initial model to obtain the trained dangerous goods identification model. For dangerous goods containers, there are usually signs of dangerous goods on their sides. During the training process of the model, the model can identify whether a dangerous goods sign is collected by recognizing the image on the side of the vehicle, thereby serving as the basis for the model to determine whether it is a vehicle transporting dangerous goods. At the same time, it is also necessary to ensure that the collected side image is complete to avoid the situation where the side image is incompletely collected and the dangerous goods sign is not recognized. Specifically, the above first initial model can be based on a deep learning architecture, such as a convolutional neural network (CNN) or a vision Transformer (ViT) in particular. Through the above training process, the large model can accurately identify whether the target vehicle is a vehicle transporting dangerous goods, the dangerous goods category information of the target vehicle, and the vehicle type category information of the target vehicle through the first sample images.

[0073] In an alternative embodiment, the virtual image generation model is trained in the following manner:

[0074] Obtain second sample data, where the second sample data includes first target images, second target images corresponding to a number of sample vehicles, second result information, third result information, fourth result information, and fifth result information corresponding to the sample vehicles;

[0075] Taking the first target image, the second target image, the second result information, the third result information, the fourth result information, and the fifth result information in the second sample data as input data, and using a virtual image as output data to train the second initial model to obtain the virtual image generation model. Further, the above-mentioned second initial model can be an architecture based on a diffusion model, such as a model of the Latent Diffusion architecture, which can implement the function of generating an image from an image. For another example, it can be a model of the VQ-VAE-2 architecture. Further, some elements related to the target vehicle or user-defined elements can also be preset in the second initial model. For example, elements with different virtual image styles can enable the second initial model to generate virtual images related to the preset elements during the training process.

[0076] Through the above technical solution, it is possible to achieve real-time monitoring of the vehicles transporting dangerous goods in the target area, display the real-time positions of all dangerous goods vehicles in the target area in real time on the map interface, and display them on the map in the form of a virtual image corresponding to the target vehicle, which is convenient for users to intuitively obtain information about each dangerous goods vehicle on the map interface. The display method of the virtual image can display more key information within a limited display area as much as possible, and the display effect is better.

[0077] In an optional embodiment, cameras are set at preset intervals in the target area for shooting video information within the shooting range of the cameras, so as to achieve seamless coverage of the target area. Further, in the background server, each video interface uploaded by a camera is provided with a corresponding access link, and each access link corresponding to a camera has a regional code according to the position information of the camera. That is, assuming that there are 20 cameras in the target area that divide the target area into 20 sub-areas, as Figure 2 shown, (the black solid dots in the figure represent cameras) there are 20 access links, and each access link contains a regional code. Assuming the regional codes are 001 - 0020, then 001 - 0020 are respectively part of each access link. The method provided in this embodiment further includes: based on the position information of the target vehicle, mapping the access link corresponding to the position information to the virtual image.

[0078] Specifically, the area code in the access link can correspond to any location information of the target vehicle within the target area, that is, all locations within the target area are pre-configured with area codes. For example, when the vehicle appears in area 001, the area code corresponding to area 001 is 001, and then the access link corresponding to 001 is mapped to the virtual image corresponding to the target vehicle; when the target vehicle moves from area 001 to area 002, the area code corresponding to area 002 is 002, and then the access link corresponding to 002 is mapped to the virtual image, and so on, until the vehicle travels to the last area 0020, and the access link corresponding to 0020 is mapped to the virtual image.

[0079] In one embodiment, the method further includes:

[0080] In response to the trigger of the access link, play the video picture corresponding to the access link.

[0081] The user can click on the virtual image at any time in the map interface. Since the access link is mapped on the virtual image, when the server receives the signal that the access link is triggered, it can pop up the video picture associated with the access link, so that the user can directly switch from the map interface to the implementation video picture of the target vehicle, facilitating the user to monitor the on-site situation.

[0082] Embodiment 2

[0083] This embodiment provides a control system for dangerous goods vehicles, as Figure 3 shown, including:

[0084] An image acquisition module 210, configured to acquire image information of the target vehicle in the target area;

[0085] A dangerous goods identification module 220, configured to determine whether the target vehicle is a dangerous goods vehicle according to the image information;

[0086] A virtual image generation module 230, configured to generate a virtual image corresponding to the target vehicle when it is determined to be a dangerous goods vehicle;

[0087] A position acquisition module 240, configured to acquire the real-time position information of the target vehicle;

[0088] A result display module 250, configured to display the virtual image in the map interface corresponding to the target area according to the real-time position information.

[0089] This embodiment further provides an electronic device, including a memory and a processor, where the memory is used to store computer application programs, and the processor is configured to execute any of the methods in the above embodiments when executing the computer application programs.

[0090] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0091] This embodiment of the present invention also provides an electronic device, as Figure 4 shown, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, each process of the above-mentioned dangerous goods vehicle monitoring method is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be elaborated here. In addition, the electronic device provided in this embodiment can be a server.

[0092] This embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned charging voltage threshold determination method or charging strategy determination method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be elaborated here. Among them, the computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0093] It should be noted that in this article, the terms "including", "comprising", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article, or device. Without further limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article, or device including that element.

[0094] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0095] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the purpose and claims of the present invention, and all of them fall within the protection scope of the present invention.

[0096] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for monitoring dangerous goods vehicles, characterized in that, Including the following steps: Obtain the target image information of the target vehicle; Input the target image information into the dangerous goods recognition model to obtain the recognition result; When the recognition result is a dangerous goods vehicle, generate a virtual image corresponding to the target image information; Obtain the real-time position information of the target vehicle; Display the virtual image in the map interface corresponding to the target area according to the real-time position information.

2. The method according to claim 1, wherein Including the following steps: The obtaining of the target image information of the target vehicle includes: Receive the target image information at at least two angles uploaded by the camera at the entrance of the target area. The target image information at at least two angles includes a first target image at a first angle for displaying the license plate number and a second target image at a second angle for displaying the goods loaded on the vehicle.

3. The method according to claim 1, characterized in that The recognition result includes at least one of the following: First result information for characterizing whether the target vehicle belongs to a dangerous goods vehicle; Second result information for characterizing the type of dangerous goods on the target vehicle; Third result information for characterizing the license plate number information of the target vehicle; Fourth result information for characterizing the vehicle type category of the target vehicle; Fifth result information for characterizing the body color of the target vehicle.

4. The method according to claim 3, wherein When the recognition result is a dangerous goods vehicle, generating a virtual image corresponding to the target vehicle; includes the following steps: Send the second result information, the third result information, the fourth result information, and the fifth result information to the virtual image generation model to obtain a virtual image corresponding to the target vehicle.

5. The method according to claim 1, wherein Obtaining the real-time position information of the target vehicle includes the following steps: Obtain the position information of the target vehicle through the GPS module on the target vehicle.

6. The method according to claim 3, characterized in that, The dangerous goods recognition model is trained in the following manner: Obtain first sample data, where the first sample data includes a first target image, a second target image corresponding to a number of sample vehicles, first result information, second result information, third result information, fourth result information, and fifth result information corresponding to the sample vehicles; Use the first target image and the second target image in the first sample data as input data, and the first result information, second result information, third result information, first result information, and fifth result information as output data to train the first initial model to obtain the trained dangerous goods recognition model.

7. The method according to claim 1, characterized in that, The virtual image generation model is trained in the following manner: Obtain second sample data, where the second sample data includes a first target image, a second target image corresponding to a number of sample vehicles, second result information, third result information, fourth result information, and fifth result information corresponding to the sample vehicles; Use the first target image, the second target image, the second result information, third result information, fourth result information, and fifth result information corresponding to the sample vehicles in the second sample data as input data, and use the virtual image as output data to train the second initial model to obtain the virtual image generation model.

8. A dangerous goods vehicle monitoring system, characterized in that, Including: An image acquisition module for acquiring the image information of the target vehicle in the target area; A hazardous material identification module, configured to determine whether the target vehicle is a hazardous material vehicle according to the image information; A virtual image generation module, configured to generate a virtual image corresponding to the target vehicle when it is determined that the vehicle is a hazardous material vehicle; A location acquisition module, configured to acquire real-time location information of the target vehicle; A result display module, configured to display the virtual image in a map interface corresponding to the target area according to the real-time location information.

9. An electronic device, characterized in that, It includes a memory and a processor. The memory is used to store computer application programs, and the processor is configured to execute the method according to any one of claims 1-7 when executing the computer application programs.

Images