A method, device, terminal and storage medium for classifying and identifying dense fog on highways.

Abstract

This invention discloses a method, device, terminal, and storage medium for classifying and identifying dense fog on highways in the field of traffic meteorology. It aims to solve the problem that existing technologies suffer from low precision in dense fog classification and early warning due to the sparse monitoring density of traffic meteorological stations and video surveillance points along highways. The method includes obtaining the visibility level of each unit using optimal interpolation based on unitized highway network information, the visibility levels of each traffic meteorological station, and the visibility levels of each video surveillance point; and obtaining a dataset based on the visibility levels of each unit. This invention utilizes optimal interpolation to obtain the visibility levels of each unit, thereby achieving accurate classification and identification of dense fog on highways even when the density of traffic meteorological stations and video surveillance points is insufficient, providing a reference for highway safety operation scheduling and management.

Description

Technical Field

[0001] This invention relates to a method, device, terminal, and storage medium for classifying and identifying dense fog on highways, belonging to the field of traffic meteorology technology. Background Technology

[0002] Highways, or expressways for short, are roads specifically designed for high-speed automobile travel. Due to the high speeds vehicles travel on highways, it is necessary to monitor meteorological factors that may affect traffic flow. With the rapid expansion of the highway network, traffic safety issues under complex weather conditions are becoming increasingly prominent. Dense fog, due to its sudden drop in visibility and unpredictable nature, has become a major cause of serious accidents such as multi-vehicle pileups. Therefore, it is essential to monitor dense fog using traffic meteorological stations and video surveillance points along highways.

[0003] In existing technologies, the sparse monitoring density of traffic meteorological stations and video surveillance points along highways cannot support refined fog classification early warnings, resulting in low precision in fog classification early warnings and affecting the safe operation, scheduling, and management of highways. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, device, terminal and storage medium for classifying and identifying dense fog on highways. It makes full use of highway network information, traffic and meteorological station information along the highway and video surveillance point information, and uses the optimal interpolation method to obtain the visibility level of each unit. Thus, even when the density of traffic and meteorological stations and video surveillance points is insufficient, it can achieve accurate classification and identification of dense fog on highways, and provide a reference for the safe operation scheduling and management of highways.

[0005] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution: In a first aspect, the present invention provides a method for classifying and identifying dense fog on highways, comprising the following steps: Obtain highway network information; The highway network information is modularized to obtain modularized highway network information; wherein the modularized highway network information includes multiple units; Based on the highway network information, obtain information on traffic and meteorological stations along the highway and video surveillance points; Based on the information from each of the aforementioned traffic and meteorological stations, obtain the visibility level for each traffic and meteorological station; Based on the information of each video monitoring point, obtain images of each video monitoring point, input the images of the video monitoring points into a pre-built and trained hybrid model, and output the visibility level of each video monitoring point. Based on the unitized highway network information, the visibility levels of various traffic and meteorological stations, and the visibility levels of various video surveillance points, the visibility level of each unit is obtained using the optimal interpolation method. Based on the visibility level of each unit, obtain the dataset; Based on a preset filling scheme, the dataset is filled to obtain a highway fog level warning map, thus completing the highway fog level identification work.

[0006] Furthermore, the highway network information includes the latitude and longitude of the highway network, as well as the starting and ending points of the highway network. Based on the starting and ending points of the highway network, the mileage of the highway network is obtained.

[0007] Furthermore, the step of unitizing the highway network information to obtain the unitized highway network information specifically includes: Based on the starting and ending points of the highway network, the highway network information is modularized using ArcGIS software at preset intervals to obtain multiple units.

[0008] Furthermore, obtaining the visibility level of each traffic weather station based on the information from each traffic weather station specifically includes: Based on the information from each of the aforementioned traffic and weather stations, obtain the visibility of each traffic and weather station; Based on the visibility classification standard, the visibility level of each traffic weather station is obtained according to the visibility of each traffic weather station.

[0009] Furthermore, the hybrid model is a hybrid model of convolutional neural network and transformer.

[0010] Furthermore, the method of obtaining the visibility level of each cell using optimal interpolation specifically includes: Obtain real-time visibility data within a preset grid point; Based on the visibility data within the preset grid points, obtain the visibility grid point values; Based on the unitized highway network information, highway traffic and meteorological station information, and video surveillance point information, the visibility grid value is interpolated into each unit, each traffic and meteorological station, and each video surveillance point using bilinear interpolation. Let the initial estimated values ​​of visibility grid points inserted into each unit, each traffic and weather station, and each video surveillance point be classified as follows: ; Determine whether there is a traffic weather station or video surveillance point within the unit. If so, use the visibility level of the traffic weather station or the visibility level of the video surveillance point as the visibility level of the unit. If it does not exist, calculate the visibility level of this unit using the following formula: ; ; In the formula: For the output result; This is a round-down operation; The visibility level for this unit; W is the preset weight; This refers to the visibility level of the nearest traffic weather station or video surveillance point in the direction of the starting point of the highway network to this unit. This refers to the visibility level of the nearest traffic weather station or video surveillance point to the end of the highway network. The The acquisition specifically includes: Based on the highway network information, obtain the location information of each unit; Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the starting point of the highway network; Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point; The The acquisition specifically includes: Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the end of the highway network. Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point.

[0011] Furthermore, the step of obtaining the dataset based on the visibility level of each unit specifically includes: Based on the unitized highway network information and the starting and ending points of the highway network, the visibility levels of each unit are arranged sequentially to obtain the dataset. The process of filling the dataset based on a preset filling scheme specifically includes: The preset filling scheme includes a variety of colors that correspond one-to-one with the visibility level; Based on the visibility level of each unit, the unit is filled with the color corresponding to the visibility level.

[0012] Secondly, the present invention provides a highway fog classification identification device, specifically comprising: Data acquisition module: used to acquire highway network information; Unitization module: used to unitize the highway network information to obtain the unitized highway network information; wherein, the unitized highway network information includes multiple units; The first calculation module is used to obtain information on traffic and meteorological stations along the highway and video surveillance points based on the highway network information; to obtain the visibility level of each traffic and meteorological station based on the information on each traffic and meteorological station; to obtain images of each video surveillance point based on the information on each video surveillance point; to input the images of the video surveillance points into a pre-built and trained hybrid model; and to output the visibility level of each video surveillance point. The second calculation module is used to obtain the visibility level of each unit based on the unitized highway network information, the visibility level of each traffic and meteorological station, and the visibility level of each video surveillance point using the optimal interpolation method. Identification module: used to obtain a dataset based on the visibility level of each unit; fill the dataset based on a preset filling scheme to obtain a highway fog classification warning map and complete the highway fog classification identification work.

[0013] Thirdly, the present invention provides a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to the first aspect.

[0014] Fourthly, a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in the first aspect.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: This method for classifying and identifying dense fog on highways makes full use of highway network information, traffic and meteorological station information along the highway, and video surveillance point information. It uses the optimal interpolation method to obtain the visibility level of each unit, thereby achieving accurate classification and identification of dense fog on highways even when the density of traffic and meteorological stations and video surveillance points is insufficient, providing a reference for highway safety operation scheduling and management. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for classifying and identifying dense fog on highways according to an embodiment of the present invention. Figure 2This is a schematic flowchart of a highway fog classification and identification device according to an embodiment of the present invention; Figure 3 This is a diagram of the cTrans-Net network model structure provided in an embodiment of the present invention; Figure 4 This is a highway fog classification warning map based on a dataset provided in an embodiment of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0018] The term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Example 1:

[0019] like Figure 1 As shown, the present invention provides a method for classifying and identifying dense fog on highways, comprising the following steps: Obtain highway network information; The highway network information includes the latitude and longitude of the highway network, as well as the starting and ending points of the highway network. The mileage of the highway network is obtained based on the starting and ending points of the highway network.

[0020] The highway network information is modularized to obtain modularized highway network information; wherein the modularized highway network information includes multiple units; Based on the highway network information, obtain information on traffic and meteorological stations along the highway and video surveillance points; Based on the information from each of the aforementioned traffic and meteorological stations, obtain the visibility level for each traffic and meteorological station; Based on the information of each video monitoring point, obtain images of each video monitoring point, input the images of the video monitoring points into a pre-built and trained hybrid model, and output the visibility level of each video monitoring point. Based on the unitized highway network information, the visibility levels of various traffic and meteorological stations, and the visibility levels of various video surveillance points, the visibility level of each unit is obtained using the optimal interpolation method. Based on the visibility level of each unit, obtain the dataset; Based on a preset filling scheme, the dataset is filled to obtain a highway fog level warning map, thus completing the highway fog level identification work.

[0021] One embodiment involves unitizing highway network information to obtain unitized highway network information, specifically including: Based on the starting and ending points of the highway network, the highway network information is modularized using ArcGIS software at preset intervals to obtain multiple units.

[0022] Specifically, the starting point of the highway is taken as S0, and the first integer mileage point is taken as the initial point S1. Using ArcGIS software, each highway is divided into units with a length of 1KM. The length of the highway is nKM. Starting from the starting point of the highway, each unit is labeled sequentially as S1, S2, S3...Sn. Based on the obtained highway network information, the specific location information of each unit can be obtained.

[0023] In one embodiment, obtaining the visibility level of each traffic weather station based on the information from each traffic weather station specifically includes: Based on the information from each of the aforementioned traffic and weather stations, obtain the visibility of each traffic and weather station; Based on the visibility classification standard, the visibility level of each traffic meteorological station is obtained according to the visibility of each station, so that the visibility level of each traffic meteorological station corresponds to its respective traffic meteorological station information.

[0024] Specifically, the visibility grading standards are shown in Table 1:

[0025] In one embodiment, images of each video surveillance point are obtained based on the information of each video surveillance point, and the images of each video surveillance point are input into a pre-built and trained hybrid model to output the visibility level of each video surveillance point. The hybrid model is a hybrid model of convolutional neural network and transformer.

[0026] Specifically, a dataset is constructed by collecting video surveillance images along highways for model training and testing. Historical data is graded and labeled based on traffic station observation data. The video surveillance images are normalized. The ability of convolutional neural networks (CNNs) to learn local features and the ability of Transformers (a neural network architecture based on self-attention mechanism) to learn global features are utilized to build a CNN + Transformer hybrid model (cTrans-Net). The hybrid model is then used to carry out visibility grading recognition.

[0027] In one embodiment, obtaining the visibility level of each cell using optimal interpolation specifically includes: Obtain real-time visibility data within a preset grid; the preset grid consists of 5*5KM grid points. Based on the visibility data within the preset grid points, obtain the visibility grid point values; Based on the unitized highway network information, highway traffic and meteorological station information, and video surveillance point information, the visibility grid value is interpolated into each unit, each traffic and meteorological station, and each video surveillance point using bilinear interpolation. Let the initial estimated values ​​of visibility grid points inserted into each unit, each traffic and weather station, and each video surveillance point be classified as follows: ; Determine whether there is a traffic weather station or video surveillance point within the unit. If so, use the visibility level of the traffic weather station or the visibility level of the video surveillance point as the visibility level of the unit. If it does not exist, calculate the visibility level of this unit using the following formula: ; ; In the formula: For the output result; This is a round-down operation; The visibility level for this unit; W is the preset weight; This refers to the visibility level of the nearest traffic weather station or video surveillance point in the direction of the starting point of the highway network to this unit. This refers to the visibility level of the nearest traffic weather station or video surveillance point to the end of the highway network. The The acquisition specifically includes: Based on the highway network information, obtain the location information of each unit; Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the starting point of the highway network; Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point; The The acquisition specifically includes: Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the end of the highway network. Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point.

[0028] In one embodiment, obtaining the dataset based on the visibility level of each unit specifically includes: Based on the unitized highway network information and the starting and ending points of the highway network, the visibility levels of each unit are arranged sequentially to obtain the dataset. The process of filling the dataset based on a preset filling scheme specifically includes: The preset filling scheme includes a variety of colors that correspond one-to-one with the visibility level; Based on the visibility level of each unit, the unit is filled with the color corresponding to the visibility level.

[0029] Specifically, using the dataset, the road network units are color-coded according to visibility level. Road network cells with visibility level 0 are filled with red, those with visibility level 1 are filled with orange, those with visibility level 2 are filled with yellow, those with visibility level 3 are filled with blue, and those with visibility level 4 do not need to be filled. This completes the color filling work for each unit, thereby realizing the classification and identification of dense fog on highways.

[0030] In one embodiment, the latitude and longitude information of a certain expressway within the jurisdiction is obtained, with the starting mileage being K1508 and the ending mileage being K1630. K1508 is taken as the starting point S0 of the expressway, and the first integer mileage point is taken as the initial point S1 with a mileage of K1509. Using ArcGIS software, each expressway is divided into units with a length of 1KM. The expressway is 122KM long. Starting from the starting point of the expressway, each unit is labeled sequentially as S1, S2, S3...Sn. Information was obtained from four traffic weather stations and eight video surveillance points along the highway. Based on the information from each of the aforementioned traffic and meteorological stations, obtain the visibility level for each traffic and meteorological station; Based on the information from each video surveillance point, images of each video surveillance point are obtained. These images are then input into a pre-built and trained hybrid model, which outputs the visibility level of each video surveillance point. After evaluation, the cTrans-Net model, such as... Figure 3The accuracy (Acc) was 0.8917, recall (Rec) was 0.8947, precision (Pre) was 0.8942, area under the receiver operating characteristic curve (AUC) was 0.9822, and F1 score was 0.8942, indicating that cTrans-Net can not only accurately identify positive samples but also effectively distinguish between positive and negative samples, demonstrating a high comprehensive classification ability. The accuracy for identifying visibility levels L0-4 of the test samples was 92.31% for L0, 85.97% for L1, 90.66% for L2, 87.95% for L3, and 90.45% for L4, indicating that cTrans-Net can achieve accurate classification of different samples.

[0031] The cTrans-Net model was used to output images from all video surveillance points along the highway at a certain moment. The results of the video image classification and recognition are shown in Table 2.

[0032] Then, the visibility level of each unit is obtained using the optimal interpolation method. Based on the visibility level of each unit, the dataset is obtained. The visibility levels of each unit are shown in Table 3.

[0033] Finally, based on a preset filling scheme, the dataset is filled to obtain a highway fog level warning map, thus completing the highway fog level identification work. Figure 4 As shown.

[0034] This invention makes full use of highway network information, highway traffic and meteorological station information, and video surveillance point information, and uses the optimal interpolation method to obtain the visibility level of each unit. Thus, even when the density of traffic and meteorological stations and video surveillance points is insufficient, it can achieve accurate identification of dense fog levels on highways, providing a reference for highway safety operation scheduling and management. Example 2:

[0035] like Figure 2 As shown, the present invention provides a highway fog classification identification device, specifically comprising: Data acquisition module: used to acquire highway network information; Unitization module: used to unitize the highway network information to obtain the unitized highway network information; wherein, the unitized highway network information includes multiple units; The first calculation module is used to obtain information on traffic and meteorological stations along the highway and video surveillance points based on the highway network information; to obtain the visibility level of each traffic and meteorological station based on the information on each traffic and meteorological station; to obtain images of each video surveillance point based on the information on each video surveillance point; to input the images of the video surveillance points into a pre-built and trained hybrid model; and to output the visibility level of each video surveillance point. The second calculation module is used to obtain the visibility level of each unit based on the unitized highway network information, the visibility level of each traffic and meteorological station, and the visibility level of each video surveillance point using the optimal interpolation method. The identification module is used to acquire a dataset based on the visibility level of each unit; to fill the dataset according to a preset filling scheme, thereby acquiring a highway fog classification warning map and completing the highway fog classification identification work. Example 3:

[0036] This invention also provides a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method described in Embodiment 1. Example 4:

[0037] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0038] Since the storage medium provided in this embodiment of the invention can execute the method provided in Embodiment 1 of the invention, it has the corresponding functional modules and beneficial effects for executing the method.

[0039] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0040] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, 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.

[0041] These computer program instructions 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.

[0042] These computer program instructions 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.

[0043] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for classifying and identifying dense fog on highways, characterized in that, Includes the following steps: Obtain highway network information; The highway network information is modularized to obtain modularized highway network information; wherein the modularized highway network information includes multiple units; Based on the highway network information, obtain information on traffic and meteorological stations along the highway and video surveillance points; Based on the information from each of the aforementioned traffic and meteorological stations, obtain the visibility level for each traffic and meteorological station; Based on the information of each video monitoring point, obtain images of each video monitoring point, input the images of the video monitoring points into a pre-built and trained hybrid model, and output the visibility level of each video monitoring point. Based on the unitized highway network information, the visibility levels of various traffic and meteorological stations, and the visibility levels of various video surveillance points, the visibility level of each unit is obtained using the optimal interpolation method. Based on the visibility level of each unit, obtain the dataset; Based on a preset filling scheme, the dataset is filled to obtain a highway fog level warning map and complete the highway fog level identification work. The highway network information includes the latitude and longitude of the highway network, as well as the starting and ending points of the highway network. The mileage of the highway network is obtained based on the starting and ending points of the highway network. Specifically, the step of unitizing the highway network information to obtain the unitized highway network information includes: Based on the starting and ending points of the highway network, the highway network information is modularized using ArcGIS software at preset intervals to obtain multiple units. The hybrid model is a hybrid model of convolutional neural network and transformer; Specifically, obtaining the visibility level of each cell using the optimal interpolation method includes: Obtain real-time visibility data within a preset grid point; Based on the visibility data within the preset grid points, obtain the visibility grid point values; Based on the unitized highway network information, highway traffic and meteorological station information, and video surveillance point information, the visibility grid value is interpolated into each unit, each traffic and meteorological station, and each video surveillance point using bilinear interpolation. Let the initial estimated values ​​of visibility grid points inserted into each unit, each traffic and weather station, and each video surveillance point be classified as follows: ; Determine whether there is a traffic weather station or video surveillance point within the unit. If so, use the visibility level of the traffic weather station or the visibility level of the video surveillance point as the visibility level of the unit. If it does not exist, calculate the visibility level of the unit using the following formula: ； ； In the formula: For the output result; This is a round-down operation; The visibility level for this unit; W is the preset weight; This refers to the visibility level of the nearest traffic weather station or video surveillance point in the direction of the starting point of the highway network to this unit. This refers to the visibility level of the nearest traffic weather station or video surveillance point to the end of the highway network. The The acquisition specifically includes: Based on the highway network information, obtain the location information of each unit; Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the starting point of the highway network; Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point; The The acquisition specifically includes: Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the end of the highway network. Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point.

2. The method for classifying and identifying dense fog on highways according to claim 1, characterized in that, The step of obtaining the visibility level of each traffic weather station based on the information from each traffic weather station specifically includes: Based on the information from each of the aforementioned traffic and weather stations, obtain the visibility of each traffic and weather station; Based on the visibility classification standard, the visibility level of each traffic weather station is obtained according to the visibility of each traffic weather station.

3. The method for classifying and identifying dense fog on highways according to claim 1, characterized in that, The step of obtaining the dataset based on the visibility level of each unit specifically includes: Based on the unitized highway network information and the starting and ending points of the highway network, the visibility levels of each unit are arranged sequentially to obtain the dataset. The process of filling the dataset based on a preset filling scheme specifically includes: The preset filling scheme includes a variety of colors that correspond one-to-one with the visibility level; Based on the visibility level of each unit, the unit is filled with the color corresponding to the visibility level.

4. A highway fog classification and identification device, characterized in that, Specifically, it includes: Data acquisition module: used to acquire highway network information; Unitization module: used to unitize the highway network information to obtain the unitized highway network information; wherein, the unitized highway network information includes multiple units; The first calculation module is used to obtain information on traffic and meteorological stations along the highway and video surveillance points based on the highway network information; to obtain the visibility level of each traffic and meteorological station based on the information on each traffic and meteorological station; to obtain images of each video surveillance point based on the information on each video surveillance point; to input the images of the video surveillance points into a pre-built and trained hybrid model; and to output the visibility level of each video surveillance point. The second calculation module is used to obtain the visibility level of each unit based on the unitized highway network information, the visibility level of each traffic and meteorological station, and the visibility level of each video surveillance point using the optimal interpolation method. Identification module: used to obtain a dataset based on the visibility level of each unit; fill the dataset based on a preset filling scheme to obtain a highway fog classification warning map and complete the highway fog classification identification work; The highway network information includes the latitude and longitude of the highway network, as well as the starting and ending points of the highway network. The mileage of the highway network is obtained based on the starting and ending points of the highway network. Specifically, the step of unitizing the highway network information to obtain the unitized highway network information includes: Based on the starting and ending points of the highway network, the highway network information is modularized using ArcGIS software at preset intervals to obtain multiple units. The hybrid model is a hybrid model of convolutional neural network and transformer; Specifically, obtaining the visibility level of each cell using the optimal interpolation method includes: Obtain real-time visibility data within a preset grid point; Based on the visibility data within the preset grid points, obtain the visibility grid point values; Based on the unitized highway network information, highway traffic and meteorological station information, and video surveillance point information, the visibility grid value is interpolated into each unit, each traffic and meteorological station, and each video surveillance point using bilinear interpolation. Let the initial estimated values ​​of visibility grid points inserted into each unit, each traffic and weather station, and each video surveillance point be classified as follows: ; Determine whether there is a traffic weather station or video surveillance point within the unit. If so, use the visibility level of the traffic weather station or the visibility level of the video surveillance point as the visibility level of the unit. If it does not exist, calculate the visibility level of the unit using the following formula: ； ； In the formula: For the output result; This is a round-down operation; The visibility level for this unit; W is the preset weight; This refers to the visibility level of the nearest traffic weather station or video surveillance point in the direction of the starting point of the highway network to this unit. This refers to the visibility level of the nearest traffic weather station or video surveillance point to the end of the highway network. The The acquisition specifically includes: Based on the highway network information, obtain the location information of each unit; Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the starting point of the highway network; Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point; The The acquisition specifically includes: Based on the location information of each unit, the traffic and meteorological station information along the highway, and the video surveillance point information, obtain the traffic and meteorological station information or video surveillance point information that is closest to this unit in the direction of the end of the highway network. Based on information from traffic weather stations or video surveillance points, obtain the visibility level of the corresponding traffic weather station or the visibility level of the corresponding video surveillance point.

5. A terminal, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 3.

Images