Traffic volume evaluation method and system and electronic equipment

Abstract

The invention relates to the technical field of data processing, in particular to a traffic volume evaluation method and system and electronic equipment, and the method comprises the steps: obtaining the real-time video data of a vehicle passing through a target monitoring point; identifying the vehicle types of the vehicles in the real-time video data based on an identification model, and recording the driving speed of each vehicle and the traffic volume of passing through the target monitoring point; and performing time segmentation on the traffic data of the target monitoring point according to the identified vehicle type, the driving speed and the traffic volume of the vehicle. Therefore, the driving speed of each vehicle and the traffic volume passing through the target monitoring point can be recorded, and time segmentation is performed on the traffic data of the target monitoring point according to the vehicle type of the vehicle, the driving speed of the vehicle and the traffic volume. Therefore, the traffic data of different time periods are subdivided and counted, accurate flow data of each time period is provided for a traffic management department, and the refined management requirement is met.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method, a system and an electronic device for evaluating traffic volume. Background Art

[0002] In the field of traffic volume monitoring and statistics, accurate and detailed data plays a crucial role in traffic management decisions. However, there are many deficiencies in current technical means. The traditional geomagnetic / coil detection method, which was once widely used in traffic volume monitoring, exposes significant limitations. It lacks the ability to identify vehicle types accurately, making it difficult to precisely distinguish different vehicle models, resulting in rough traffic volume data that cannot reflect the actual vehicle composition on the road, and further leading to inaccurate data. Such inaccurate data is prone to cause decision-making deviations in applications such as traffic planning and congestion control, affecting the pertinence and effectiveness of traffic management measures.

[0003] Common video recognition systems have made progress in traffic volume statistics and can obtain the total traffic flow information, but there are obvious shortcomings in the detailed statistics in the time dimension. Traffic flow shows significant variation patterns at different times. For example, the traffic characteristics during peak hours and off-peak hours are very different. However, existing video recognition systems lack detailed statistics for different time periods and cannot provide accurate traffic flow data for each time period to traffic management departments, making it difficult to meet the current requirements of traffic fine management. Summary of the Invention

[0004] In order to solve the technical problem of poor detailed statistics of traffic flow in the time dimension, the purpose of the present invention is to provide a method, a system and an electronic device for evaluating traffic volume, and the specific technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a method for evaluating traffic volume, including: obtaining real-time video data of vehicles passing through a target monitoring point; identifying the vehicle models of the vehicles in the real-time video data based on an identification model, and recording the driving speeds of each vehicle and the traffic volume passing through the target monitoring point; segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, vehicle driving speeds and traffic volume.

[0006] Optionally, after segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, vehicle driving speeds and traffic volume, the method further includes: converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on a dynamic coefficient table.

[0007] Optionally, converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on the dynamic coefficient table includes: determining the current network status; when the network status is at the first level, obtaining the latest updated dynamic coefficient table from the cloud, looking up the first conversion coefficient corresponding to the vehicle model of the vehicle in the latest updated dynamic coefficient table, and converting the vehicles passing through the target monitoring point into standard vehicle equivalents using the first conversion coefficient.

[0008] Optionally, when the network status is at the second level, look up the second conversion coefficient corresponding to the vehicle model of the vehicle in the locally stored highway engineering standard coefficient table, and convert the vehicles passing through the target monitoring point into standard vehicle equivalents using the second conversion coefficient.

[0009] Optionally, time-segmenting the traffic data of the target monitoring point according to the identified vehicle model, vehicle driving speed, and traffic volume includes: triggering time segmentation when the same type of vehicle exceeds the first threshold within a predetermined time period and the traffic volume exceeds the second threshold within the predetermined time period; or triggering time segmentation when the change rate of the traffic volume exceeds the third threshold within the predetermined time period; or triggering time segmentation when the average value of the vehicle driving speed exceeds the fourth threshold within the predetermined time period.

[0010] Optionally, after time-segmenting the traffic data of the target monitoring point according to the identified vehicle model, vehicle driving speed, and traffic volume, the method further includes: time-segmenting the traffic data of the target monitoring point based on a predefined time period, and the predefined time period includes a user-defined time period and a legal holiday time period.

[0011] Optionally, after time-segmenting the traffic data of the target monitoring point according to the identified vehicle model, vehicle driving speed, and traffic volume, the method further includes: storing the time-segmented traffic data in the cloud; receiving a user's viewing request for the traffic data of the target time period; and in response to the viewing request, extracting the traffic data of the target time period from the cloud.

[0012] Optionally, identifying the vehicle model in the real-time video data based on the identification model includes: frame-dividing the real-time video data to obtain multiple frames of images, performing defogging and light compensation on each frame of image to obtain each frame of target image; performing differential calculation on consecutive multiple frames of target images to mark the moving area, and the moving area includes vehicles; and identifying the vehicle model in the moving area based on the pre-trained lightweight YOLOv7 model.

[0013] In a second aspect, an embodiment of the present invention provides a traffic volume evaluation system, including: a camera device and an edge computing node, where the camera device is connected to the edge computing node; the camera device is used to capture real-time images of vehicles passing through a target monitoring point; the edge computing node is used to obtain real-time video data of the vehicles passing through the target monitoring point; identify the vehicle models in the real-time video data based on an identification model, and record the driving speeds of each vehicle and the traffic volume passing through the target monitoring point; and perform time segmentation on the traffic data of the target monitoring point according to the identified vehicle models, driving speeds of the vehicles, and traffic volume.

[0014] In a third aspect, an embodiment of the present invention provides an electronic device, including: a processor and a memory; wherein, the memory is used to store a computer program that can run on the processor; the processor is used to execute the program stored in the memory to implement the steps of the traffic volume evaluation method as mentioned in the first aspect.

[0015] The present invention has the following beneficial effects: First, obtain real-time video data of vehicles passing through a target monitoring point; then identify the vehicle models in the real-time video data based on an identification model, and record the driving speeds of each vehicle and the traffic volume passing through the target monitoring point; finally, perform time segmentation on the traffic data of the target monitoring point according to the identified vehicle models, driving speeds of the vehicles, and traffic volume.

[0016] In this way, the embodiment of the present invention can record videos of vehicles passing through a target monitoring point, and identify and distinguish the vehicle models in the videos through an identification model. It can record the driving speeds of each vehicle and the traffic volume passing through the target monitoring point, and perform time segmentation on the traffic data of the target monitoring point according to the vehicle models, driving speeds of the vehicles, and traffic volume. Thereby, the traffic data in different time periods can be subdivided and statistically analyzed, providing accurate traffic flow data for each time period for the traffic management department and meeting the refined management requirements. Description of the Drawings

[0017] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required to be used in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of a traffic volume evaluation method provided by an embodiment of the present invention;

[0019] Figure 2 It is a schematic structural diagram of a traffic volume evaluation system provided by an embodiment of the present invention

[0020] Figure 3 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. Specific implementation manners

[0021] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and effects of an evaluation method, system and electronic device for traffic volume proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0023] The following specifically describes the specific solution of an evaluation method for traffic volume provided by the present invention with reference to the accompanying drawings.

[0024] Embodiment 1:

[0025] Please refer to Figure 1 , which shows a flowchart of an evaluation method for traffic volume provided by an embodiment of the present invention. The execution subject of this method can be an edge computing node. The evaluation method for traffic volume includes:

[0026] S101, obtaining real-time video data of vehicles passing through a target monitoring point.

[0027] Specifically, in the embodiment of the present invention, a high-definition camera is installed directly above or obliquely in front of a selected traffic intersection or section to ensure that the viewing angle of the camera covers all vehicle driving directions and clear license plates are captured. For example, the high-definition camera can have a resolution of 1080P / 4K and cover a wide-angle or close-up area. After shooting the video of the target monitoring point through the high-definition camera, the video can be compressed by H.264 / H.265 encoding and transmitted to the edge computing node based on the RTSP / RTMP streaming media protocol. Among them, using H.264 / H.265 encoding to compress the video can reduce bandwidth occupancy. The subsequent steps of reasoning are performed by the edge computing node, thereby reducing latency and improving real-time performance.

[0028] Furthermore, the edge computing node in the embodiment of the present invention has a built-in high-performance graphics processing unit (GPU) to support the rapid reasoning of deep learning models and improve data processing speed. It is equipped with a fifth-generation mobile communication technology (5th Generation Mobile Communication Technology, 5G) data transmission module to ensure fast and stable data transmission to the cloud. It also has a local storage server for temporary storage of key data to ensure data integrity in the case of network instability.

[0029] Furthermore, the real-time video data includes but is not limited to key information such as the vehicle's license plate number, color, brand, model, vehicle speed, abnormal behavior (such as driving against traffic, illegal lane change), etc. Vehicle models include but are not limited to cars, passenger cars, small trucks, medium-sized buses, medium-sized trucks, large trucks, trailers, articulated trains, tractors, new energy vehicles, etc.

[0030] S102, identifying the type of vehicles in the real-time video data based on the recognition model, and recording the driving speed of each vehicle and the traffic volume passing through the target monitoring point.

[0031] Specifically, the recognition model in the embodiment of the present invention can be a pre-trained lightweight YOLOv7 model. The lightweight YOLOv7 model is improved on the basis of the original YOLOv7 to reduce the amount of calculation and parameter of the model, while maintaining the detection accuracy as much as possible, so that it is more suitable for running on resource-constrained devices. In the embodiment of the present invention, the lightweight YOLOv7 model is configured in the edge computing node, which can reduce the occupation of resources in the edge computing node by the recognition model and optimize the resource utilization in the edge computing node. It is worth noting that the lightweight process of the YOLOv7 model can refer to the known technology, and the embodiment of the present invention will not be repeated here.

[0032] Furthermore, the training process of the lightweight YOLOv7 model in the embodiments of the present invention is as follows: First, collect image data containing different vehicle models, which can be obtained from the network, surveillance cameras, or professional dataset websites. Ensure that the dataset contains images of various vehicle models, different angles, and lighting conditions. Then use annotation tools (such as LabelImg) to annotate the collected images, mark the bounding boxes of each vehicle, and assign the corresponding vehicle model category to each bounding box. The annotation files are usually saved in the YOLO format, that is, each annotation file is a text file, and each line represents a target, with the format "category number center point x coordinate center point y coordinate width height", and all coordinates and dimensions are proportional values relative to the image width and height. Secondly, divide the annotated dataset into a training set and a test set, generally in a ratio of 7:3. Ensure that each set contains samples of various vehicle models to ensure the generalization ability of the model. Then select a suitable pre-trained model from the official YOLOv7 repository as the base model, such as yolov7-tiny.pt, which has fewer parameters and is easier to lightweight. Lightweight the base model, such as pruning, quantization, and using lightweight modules to lightweight the base model. Input the samples in the training set into the lightweight YOLOv7 model for iterative training to obtain the trained lightweight YOLOv7 model. After training, use the test set to evaluate the trained lightweight YOLOv7 model, calculate metrics such as the accuracy, recall rate, and mean average precision of the trained lightweight YOLOv7 model to evaluate the performance of the trained lightweight YOLOv7 model, and optimize and adjust the trained lightweight YOLOv7 model based on the evaluation results, such as trying to adjust training parameters (such as learning rate, batch size), increasing training data, improving data augmentation methods, etc., to improve the performance of the model.

[0033] Furthermore, after the lightweight YOLOv7 model is trained, deploy the pre-trained lightweight YOLOv7 model to the edge computing node for application.

[0034] Furthermore, as an optional embodiment of the present invention, the recognition of the vehicle model in the real-time video data based on the recognition model includes: dividing the real-time video data into frames to obtain multiple frames of images, performing defogging and light compensation on each frame of image to obtain each frame of target image; performing differential calculation on consecutive multiple frames of target images to mark the moving areas, and the moving areas include vehicles; recognizing the vehicle models in the moving areas based on the pre-trained lightweight YOLOv7 model.

[0035] Specifically, in the embodiments of the present invention, frame operations are performed on real-time video data at a fixed frame rate through video processing libraries such as OpenCV. For each frame of the obtained image, atmospheric scattering causes fog in the image, reducing the clarity and recognition rate of the image. The embodiments of the present invention use the dark channel prior algorithm to remove the fog, thereby improving the clarity and recognition rate of the image. Further, uneven illumination will affect the feature extraction of vehicles in the image. The embodiments of the present invention use the histogram equalization algorithm to perform illumination compensation on the de-fogged image, thereby redistributing the pixel values of the image, enhancing the contrast of the image, and highlighting the detailed features of the vehicle.

[0036] Further, after de-fogging and illumination compensation are performed on each frame of the image, a target image is obtained. The embodiments of the present invention use the method of continuous multi-frame image difference to identify the moving regions in the video. Specifically, the pixel differences between adjacent frames of the target image are calculated, and the regions with larger differences are marked as moving regions. Therefore, the embodiments of the present invention use the frame difference method to exclude the interference of stationary vehicles, improve the accuracy and reliability of the data, and further improve the recognition accuracy of the lightweight YOLOv7 model.

[0037] Further, the embodiments of the present invention input the target image with the marked moving regions into a pre-trained lightweight YOLOv7 model to identify the vehicle models.

[0038] Further, the embodiments of the present invention also record information such as the driving speeds of each vehicle and the traffic volume passing through the target monitoring point for subsequent step analysis.

[0039] S103, perform time segmentation on the traffic data of the target monitoring point according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume.

[0040] Specifically, the embodiments of the present invention can perform time segmentation on the traffic data of the target monitoring point based on information such as the vehicle models, driving speeds, and traffic roads, so as to perform sub-divided storage on the traffic data. Among them, the embodiments of the present invention can use various rules to perform time segmentation on the traffic data of the target monitoring point.

[0041] Further, as an optional embodiment of the present invention, performing time segmentation on the traffic data of the target monitoring point according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume includes: triggering time segmentation when vehicles of the same type exceed a first threshold within a predetermined time period and the traffic volume exceeds a second threshold within the predetermined time period; or triggering time segmentation when the change rate of the traffic volume exceeds a third threshold within the predetermined time period; or triggering time segmentation when the average value of the driving speeds of the vehicles exceeds a fourth threshold within the predetermined time period.

[0042] Specifically, the predetermined time period, the first threshold, the second threshold, the third threshold, and the fourth threshold can be set according to the actual situation. In the embodiments of the present invention, the first threshold is the maximum allowable number of vehicles of the same type that appear within the predetermined time period. For example, on the roads around a certain commercial area, in order to prevent a certain type of delivery truck from being overly concentrated and affecting traffic flow, the upper limit of the number of such trucks passing through per hour is set at 50. The second threshold is the maximum allowable value of the traffic volume within the predetermined time period. For example, for a main road in a certain city, according to its designed traffic capacity, the upper limit of the total traffic volume per hour is set at 2000. The third threshold is the maximum allowable change rate of the traffic volume within the predetermined time period. Assuming a 15-minute statistical cycle, it is stipulated that when the change rate of the traffic volume exceeds 30%, the corresponding mechanism is triggered. The fourth threshold is the upper or lower limit of the average vehicle speed within the predetermined time period. For example, on a highway, when the average vehicle speed exceeds 120 km / h, or in a congested section, when the average speed is lower than 10 km / h, relevant operations are triggered.

[0043] Furthermore, within a predetermined one-hour period, the number of a certain type of vehicle exceeds the first threshold. For example, the number of delivery trucks reaches 60, and at this time, the total traffic volume on this section of the road also exceeds the second threshold, such as reaching 2200. This situation indicates that within the current time period on this section of the road, a particular type of vehicle is overly concentrated, and the overall traffic flow is too large, exceeding the normal carrying capacity of the road, and is extremely likely to cause traffic congestion. At this time, the system will trigger a time segmentation mechanism, mark the current time period as a traffic peak period, and initiate corresponding traffic control strategies, such as adjusting the signal timing to increase the passing time of the main road.

[0044] Furthermore, in the embodiments of the present invention, with a 15-minute statistical cycle, the system calculates the change rate of the traffic volume in real time. When within a certain cycle, the traffic volume rapidly increases from 1000 vehicles to 1400 vehicles, and the change rate reaches 40%, exceeding the third threshold of 30%. This means that the traffic flow has changed sharply in a short period of time, which may be caused by sudden traffic accidents, the dispersal of large-scale events, etc. To respond to this change in a timely manner, the system will trigger time segmentation, divide this time period into a special traffic period, and send a warning message to the traffic management department so that timely dredging measures can be taken.

[0045] Furthermore, in the embodiments of the present invention, within one hour, the average vehicle speed reaches 130 km / h, exceeding the fourth threshold of 120 km / h. This indicates that the vehicle speed is too fast and there are significant safety hazards. The system will trigger time segmentation, mark this time period as an overspeed risk period, and issue a warning message to drivers through a variable message sign to remind them to slow down. On the contrary, in a congested section of the city, if the average vehicle speed is lower than 10 km / h, the system will also trigger time segmentation, divide this time period into a congested period, and recommend that drivers choose other routes to travel.

[0046] Furthermore, in the embodiments of the present invention, traffic data of a target monitoring point can also be time-segmented based on a predefined time period, and the predefined time period includes a user-defined time period and a legal holiday time period.

[0047] Specifically, to meet the personalized needs of different traffic management scenarios, the embodiments of the present invention provide a function for setting a user-defined time period. Staff members of the traffic management department can set different time periods according to the historical traffic data of a specific road section and actual management requirements. For example, for the roads around schools, considering the time of students going to and from school, the periods from 7:00 to 8:30 and from 16:00 to 18:00 on weekdays can be set as key control periods. In addition, for large-scale events such as concerts and sports events, the temporary traffic control time periods can be flexibly set according to the start and end times of the events.

[0048] Furthermore, the embodiments of the present invention can also pre-embed a schedule of legal holidays, covering New Year's Day, Spring Festival, Tomb-Sweeping Day, Labor Day, Dragon Boat Festival, Mid-Autumn Festival, National Day, etc. Before the holidays come, the corresponding time periods will be automatically identified, and combined with the traffic data of historical holidays, the changing trend of traffic flow will be predicted. For example, during the Spring Festival, the peak periods of returning home and coming back to the city usually occur in specific time periods before and after the festival, and the system will mark these periods as key periods accordingly.

[0049] Furthermore, the embodiments of the present invention perform time-segmentation processing on traffic data according to a predefined time period. Within each time period, key indicators such as traffic volume, average vehicle speed, and vehicle type distribution will be statistically analyzed in detail, and corresponding statistical reports will be generated. Through data mining and machine learning algorithms, the changing rules of traffic data in different time periods are analyzed, and a traffic flow prediction model is established. For example, a long short-term memory (LSTM) model is trained using historical data to predict the traffic flow in a specific future time period, providing support for traffic management decisions.

[0050] The embodiments of the present invention can record videos of vehicles passing through a target monitoring point, and identify and distinguish the vehicle types in the videos through an identification model. The driving speed of each vehicle and the traffic volume passing through the target monitoring point can be recorded, and the traffic data of the target monitoring point is time-segmented according to the vehicle type, driving speed, and traffic volume of the vehicle. Thus, the traffic data in different time periods is statistically analyzed in detail, providing accurate traffic volume data for each time period for the traffic management department and meeting the refined management requirements.

[0051] Furthermore, as an optional embodiment of the present invention, after time-segmenting the traffic data of the target monitoring point according to the identified vehicle type, driving speed, and traffic volume of the vehicle, the method further includes:

[0052] Store the traffic data after time segmentation in the cloud; receive the user's request to view the traffic data for the target time period; in response to the view request, extract the traffic data for the target time period from the cloud.

[0053] Specifically, in the embodiments of the present invention, with the help of a high-speed and stable network connection, the traffic data processed by time segmentation is transmitted to the cloud storage platform. To ensure the security and integrity of data transmission, the SSL / TLS encryption protocol is used to encrypt the data during the transmission process to prevent data leakage and tampering. At the same time, to ensure the stability of data transmission and cope with network fluctuations, the system sets up a breakpoint resumption mechanism. When the network connection is interrupted, the transmission progress is recorded. After the network resumes, the transmission continues from the breakpoint to avoid duplicate data transmission and data loss.

[0054] Furthermore, in the cloud, a distributed file system and object storage technology are used to build a highly reliable and scalable storage architecture. With the help of mature cloud storage services such as AWS S3 and Alibaba Cloud OSS, the traffic data is classified and stored according to dimensions such as time segmentation and data type. Taking time as the main line, a hierarchical directory structure is established according to year, month, day, and specific time periods to facilitate data management and retrieval. For example, the traffic data for the morning rush hour on October 1, 2024 is stored in the directory " / 2024 / 10 / 01 / morning rush hour / ". And, to improve the reliability of data storage, the data is backed up to multiple geographical locations through cross-region multi-copy storage to reduce the risk of data loss caused by natural disasters, hardware failures, etc.

[0055] Furthermore, users can submit requests to view traffic data for the target time period through various channels such as the official website of the traffic management department and mobile APPs. The system sets up a unified request access interface for each channel to ensure that requests can be received in a timely and accurate manner. When the user enters the time period of the traffic data they expect to view on the mobile APP and clicks the submit button, the APP encapsulates the request data into an HTTP / HTTPS request and sends it to the backend server. After receiving the request, the backend server first parses the request to extract key information such as the target time period and user identity. Through the identity verification mechanism, the legitimacy of the user's identity is verified. Only legitimate users can access the corresponding traffic data. At the same time, the validity of the target time period parameter is verified to ensure that the requested time period meets the format and range specified by the system. For example, the system only supports querying traffic data within the past 5 years. If the user requests to view data for a time period beyond this range, the system will return an error prompt message.

[0056] Furthermore, after verifying the legitimacy of the request, retrieve and extract the corresponding traffic data from the cloud storage platform according to the target time period information. By using the query interface provided by the cloud storage service and combining with the previously established directory structure, quickly locate the target data file. For example, when the user requests to view the traffic data during the morning rush hour on October 1, 2024, the system accesses the directory " / 2024 / 10 / 01 / morning rush hour / " to obtain the corresponding traffic data file.

[0057] Furthermore, to improve the user experience, the system preprocesses the extracted traffic data. For example, according to the user's terminal type and network conditions, operations such as data compression and format conversion are performed on the data. For mobile APP users, the data is converted into a format suitable for mobile display and appropriately compressed to reduce the data transmission volume and speed up the data loading speed. Finally, the processed traffic data is returned to the user through an HTTP / HTTPS response, and the user can view the traffic data for the target time period on the front-end interface, such as traffic flow statistics charts and vehicle speed change curves.

[0058] Furthermore, the key link of standard vehicle conversion has long relied on manual post-processing and cannot output the conversion result in real time, seriously affecting the timeliness and availability of the data. Therefore, as an optional embodiment of the present invention, after time-segmenting the traffic data of the target monitoring point according to the vehicle type, driving speed, and traffic volume of the identified vehicle, the method further includes: converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on the dynamic coefficient table.

[0059] Specifically, in the embodiment of the present invention, the edge computing node converts the vehicles passing through the target monitoring point into standard vehicle equivalents based on the dynamic coefficient table, so as to be able to perform standard vehicle conversion on the vehicles passing through the target monitoring point in real time and improve the timeliness.

[0060] Among them, the dynamic coefficient table in the embodiments of the present invention can be constructed and updated in real time in the cloud, or it can be a highway engineering standard coefficient table. The process of constructing the dynamic coefficient table in the cloud in real time is as follows: Collect a large amount of operation data of different types of vehicles in actual traffic scenarios, including vehicle speed, road occupancy rate, acceleration performance, braking performance, etc. At the same time, with the help of traffic simulation software, simulate the driving characteristics of various vehicles under different traffic conditions to obtain supplementary data. For example, it is found that during the morning and evening rush hours, the impact of buses on road capacity is different from that during the off-peak period due to frequent stops at stations. Use mathematical statistical methods to analyze the collected data. Taking small passenger cars as the standard vehicles, calculate the equivalent coefficients corresponding to various vehicles by comparing the differences between other vehicles and small passenger cars in terms of occupying road space and affecting traffic flow operation. For example, large trucks have larger body sizes and relatively poor acceleration and braking performance, so their equivalent coefficients in the traffic flow will be higher than those of small passenger cars. Considering the changes in traffic conditions, road conditions, and vehicle technologies, the dynamic coefficient table needs to be updated regularly. By real-time monitoring traffic data and combining new research results, timely adjust the equivalent coefficients of various vehicles. For example, with the popularization of new energy vehicles, their driving characteristics are different from those of traditional fuel vehicles, and it is necessary to re-evaluate and update the equivalent coefficients of new energy vehicles.

[0061] Furthermore, when a vehicle passes through the target monitoring point, the acquisition device obtains the relevant information of the vehicle and matches it with the vehicle types in the dynamic coefficient table. For example, identify a vehicle as a large bus through a camera and find the corresponding equivalent coefficient in the dynamic coefficient table. According to the matched equivalent coefficient, convert the actual number of vehicles into the equivalent of standard vehicles. The calculation formula is: equivalent of standard vehicles = actual number of vehicles × corresponding equivalent coefficient. For example, if 10 large buses pass through the monitoring point during a certain period and the equivalent coefficient of large buses is 2.5, then the equivalent of standard vehicles after conversion of these 10 large buses is 10 × 2.5 = 25. By converting different types of vehicles into the equivalent of standard vehicles, the traffic flow of the road can be evaluated more accurately. The traffic management department can judge whether the road is in a congested state based on the equivalent data of standard vehicles, providing a basis for traffic guidance and signal timing optimization. For example, when the equivalent of standard vehicles on a certain road section exceeds its designed traffic capacity, the traffic management department can take diversion measures in time to relieve traffic congestion.

[0062] Furthermore, changes in network status have a significant impact on traffic data processing and applications. When the network status is at the first level, it means that the network has the characteristics of high speed, stability, and low latency, providing a strong guarantee for timely obtaining the latest data from the cloud. Therefore, converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on the dynamic coefficient table includes: determining the current network status; when the network status is at the first level, obtaining the latest updated dynamic coefficient table from the cloud, finding the first conversion coefficient corresponding to the vehicle model from the latest updated dynamic coefficient table, and using the first conversion coefficient to convert the vehicles passing through the target monitoring point into standard vehicle equivalents.

[0063] Specifically, the first level refers to a good network state. The edge computing node continuously monitors key indicators such as network connectivity, bandwidth, latency, and packet loss rate in real time through a built-in network monitoring module. By setting clear network state classification criteria, when the network bandwidth reaches 100 Mbps or more, the network latency is stable within 10 ms, and the packet loss rate is less than 0.1%, the network state is determined to be in the first level. Once the network meets this standard, it triggers the dynamic coefficient table update mechanism. When standard quantity conversion is required, an HTTP / HTTPS request carrying authentication information and request parameters is sent to the cloud server. The request contains the version identifier and update timestamp of the required dynamic coefficient table to ensure that the latest version is obtained. After receiving the request, the cloud server verifies and parses the request to confirm the legitimacy of the request and the permissions of the requester. The cloud server retrieves and extracts the latest dynamic coefficient table, and uses an efficient data compression algorithm such as gzip to compress the data to reduce the amount of transmitted data and improve the transmission efficiency. During the transmission process, to ensure the integrity and accuracy of the data, a hash algorithm such as MD5 or SHA-1 is used to sign the data. The receiving end verifies the signature to ensure that the data has not been tampered with during transmission. After receiving the data, the receiving end performs decompression and verification operations. After confirmation, the dynamic coefficient table is stored in the local cache. When the vehicle passes through the target monitoring point, the vehicle model information is collected. The vehicle model can be accurately obtained through a high-definition camera combined with image recognition technology or by using a vehicle electronic tag recognition system. The collected vehicle model information is matched with the latest dynamic coefficient table stored in the local cache, using multiple dimensions such as vehicle model name, vehicle size, and usage as the matching basis. For example, when a 12-meter-long city bus is identified, the corresponding first conversion coefficient is accurately found in the dynamic coefficient table. Based on the obtained first conversion coefficient, the vehicle passing through the target monitoring point is converted into standard vehicle equivalents using a standard calculation formula. The calculation formula is: Standard vehicle equivalents = Number of passing vehicles × First conversion coefficient. For example, during a certain period, 20 vehicles identified as 12-meter-long city buses pass through the monitoring point. The first conversion coefficient of this vehicle model is found to be 3.0 from the dynamic coefficient table. Then the standard vehicle equivalents of these buses after conversion are 20 × 3.0 = 60. The calculated standard vehicle equivalent data is transmitted to the traffic flow analysis module in real time for evaluating traffic congestion conditions, analyzing road capacity, and optimizing traffic signal timing. At the same time, the data is also stored in the local database to provide historical data support for subsequent traffic planning, policy formulation, and traffic management decisions.

[0064] Further, as an optional embodiment of the present invention, in the case where the network state is in the second level, the second conversion coefficient corresponding to the vehicle model is found from the highway engineering standard coefficient table stored locally, and the vehicle passing through the target monitoring point is converted into standard vehicle equivalents using the second conversion coefficient.

[0065] Specifically, the second level refers to a relatively poor network state. When it is determined that the network state is at the second level, it means that there are certain limitations in network transmission, such as relatively low bandwidth, large latency, or high packet loss rate. To ensure the continuous and stable progress of the conversion work from vehicle to standard vehicle equivalent, the locally stored highway engineering standard coefficient table will be enabled to achieve real-time analysis and processing of traffic data. Among them, the edge computing node with a network monitoring module sends network probe packets to a preset server regularly to obtain key performance indicators of the network in real time, such as bandwidth, latency, and packet loss rate. According to the pre-set grading criteria, the network state is evaluated. When the network bandwidth is between 20 - 100 Mbps, the latency is in the range of 10 - 50 ms, and the packet loss rate is between 0.1% - 1%, it is determined that the network state is at the second level. Once the network state enters this level, the system will trigger the local data call mechanism to reduce the dependence on cloud data and ensure the continuity of data processing.

[0066] Further, in the edge computing node, the authoritative highway engineering standard coefficient table (such as the JTGB01-2014 coefficient table) will be downloaded and stored in the high-performance storage device of the local server, such as a solid-state drive (SSD), to ensure the rapid reading of data. At the same time, the edge computing node will perform version management on the coefficient table, recording the time and content of each update for easy traceability and management. When the network status is at the second level, the locally cached highway engineering standard coefficient table will be quickly called. To improve data retrieval efficiency, data structures such as hash tables or B+ trees are used to construct an index for the coefficient table to achieve fast vehicle type matching and coefficient lookup. Specifically, the collected vehicle type information is compared with the local highway engineering standard coefficient table. Based on the vehicle type classification standard, multiple dimensions such as the use, number of axles, and length of the vehicle are comprehensively considered. For example, when a vehicle is identified as a three-axle large truck, the system quickly locates the corresponding record in the coefficient table through the index and obtains the second conversion coefficient. According to the obtained second conversion coefficient, vehicle equivalent conversion is performed according to the standard calculation formula. The calculation formula is: standard vehicle equivalent = number of passing vehicles × second conversion coefficient. For example, if 15 three-axle large trucks pass through the monitoring point during a certain period, and the second conversion coefficient of this vehicle type is found to be 2.5 from the local coefficient table, then the standard vehicle equivalent after conversion of these trucks is 15 × 2.5 = 37.5. The calculated standard vehicle equivalent data is transmitted to the traffic flow analysis module in real time for evaluating the traffic load of the current section and judging the congestion status. At the same time, these data will also be stored in the local database to provide historical data support for the traffic management department, helping it to formulate scientific and reasonable traffic plans and management decisions, such as optimizing signal timing and adjusting road traffic rules. In addition, the network monitoring module continuously monitors the network status. Once the network resumes to the first level, the dynamic coefficient table will be automatically updated to ensure the accuracy and timeliness of data processing.

[0067] Embodiment 2:

[0068] Corresponding to the traffic volume evaluation method provided in the above embodiment, based on the same technical concept, the embodiment of the present invention also provides a traffic volume evaluation system, and this traffic volume evaluation system is used to execute the above traffic volume evaluation method. Figure 2 The structural schematic diagram of a traffic volume evaluation system provided by another embodiment of the present invention is as Figure 2As shown in the figure. The traffic volume evaluation system 200 includes: a camera device 201 and an edge computing node 202, and the camera device 201 is connected to the edge computing node 202; the camera device 201 is used for taking real-time pictures of vehicles passing through the target monitoring point; the edge computing node 202 is used for obtaining real-time video data of the vehicles passing through the target monitoring point; identifying the vehicle models in the real-time video data based on an identification model, and recording the driving speeds of the vehicles and the traffic volume passing through the target monitoring point; and segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, driving speeds of the vehicles, and traffic volume.

[0069] It should be noted that the traffic volume evaluation system in the embodiments of the present invention and the above-mentioned traffic volume evaluation method belong to the same inventive concept, and their similarities or similarities and beneficial effects can be referred to each other, and will not be elaborated herein in the embodiments of the present invention. In addition, the application of edge computing technology in the embodiments of the present invention ensures the real-time processing and transmission of data, greatly improving the system response speed. And the data is processed at the edge computing node, ensuring the security and privacy of the data, and at the same time realizing cloud backup to ensure that the data is not lost.

[0070] Embodiment 3:

[0071] Corresponding to the traffic volume evaluation method provided in the above embodiment, based on the same technical concept, the embodiments of the present invention also provide an electronic device, which is used to execute the above traffic volume evaluation method. Figure 3 As shown in the figure, it is a schematic structural diagram of an electronic device provided in another embodiment of the present invention. Figure 3 As shown. The electronic device may vary greatly due to configuration or performance, and may include one or more processors 301 and a memory 302. The memory 302 is used to store a computer program that can run on the processor 301. The processor 301 is used to execute the program stored in the memory 302 to implement each step in the method embodiments above. Figure 1 Among them, the memory 302 can be short-term storage or persistent storage. The application program stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the electronic device.

[0072] Furthermore, the processor 301 can be set to communicate with the memory 302 and execute a series of computer executable instructions in the memory 302 on the electronic device. The electronic device may further include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0073] Specifically, in this embodiment, the electronic device includes a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the bus. The memory is used to store computer programs. The processor is used to execute the programs stored on the memory to implement the above Figure 1 steps in the method embodiments above, and has the beneficial effects of the above method embodiments. To avoid repetition, the embodiments of the present invention will not be described in detail here.

[0074] It should be noted that the traffic volume evaluation system provided by the embodiments of the present invention and the traffic volume evaluation method provided by the embodiments of the present invention are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned traffic volume evaluation method and has the same or similar beneficial effects. The repeated parts will not be described again.

[0075] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0076] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A method for evaluating traffic volume, characterized in that, The method for evaluating traffic volume includes: Obtaining real-time video data of vehicles passing through a target monitoring point; Identifying the vehicle models in the real-time video data based on an identification model, and recording the driving speeds of each vehicle and the traffic volume passing through the target monitoring point; Segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume.

2. The traffic volume evaluation method according to claim 1, characterized in that, After segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume, the method further includes: Converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on a dynamic coefficient table.

3. The traffic volume evaluation method according to claim 2, characterized in that The converting the vehicles passing through the target monitoring point into standard vehicle equivalents based on a dynamic coefficient table includes: Determining the current network state; In the case where the network state is at the first level, obtaining the latest updated dynamic coefficient table from the cloud, searching for a first conversion coefficient corresponding to the vehicle model in the latest updated dynamic coefficient table, and converting the vehicles passing through the target monitoring point into standard vehicle equivalents by using the first conversion coefficient.

4. The traffic volume evaluation method according to claim 3, characterized in that, The method further includes: In the case where the network state is at the second level, searching for a second conversion coefficient corresponding to the vehicle model in the locally stored highway engineering standard coefficient table, and converting the vehicles passing through the target monitoring point into standard vehicle equivalents by using the second conversion coefficient.

5. The traffic volume evaluation method according to any one of claims 1 to 4, characterized in that The segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume includes: Triggering time segmentation in the case where vehicles of the same type exceed a first threshold within a predetermined time period and the traffic volume exceeds a second threshold within the predetermined time period; Or triggering time segmentation in the case where the change rate of the traffic volume exceeds a third threshold within the predetermined time period; Or triggering time segmentation in the case where the average value of the driving speeds of the vehicles exceeds a fourth threshold within the predetermined time period.

6. The traffic volume evaluation method according to claim 1, wherein After segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume, the method further includes: Segmenting the traffic data of the target monitoring point by time based on a predefined time period, where the predefined time period includes a user-defined time period and a legal holiday time period.

7. The traffic volume evaluation method according to claim 1, characterized in that, After segmenting the traffic data of the target monitoring point by time according to the identified vehicle models, the driving speeds of the vehicles, and the traffic volume, the method further includes: Storing the traffic data after time segmentation to the cloud; Receiving a viewing request from a user for the traffic data of a target time period; In response to the viewing request, extracting the traffic data of the target time period from the cloud.

8. The traffic volume evaluation method according to claim 1, characterized in that, The identifying the vehicle models in the real-time video data based on an identification model includes: Framing the real-time video data to obtain multiple frames of images, and performing defogging and light compensation on each frame of image to obtain each frame of target image; Perform differential calculation on consecutive multi-frame target images, mark the moving area, and the moving area includes the vehicle; Identify the vehicle model of the vehicle in the moving area based on the pre-trained lightweight YOLOv7 model.

9. An evaluation system for traffic volume, characterized in that, It includes: A camera device and an edge computing node, and the camera device is connected to the edge computing node; The camera device is used to take real-time pictures of the vehicles passing through the target monitoring point; The edge computing node is used to obtain the real-time video data of the vehicles passing through the target monitoring point; Identify the vehicle models of the vehicles in the real-time video data based on the recognition model, and record the driving speeds of the vehicles and the traffic volume passing through the target monitoring point; Segment the traffic data of the target monitoring point by time according to the recognized vehicle models, the driving speeds of the vehicles, and the traffic volume.

10. An electronic device, characterized in that, It includes: A processor and a memory; wherein, the memory is used to store computer programs that can run on the processor; The processor is used to execute the programs stored on the memory to implement the steps of the traffic volume evaluation method described in any one of claims 1-8.

Images