A management method and system based on highway overloading

Abstract

The application relates to the technical field of expressway overloading and overlength management, in particular to a management method and system based on expressway overloading and overlength, wherein the method comprises the following steps: S1, comparing the front and rear license plate symbols of a vehicle, if the symbols are inconsistent, combining infrared image edge detection to judge whether the vehicle is followed by multiple vehicles or the license plate is wrong, if the symbols are consistent, S2, classifying the vehicle type based on the front picture of the vehicle, and checking whether the wheel shaft detection result matches the vehicle type, if the result matches, S3, if the speed difference is large, calculating the fitting degree of the weight curve and the escape behavior curve, if the fitting degree is high, requiring the vehicle to be weighed again, otherwise, taking the robust average value of the weight curve as the weighing result; if the speed is stable, directly calculating the robust average value as the weighing result; S4, obtaining the load upper limit according to the vehicle type, if the weighing result is not over the limit, the vehicle is allowed to pass. The application has high credibility in multi-layer judgment, and the accuracy of effective weighing data is improved.

Description

Technical Field

[0001] This invention relates to the field of highway overload management technology, specifically a management method and system for highway overload. Background Technology

[0002] Controlling overloaded vehicles at highway entrances is a crucial aspect of ensuring highway safety.

[0003] The existing overload control system has obvious technical defects. First, it relies on a single infrared beam or inductive loop for vehicle separation. When vehicles are too close, it is easy to misidentify multiple vehicles as a single vehicle, resulting in all subsequent detection data being disordered and invalid records being generated. Second, it lacks an effective anti-cheating mechanism and cannot accurately identify behaviors such as drivers altering the vehicle's dynamic state by jumping over or rushing the scale to interfere with the weighing results, resulting in inaccurate weighing data and leaving loopholes for overloaded vehicles to evade supervision. Data matching relies only on simple timing, which has low reliability in complex traffic scenarios. Summary of the Invention

[0004] The purpose of this invention is to provide a management method and system for overloading on highways.

[0005] The technical solution of this invention is as follows: A management method for overloaded and oversized vehicles on highways, comprising: S1. Obtain the license plates from the front and rear of the vehicle to be detected, extract the features and symbols of the license plates from the front and rear, and compare them. If the license plate symbols at the front and rear of the vehicle are different and the infrared light is blocked, an infrared image is acquired and edge detection is performed. If the edge detection is not continuous, the detection is repeated. If the edge detection is continuous, an error message is displayed indicating that the license plate is incorrect. If the license plate symbols at the front and rear of the vehicle are the same, then execute S2; S2. Obtain the front image of the vehicle, classify the model of the vehicle to be detected based on the front image, obtain the wheel axle detection results of the vehicle to be detected, and determine whether they match the model of the vehicle to be detected. If they match, proceed to S3. S3. Obtain the side video and weight curve of the vehicle to be tested. Based on the known length of the weighing platform, obtain the speed change curve of the vehicle to be tested. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, calculate the fitting degree between the weight curve and the escape behavior curve based on the vehicle length. If the fitting degree exceeds the safety threshold, re-weigh the vehicle. If the fitting degree does not exceed the safety threshold, take the average value of the weight curve after removing the maximum and minimum values ​​as the weighing result. If the difference between the maximum and minimum values ​​in the velocity change curve is less than the difference threshold, then the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result. S4. Obtain the maximum load capacity of the vehicle model to be inspected based on the model of the vehicle to be inspected, compare the weighing result with the maximum load capacity, and if the weighing result is less than or equal to the maximum load capacity, then the vehicle is not overloaded and is allowed to pass.

[0006] Based on the vehicle length, the goodness of fit between the weight curve and the escape behavior curve is calculated. The specific operation is as follows: For vehicle models whose length exceeds the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, the weight data sequence is obtained from the weight curve, and the escape curve data sequence corresponding to the difference between the maximum and minimum values ​​in the current speed change curve is retrieved from the escape behavior curve. The minimum matching path between the weight data sequence and the escape curve data sequence is found and mapped to the goodness of fit. For vehicle models with a length less than or equal to the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, features in the weight curve and escape behavior are extracted, including the maximum peak amplitude, fluctuation duration, fluctuation period, and phase difference between the front and rear axle fluctuations. The relative error between each pair of features is calculated, a weight is assigned to each relative error, and the weighted comprehensive error is calculated. The comprehensive error is used as the goodness of fit.

[0007] Finding the minimum matching path between the weight data sequence and the escape curve data sequence involves: standardizing both sequences, calculating the Euclidean distance between any two points in the standardized sequences to form a local distance matrix, and using dynamic programming to create a cumulative cost matrix starting from the top left corner (1,1). The value at each position is equal to the value at the corresponding position in the local distance matrix plus the minimum cumulative cost among its three adjacent positions to the left, below, and to the bottom left. The value at the bottom right corner of the cumulative cost matrix is ​​the minimum global cumulative cost between the two sequences, and the path containing the minimum global cumulative cost is the minimum matching path.

[0008] The standardization process is as follows: the difference between each data point in the sequence and the sequence mean is divided by the sequence standard deviation.

[0009] The symbols described in S1 include characters, letters, and numbers.

[0010] In S2, edge detection coherence is specifically defined as follows: For the image with edge detection, connected components are extracted, and the connected component with the largest area is found. The pixel size of the largest connected region in the image is converted into the actual length and width. If the converted length is greater than or equal to the minimum vehicle length threshold, the width is greater than or equal to the minimum effective lane width threshold, and the aspect ratio is within the range of typical vehicle aspect ratios, then the edge detection is determined to be coherent; otherwise, the edge detection is determined to be discontinuous.

[0011] A management system for overloaded and oversized vehicles on highways, used to implement the aforementioned management method for overloaded and oversized vehicles on highways, includes: The license plate detection module acquires the license plates from the front and rear of the vehicle to be detected, extracts the features and symbols of the license plates from the front and rear, and compares them. If the license plate symbols from the front and rear are different and the infrared light is blocked, an infrared image is acquired and edge detection is performed. If the edge detection is not continuous, the detection is repeated. If the edge detection is continuous, a license plate error is indicated. If the license plate symbols from the front and rear are the same, wheel axle detection is performed. The wheel axle detection module acquires an image of the vehicle's front end, classifies the vehicle model based on the image, obtains the wheel axle detection results, and determines whether the results match the vehicle model. If they match, the module then obtains the weighing results. The weighing result acquisition module acquires the side video and weight curve of the vehicle to be inspected. Based on the known length of the weighing platform, it obtains the speed change curve of the vehicle to be inspected. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, it calculates the fit between the weight curve and the escape behavior curve based on the vehicle length. If the fit exceeds the safety threshold, the vehicle is weighed again. If the fit does not exceed the safety threshold, the average value of the weight curve after removing the maximum and minimum values ​​is taken as the weighing result. If the difference between the maximum and minimum values ​​in the speed change curve is less than the difference threshold, the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result. The passage judgment module obtains the upper limit of the vehicle model based on the model of the vehicle to be detected, compares the weighing result with the upper limit of the load, and if the weighing result is less than or equal to the upper limit of the load, then the vehicle is not overloaded and is allowed to pass.

[0012] It also includes cameras installed at the front parking space of the vehicle, as well as cameras on the sides of the vehicle.

[0013] It also includes an infrared beam obstruction detection sensor, which starts taking pictures when the infrared beam is blocked.

[0014] This invention achieves significant beneficial effects by using multi-layered logical judgment to control overloading: By fusing and comparing the license plate symbols and features at the front and rear of the vehicle, and combining them with infrared image contour analysis, the system effectively distinguishes between single-license plate anomalies and multiple vehicles following each other, ensuring the uniqueness and accuracy of the detected objects from the source.

[0015] This system achieves anti-cheating goals for dynamic weighing by analyzing side-view video to obtain speed changes and calculating the high-precision fit between the weight curve and evasion behavior. It can identify and handle attempts to interfere with weighing. Furthermore, a robust averaging algorithm that removes maximum and minimum values ​​further improves the accuracy of effective weighing data.

[0016] This invention forms a closed-loop verification system from vehicle identity and model verification to wheel and axle matching, and then to weighing and vehicle behavior correlation analysis, ensuring a high degree of consistency of license plate, model, wheel and axle, and weight data, and generating a complete chain of evidence with extremely high credibility.

[0017] This method reduces manual intervention and effectively avoids misjudgments and omissions through precise judgment logic, thereby enhancing overall control efficiency. It not only ensures the passage efficiency of legal vehicles but also resolutely intercepts overloaded vehicles, achieving efficient, precise, and intelligent control at highway entrances. Detailed Implementation

[0018] Example 1 A management method for overloaded and oversized vehicles on highways, comprising: S1. Obtain the license plates of the front and rear of the vehicle to be detected, extract the features and symbols of the license plates of the front and rear, and compare them. If the license plate symbols of the front and rear are different and the infrared light is blocked, then obtain the infrared image and perform edge detection. If the edge detection is not continuous, then re-detect. If the edge detection is continuous, then indicate that the license plate is wrong. If the license plate symbols of the front and rear are the same, then proceed to S2.

[0019] Specifically, the license plate is identified from images taken from the front and rear of the vehicle, and then the features and symbols of the license plate from the front and rear of the vehicle are extracted. The features include, but are not limited to, the location and color of the license plate, and the symbols include text, letters and numbers.

[0020] Compare the license plate features and symbols at the front and rear of the vehicle. If the symbols are different, it could be due to a dirty license plate, a replaced license plate, or two vehicles entering the vehicle. Therefore, further judgment is needed. Obtain an infrared image and perform edge detection. If the edge detection is discontinuous, it means two vehicles have entered the vehicle and need to be detected again. Allow the two vehicles to enter the monitoring area separately in two separate steps. If the edge detection is continuous, it indicates that the license plate is incorrect and there is a possibility of dirt or a replaced license plate. If the license plate symbols at the front and rear of the vehicle are the same, there is no problem as described above, and then proceed to step S2.

[0021] Furthermore, the determination of whether the edge detection is continuous is as follows: For the edge detection image, extract the connected components, find the connected component with the largest area, and convert the pixel size of the largest connected region in the image into the actual length and width. If the converted length is greater than or equal to the minimum vehicle length threshold, the width is greater than or equal to the minimum effective lane width threshold, and the aspect ratio is within the range of typical vehicle aspect ratios, then the edge detection is determined to be continuous; otherwise, the edge detection is determined to be discontinuous.

[0022] By comparing the license plate symbols and features at the front and rear of the vehicle, and combining them with infrared image edge detection, the system effectively distinguishes between license plate anomalies and multiple vehicles following each other, ensuring the uniqueness and accuracy of the detected objects from the source.

[0023] S2. Obtain the front image of the vehicle. Based on the front image, classify the model of the vehicle to be detected, obtain the wheel axle detection results of the vehicle to be detected, and determine whether they match the model of the vehicle to be detected. If they match, proceed to S3.

[0024] Overloading and exceeding weight limits on highways refer to vehicles whose actual cargo weight or geometric dimensions exceed the prescribed limits when driving on highways. Specifically, it can be divided into two categories: oversized and overloaded. Oversized refers to vehicles whose geometric dimensions or axle load exceed the prescribed limits. Oversized vehicles may cause potential damage to road infrastructure, have low driving safety, large turning radius, and strong obstruction of vision.

[0025] This step in the invention mainly involves the detection of exceeding limits, specifically: Extract features from the front image of the vehicle and classify the model of the vehicle to be inspected. After classification, it is necessary to obtain the corresponding limit standard for the model, including width, length, height and number of axles. Obtain the axle detection results, width, length and height from the side video of the vehicle to be inspected and compare them with the limit standard. If the limit standard is exceeded, it is an over-limit vehicle. If the limit standard is not exceeded, continue to execute S3.

[0026] S3. Obtain the side video and weight curve of the vehicle to be tested. Based on the known length of the weighing platform, obtain the speed change curve of the vehicle to be tested. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, calculate the fitting degree between the weight curve and the escape behavior curve based on the vehicle length. If the fitting degree exceeds the safety threshold, re-weigh the vehicle. If the fitting degree does not exceed the safety threshold, take the average value of the weight curve after removing the maximum and minimum values ​​as the weighing result. If the difference between the maximum and minimum values ​​in the speed change curve is less than the difference threshold, calculate the average value of the weight curve after removing the maximum and minimum values ​​as the weighing result.

[0027] This step in the invention mainly involves the detection of overload. The speed of the vehicle to be detected can be obtained from the side video. The speed of the vehicle to be detected is continuously extracted, and a speed-time curve can be plotted. The maximum and minimum values ​​in this curve are obtained, and the difference between the two is calculated to determine whether it exceeds the difference threshold. If it exceeds the difference threshold, there may be a possibility of scale skipping or scale dragging, and further judgment is required.

[0028] The escape curves for vehicles of different lengths each have their own characteristics. Shorter vehicles have better body rigidity and shorter wheelbases, and when performing weight-skipping escape maneuvers, the entire vehicle moves more like a rigid body. Their weight curves exhibit relatively simple and regular fluctuations, mainly characterized by overall, dramatic undulations. Longer vehicles, with multiple axle groups and even articulated structures, experience complex torsional and non-rigid motions during escape maneuvers. Their weight curves exhibit long-duration, multi-phase, and complex fluctuations, and feature point extraction loses a significant amount of temporal morphological information, making it impossible to accurately describe this complex pattern.

[0029] Therefore, this invention calculates the goodness of fit between the weight curve and the escape behavior curve based on the vehicle length. The specific operation is as follows: For vehicle models whose length exceeds the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, the weight data sequence is obtained from the weight curve, and the escape curve data sequence corresponding to the difference between the maximum and minimum values ​​in the current speed change curve is retrieved from the escape behavior curve. The minimum matching path between the weight data sequence and the escape curve data sequence is found and mapped as the goodness of fit.

[0030] For vehicle models whose length exceeds the length threshold, the minimum matching path between the weight data sequence and the escape curve data sequence is found. Specifically, the two sequences are standardized, and the Euclidean distance between any two points in the standardized sequences is calculated to form a local distance matrix. Using dynamic programming, a cumulative cost matrix is ​​created starting from the top left corner (1,1). The value of each position is equal to the value of the corresponding position in the local distance matrix plus the minimum cumulative cost among its three adjacent positions to the left, below, and to the bottom left. The value of the bottom right corner of the cumulative cost matrix is ​​the minimum global cumulative cost between the two sequences, and the path containing the minimum global cumulative cost is the minimum matching path.

[0031] The standardization process is as follows: the difference between each data point in the sequence and the sequence mean is divided by the sequence standard deviation.

[0032] Furthermore, the escape behavior curve in this step mainly refers to the weight curve when the jumping or dragging of the scale occurs.

[0033] If the fit exceeds the safety threshold, it is confirmed that there is skipping or dragging of the scale, and the weight is re-weighed. If the fit does not exceed the safety threshold and there is no escape behavior, the average value of the weight curve after removing the maximum and minimum values ​​is taken as the weighing result. If the difference between the maximum and minimum values ​​in the speed change curve is less than the difference threshold, the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result.

[0034] This invention enables higher recognition accuracy for long vehicles, precisely capturing their complex and non-rigid cheating curve patterns. This ensures that regardless of the speed at which a long vehicle passes, its cheating pattern can be optimally matched with the template. This reduces the false negative rate for cheating behavior from long vehicles, ensuring detection accuracy. For short vehicles, higher processing efficiency is achieved; a lightweight feature point matching method is sufficient for recognition, reducing computational burden and shortening response time. Even with heavy traffic, rapid detection is guaranteed, avoiding traffic congestion caused by complex algorithms for handling long vehicles.

[0035] For vehicle models with a length less than or equal to the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, features are extracted from the weight curve and the escape behavior curve, including the maximum peak amplitude, fluctuation duration, fluctuation period, and phase difference between the front axle and rear axle fluctuations. The relative error between each pair of features is calculated, a weight is assigned to each relative error, and the weighted comprehensive error is calculated. The comprehensive error is used as the goodness of fit.

[0036] This invention forms a closed-loop verification system from vehicle identity and model verification to wheel and axle matching, and then to weighing and vehicle behavior correlation analysis, ensuring a high degree of consistency of license plate, model, wheel and axle, and weight data, and generating a complete chain of evidence with extremely high credibility.

[0037] S4. Obtain the maximum load capacity of the vehicle model to be inspected based on the model of the vehicle to be inspected, compare the weighing result with the maximum load capacity, and if the weighing result is less than or equal to the maximum load capacity, then the vehicle is not overloaded and is allowed to pass.

[0038] This method reduces manual intervention and effectively avoids misjudgments and omissions through precise judgment logic, thereby enhancing overall control efficiency. It not only ensures the passage efficiency of legal vehicles but also resolutely intercepts overloaded vehicles, achieving efficient, precise, and intelligent control at highway entrances.

[0039] A management system for overloaded and oversized vehicles on highways, used to implement the aforementioned management method for overloaded and oversized vehicles on highways, characterized in that it includes: The license plate detection module acquires the license plates from the front and rear of the vehicle to be detected, extracts the features and symbols of the license plates from the front and rear, and compares them. If the license plate symbols from the front and rear are different and the infrared light is blocked, an infrared image is acquired and edge detection is performed. If the edge detection is not continuous, the detection is repeated. If the edge detection is continuous, a license plate error is indicated. If the license plate symbols from the front and rear are the same, wheel axle detection is performed. The wheel axle detection module acquires an image of the vehicle's front end, classifies the vehicle model based on the image, obtains the wheel axle detection results, and determines whether the results match the vehicle model. If they match, the module then obtains the weighing results. The weighing result acquisition module acquires the side video and weight curve of the vehicle to be inspected. Based on the known length of the weighing platform, it obtains the speed change curve of the vehicle to be inspected. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, it calculates the fit between the weight curve and the escape behavior curve based on the vehicle length. If the fit exceeds the safety threshold, the vehicle is weighed again. If the fit does not exceed the safety threshold, the average value of the weight curve after removing the maximum and minimum values ​​is taken as the weighing result. If the difference between the maximum and minimum values ​​in the speed change curve is less than the difference threshold, the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result. The passage judgment module obtains the upper limit of the vehicle model based on the model of the vehicle to be detected, compares the weighing result with the upper limit of the load, and if the weighing result is less than or equal to the upper limit of the load, then the vehicle is not overloaded and is allowed to pass.

[0040] The system of the present invention also includes an infrared beam obstruction detection sensor. After the vehicle to be detected obstructs the infrared beam obstruction detection sensor, the cameras set at the front and rear of the vehicle start to take pictures and extract features from the pictures. The camera set on the side of the vehicle to be detected records side video and takes infrared images.

Claims

1. A management method for overloaded and oversized vehicles on highways, characterized in that, include: S1. Obtain the license plates from the front and rear of the vehicle to be detected, extract the features and symbols of the license plates from the front and rear, and compare them. If the license plate symbols at the front and rear of the vehicle are different and the infrared light is blocked, an infrared image is acquired and edge detection is performed. If the edge detection is not continuous, the detection is repeated. If the edge detection is continuous, an error message is displayed indicating that the license plate is incorrect. If the license plate symbols at the front and rear of the vehicle are the same, then execute S2; S2. Obtain the front image of the vehicle, classify the model of the vehicle to be detected based on the front image, obtain the wheel axle detection results of the vehicle to be detected, and determine whether they match the model of the vehicle to be detected. If they match, proceed to S3. S3. Obtain the side video and weight curve of the vehicle to be tested. Based on the known length of the weighing platform, obtain the speed change curve of the vehicle to be tested. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, calculate the fitting degree between the weight curve and the escape behavior curve based on the vehicle length. If the fitting degree exceeds the safety threshold, re-weigh the vehicle. If the fitting degree does not exceed the safety threshold, take the average value of the weight curve after removing the maximum and minimum values ​​as the weighing result. If the difference between the maximum and minimum values ​​in the velocity change curve is less than the difference threshold, then the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result. S4. Obtain the maximum load capacity of the vehicle model to be inspected based on the model of the vehicle to be inspected, compare the weighing result with the maximum load capacity, and if the weighing result is less than or equal to the maximum load capacity, then the vehicle is not overloaded and is allowed to pass.

2. The management method for overloaded and oversized vehicles on highways according to claim 1, characterized in that, Based on the vehicle length, S3 calculates the goodness of fit between the weight curve and the escape behavior curve. The specific operation is as follows: For vehicle models whose length exceeds the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, the weight data sequence is obtained from the weight curve, and the escape curve data sequence corresponding to the difference between the maximum and minimum values ​​in the current speed change curve is retrieved from the escape behavior curve. The minimum matching path between the weight data sequence and the escape curve data sequence is found and mapped to the goodness of fit. For vehicle models with a length less than or equal to the length threshold, the goodness of fit between the weight curve and the escape behavior curve is calculated. Specifically, features in the weight curve and escape behavior are extracted, including the maximum peak amplitude, fluctuation duration, fluctuation period, and phase difference between the front and rear axle fluctuations. The relative error between each pair of features is calculated, a weight is assigned to each relative error, and the weighted comprehensive error is calculated. The comprehensive error is used as the goodness of fit.

3. The management method for overloaded and oversized vehicles on highways according to claim 2, characterized in that, Finding the minimum matching path between the weight data sequence and the escape curve data sequence involves: standardizing both sequences, calculating the Euclidean distance between any two points in the standardized sequences to form a local distance matrix, and using dynamic programming to create a cumulative cost matrix starting from the top left corner (1,1). The value at each position is equal to the value at the corresponding position in the local distance matrix plus the minimum cumulative cost among its three adjacent positions to the left, below, and to the bottom left. The value at the bottom right corner of the cumulative cost matrix is ​​the minimum global cumulative cost between the two sequences, and the path containing the minimum global cumulative cost is the minimum matching path.

4. The management method for overloaded and oversized vehicles on highways according to claim 3, characterized in that, The standardization process is as follows: the difference between each data point in the sequence and the sequence mean is divided by the sequence standard deviation.

5. A management method for overloaded and oversized vehicles on highways according to claim 1, characterized in that, The symbols described in S1 include characters, letters, and numbers.

6. A management method for overloaded and oversized vehicles on highways according to claim 1, characterized in that, In S2, edge detection coherence is specifically defined as follows: For the image with edge detection, connected components are extracted, and the connected component with the largest area is found. The pixel size of the largest connected region in the image is converted into the actual length and width. If the converted length is greater than or equal to the minimum vehicle length threshold, the width is greater than or equal to the minimum effective lane width threshold, and the aspect ratio is within the range of typical vehicle aspect ratios, then the edge detection is determined to be coherent; otherwise, the edge detection is determined to be discontinuous.

7. A management system for overloaded and oversized vehicles on highways, used to implement the management method for overloaded and oversized vehicles on highways as described in claims 1-6, characterized in that, include: The license plate detection module acquires the license plates from the front and rear of the vehicle to be detected, extracts the features and symbols of the license plates from the front and rear, and compares them. If the license plate symbols from the front and rear are different and the infrared light is blocked, an infrared image is acquired and edge detection is performed. If the edge detection is not continuous, the detection is repeated. If the edge detection is continuous, a license plate error is indicated. If the license plate symbols from the front and rear are the same, wheel axle detection is performed. The wheel axle detection module acquires an image of the vehicle's front end, classifies the vehicle model based on the image, obtains the wheel axle detection results, and determines whether the results match the vehicle model. If they match, the module then obtains the weighing results. The weighing result acquisition module acquires the side video and weight curve of the vehicle to be inspected. Based on the known length of the weighing platform, it obtains the speed change curve of the vehicle to be inspected. If the difference between the maximum and minimum values ​​in the speed change curve is greater than the difference threshold, it calculates the fit between the weight curve and the escape behavior curve based on the vehicle length. If the fit exceeds the safety threshold, the vehicle is weighed again. If the fit does not exceed the safety threshold, the average value of the weight curve after removing the maximum and minimum values ​​is taken as the weighing result. If the difference between the maximum and minimum values ​​in the speed change curve is less than the difference threshold, the average value of the weight curve after removing the maximum and minimum values ​​is calculated as the weighing result. The passage judgment module obtains the upper limit of the vehicle model based on the model of the vehicle to be detected, compares the weighing result with the upper limit of the load, and if the weighing result is less than or equal to the upper limit of the load, then the vehicle is not overloaded and is allowed to pass.

8. A management system based on overloading and over-limit loading on highways according to claim 7, characterized in that, It also includes cameras installed at the front parking space of the vehicle, as well as cameras on the sides of the vehicle.

9. A management system based on overloading and over-limit loading on highways according to claim 7, characterized in that, It also includes an infrared beam obstruction detection sensor, which starts taking pictures when the infrared beam is blocked.