Non-stop overload and overlimit detection method and system

Abstract

The present invention discloses a non-stop overload and overlimit detection method and system, which relates to the technical field of overload detection. The method comprises: detecting the license plate information, axle weight data and 3D point cloud data of the vehicle to be inspected by multi-source sensors, and judging whether the modification is carried out in combination with the cloud-based vehicle modification report database; if the modification report exists, further combining the 3D point cloud data to identify the private re-modification and generate a judgment result; performing local and overall overload detection through axle weight data, and using 3D point cloud data to perform volume overload detection, and finally generating two detection results, and marking the vehicle for overload and overlimit. The present invention solves the technical problems of low efficiency of existing parking overload and overlimit detection, and insufficient accuracy in identifying modified vehicles and private re-modification, and achieves the technical effect of improving detection efficiency and accurately identifying private modification of vehicles through non-stop detection and modification identification, and ensuring the accuracy of overload detection.

Description

Technical Field

[0001] The invention relates to the technical field of overload detection, and in particular to a non-stop overload and overlimit detection method and system. Background Art

[0002] Existing overload and over-limit detection technologies usually rely on parking-type detection methods, which are inefficient and easily cause traffic congestion. In addition, traditional methods have limitations in dealing with modified vehicles and privately modified vehicles. It is difficult to accurately determine whether the vehicle meets the reported information, resulting in large detection errors. At the same time, for the comprehensive over-limit detection of vehicle weight and volume, a single sensor or independent detection method is often used, lacking comprehensive data integration and efficient analysis, and it is difficult to meet the real-time and accuracy requirements of modern traffic monitoring. Summary of the invention

[0003] The present application provides a non-stop overload and over-limit detection method and system, which are used to solve the technical problems that the existing parking-type overload and over-limit detection is inefficient and the recognition accuracy of modified vehicles and unauthorized modifications is insufficient.

[0004] The first aspect of the present application provides a non-stop overload and overlimit detection method, the method comprising: when the vehicle to be inspected drives into a preset overload and overlimit detection area, generating license plate information, axle weight detection data set and 3D point cloud scanning data set through preset multi-source sensor detection; comparing the license plate information in a cloud-based vehicle modification reporting database to determine whether the vehicle to be inspected has been reported for vehicle modification; if the vehicle to be inspected has been reported for vehicle modification, extracting the modification reporting features of the vehicle to be inspected from the cloud-based vehicle modification reporting database, and combining the 3D point cloud scanning data set to identify unauthorized re-modification of the vehicle's own weight, and generating an unauthorized re-modification judgment result; performing dual overload and overlimit detection of local axle weight and overall weight based on the unauthorized re-modification judgment result and the axle weight detection data set to generate a first detection result; performing volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result; marking the vehicle to be inspected for overload and overlimit with the first detection result and the second detection result.

[0005] The second aspect of the present application provides a non-stop overload and overlimit detection system, the system comprising: a multi-source sensor detection module, the multi-source sensor detection module is used to generate license plate information, axle weight detection data set and 3D point cloud scanning data set through preset multi-source sensor detection when the vehicle to be inspected drives into a preset overload and overlimit detection area; a vehicle modification information matching module, the vehicle modification information matching module is used to compare the license plate information in the cloud-based vehicle modification reporting database to determine whether the vehicle to be inspected has been reported for vehicle modification; an unauthorized re-modification identification module, the unauthorized re-modification identification module is used to extract the vehicle to be inspected from the cloud-based vehicle modification reporting database if the vehicle to be inspected has been reported for vehicle modification. The modification reporting features of the inspected vehicle are detected, and the unauthorized re-modification of the vehicle's own weight is identified in combination with the 3D point cloud scanning data set to generate an unauthorized re-modification judgment result; a double overload and overlimit detection module is used to perform a double overload and overlimit detection of the local axle weight and the overall weight according to the unauthorized re-modification judgment result and the axle weight detection data set to generate a first detection result; a volume overload and overlimit detection module is used to perform a volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result; an overload and overlimit marking module is used to mark the vehicle to be inspected as overloaded and overlimited with the first detection result and the second detection result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The non-stop overload and overlimit detection method and system provided by the present application relate to the field of overload detection technology. The vehicle information is detected by multi-source sensors, and the modification is determined in combination with a cloud-based modification reporting database, and the unauthorized modification is further identified. The axle weight data and 3D point cloud data are used to perform local, overall weight and volume overload detection respectively, and the detection results are generated and overloaded and overlimit vehicles are marked. The technical problems of low efficiency of existing parking-type overload and overlimit detection and insufficient accuracy in identifying modified vehicles and unauthorized modifications are solved, and the technical effect of improving detection efficiency and accurately identifying unauthorized modifications of vehicles and ensuring the accuracy of overload detection is achieved through non-stop detection and modification identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1A flow chart of a non-stop overload and overlimit detection method provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a non-stop overload and over-limit detection system provided in an embodiment of the present application.

[0009] Explanation of the reference numerals: multi-source sensing detection module 11 , vehicle modification information matching module 12 , unauthorized modification identification module 13 , double overload and overlimit detection module 14 , volume overload and overlimit detection module 15 , overload and overlimit marking module 16 . DETAILED DESCRIPTION

[0010] The present application provides a non-stop overload and over-limit detection method and system, which are used to solve the technical problems that the existing parking-type overload and over-limit detection is inefficient and the recognition accuracy of modified vehicles and unauthorized modifications is insufficient.

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Embodiment 1, as Figure 1 As shown, the present application provides a non-stop overload and overlimit detection method, the method comprising: P10: When the vehicle to be inspected drives to the preset overload and overlimit detection area, the preset multi-source sensor detection generates license plate information, axle weight detection data set and 3D point cloud scanning data set.

[0014] The multi-source sensor includes a smart camera, a laser radar and an embedded axle load sensor, the embedded axle load sensor includes one or more of a quartz sensor, a piezoelectric sensor and a fiber grating sensor, and the specific configuration steps of the embedded axle load sensor include: P11: Collect the road surface flatness of the preset overload and overlimit detection area; P12: If the road surface flatness is less than or equal to the preset flatness threshold, construct the embedded axle weight sensor with a quartz sensor and a piezoelectric sensor, wherein the preset flatness threshold is the road surface flatness whose error is less than the error threshold obtained by analyzing the axle weight measurement error through historical data.

[0015] It should be understood that when the vehicle to be inspected drives to the preset overload and overlimit detection area, the system will collect data through a series of multi-source sensors, including smart cameras, lidars and embedded axle weight sensors. These sensors work together to provide important information about the vehicle. The task of the smart camera is to collect the license plate information of the vehicle and identify the license plate number through image processing technology, so that the vehicle can be compared and identified in the cloud. At the same time, the lidar is used to generate a 3D point cloud scanning data set. Through high-precision three-dimensional scanning, the system can capture the specific shape and size of the vehicle for volume overload detection. Finally, the embedded axle weight sensor is used to detect the weight of each axle of the vehicle and accurately measure the axle weight data through different sensor types (such as quartz, piezoelectric, fiber grating, etc.). The data provided by these sensors, after being processed by the system, form license plate information, axle weight data set and 3D point cloud scanning data set, laying the foundation for subsequent overload and overlimit detection.

[0016] Before performing overload detection, it is first necessary to evaluate the road surface smoothness of the area to be inspected. The road surface smoothness directly affects the accuracy of the axle load sensor because if the road surface is uneven, the sensor may produce measurement errors due to vibration or uneven pressure distribution. Therefore, the system needs to first collect road surface data in the area to be inspected to evaluate the flatness of the road surface. This process collects road surface smoothness data through sensors (such as ground sensors or road surface scanning equipment) and generates a flatness report for subsequent processing. If poor road surface flatness is detected, additional steps may need to be taken to calibrate the measurement system to ensure the accuracy of the data.

[0017] If the road surface flatness is detected to be less than or equal to the preset flatness threshold, it is considered that the road condition is not sufficient to ensure the accuracy of axle weight measurement. The type of sensor to be used for further detection is determined based on this evaluation result. In this case, in order to minimize the measurement error, quartz sensors and piezoelectric sensors can be used to build embedded axle weight sensors. Quartz sensors perform well in demanding environments due to their high stability and accuracy; piezoelectric sensors have strong dynamic response capabilities and can quickly and accurately detect instantaneous load changes. The combination of the two can provide more accurate axle weight data under uneven road conditions, thereby reducing the error caused by uneven road conditions. The preset flatness threshold is derived from the analysis of historical data. It is based on the study of axle weight measurement errors under different road conditions to determine a reasonable standard to ensure that the measurement error is controlled within an acceptable range. When the road surface flatness reaches this threshold, the system will automatically switch to a higher-precision sensor configuration to ensure that the results of overload and overrun detection are not affected by road conditions. In this way, the system can intelligently adjust the sensor configuration according to different road conditions to ensure the efficiency and accuracy of overload detection.

[0018] Furthermore, step P10 of the embodiment of the present application further includes: P13: If the road surface flatness is greater than a preset flatness threshold, the embedded axle weight sensor is constructed using a quartz sensor, a piezoelectric sensor, and a fiber grating sensor, and an error compensation network based on tire contact stress is configured; wherein the error compensation network is constructed using historical tire contact stress data and historical axle measurement data to learn the relationship between tire contact stress and axle load.

[0019] Optionally, when the road surface flatness is detected to be greater than the preset flatness threshold, it means that the road condition in the current detection area is good enough to perform more accurate axle weight measurement. In this case, a more efficient and accurate sensor combination can be selected, that is, to build an embedded axle weight sensor through quartz sensors, piezoelectric sensors and fiber grating sensors. These sensors each have their own advantages. Quartz sensors can provide stable and high-precision static measurements and are suitable for long-term load detection; piezoelectric sensors respond quickly and are suitable for capturing instantaneous pressure changes, and are often used for dynamic load measurement; and fiber grating sensors can accurately detect stress in a high-speed changing environment, have extremely high anti-interference capabilities, and are especially suitable for complex working environments.

[0020] In addition, an error compensation network based on tire contact stress can be configured. Tire contact stress refers to the pressure distribution generated when the vehicle tire contacts the road surface. This factor will affect the measurement accuracy of the sensor because the tire contact stress will change the load distribution of the axle to a certain extent. In order to reduce the measurement error caused by contact stress, the error compensation network can be used for real-time correction.

[0021] The error compensation network can be constructed based on historical tire contact stress data and historical axle load data. By collecting a large amount of historical data, the network can learn the relationship between tire contact stress and axle load. Specifically, by analyzing the rules in historical data through machine learning technology, a model is established to accurately predict the possible deviations in measurement results under different tire contact stress conditions. This network plays a corrective role in the real-time data transmission process. Once a change in tire contact stress is detected, the system will immediately adjust according to the learned relationship model to correct the axle weight data read by the sensor to ensure its accuracy.

[0022] This comprehensive configuration not only provides more accurate axle weight data on flat roads, but also effectively eliminates errors caused by changes in tire contact stress, thereby further improving the accuracy and stability of overload detection.

[0023] P20: Compare the license plate information with the cloud-based vehicle modification reporting database to determine whether the vehicle to be inspected has been reported for vehicle modification.

[0024] Specifically, when a vehicle passes through the overload and over-limit detection area, its license plate information is first obtained through the smart camera, and then transmitted to the cloud-based vehicle modification reporting database for comparison. The cloud-based database contains all reported vehicle modification information, which is usually actively reported by the vehicle owner when modifying the vehicle, including parameters such as weight and size after modification. The system will quickly locate the corresponding vehicle record based on the license plate number and determine whether the vehicle has been modified and reported in accordance with regulations.

[0025] The license plate information is the unique identifier of the vehicle. The license plate number can be accurately read and extracted through license plate recognition technology (such as OCR technology, optical character recognition). The accuracy of license plate recognition is directly related to the subsequent comparison results. Image processing algorithms can be combined with high-resolution images of smart cameras to achieve efficient and accurate license plate recognition.

[0026] The cloud-based vehicle modification reporting database is a big data platform that centrally stores all vehicle modification information. The database records the basic information of the vehicle, the reported modification items, and the modified data (such as the dead weight and axle load before and after the modification). The purpose of vehicle modification is usually to increase the load-bearing capacity or change certain structural characteristics of the vehicle, so the weight and size after modification may change.

[0027] By comparing the license plate information with the vehicle records in the cloud database, it is possible to accurately determine whether the vehicle has been reported for modification in accordance with regulations. If the vehicle cannot find a matching record in the database, it is determined that the vehicle has not been reported for modification, so that subsequent overload detection and possible inspection procedures can be carried out. Conversely, if the corresponding modification registration record is found, the modification items can be further checked to ensure that subsequent inspections are based on the correct vehicle parameters. This comparison process is efficiently completed through an automated cloud computing platform, ensuring real-time and accuracy.

[0028] This step plays a key role in the entire overload and over-limit detection process. By comparing with the cloud database, it can identify whether there is any unreported modification behavior, providing a strong basis for subsequent identification of unauthorized modifications and overload and over-limit detection.

[0029] Further, to determine whether the vehicle to be inspected has been reported for vehicle modification, the embodiment of the present application further includes step P20a, and step P20a further includes: P21a: If the vehicle to be inspected has not been reported for vehicle modification, the vehicle type is matched in the vehicle registration information database according to the license plate information; P22a: Based on the vehicle type, the local axle weight overlimit threshold and the overall weight overlimit threshold corresponding to the unmodified state are determined, and the dual overload and overlimit detection of the local axle weight and the overall weight is performed in combination with the axle weight detection dataset; P23a: Based on the vehicle type, the corresponding cargo volume overlimit threshold when the unmodified state is determined, and the volume overload and overlimit detection is performed in combination with the 3D point cloud scanning dataset.

[0030] In a possible embodiment of the present application, if the vehicle has not been reported for modification, further detailed inspection may be performed.

[0031] First, the vehicle registration information is matched against the license plate information of the vehicle to be inspected. The vehicle registration information database is a database containing the basic information of all registered vehicles, which records key information such as the model, type, and purpose of each vehicle. By matching the license plate information, the system can quickly identify the registration type of the vehicle and confirm the basic structure and specifications of the vehicle. The license plate information is the unique identifier of the vehicle and is accurately extracted through license plate recognition technology to provide basic data for subsequent judgments.

[0032] Once the type of vehicle is determined, the normal load-bearing capacity of the vehicle without modification can be determined based on the vehicle type, especially its local axle weight overlimit threshold and overall weight overlimit threshold. These thresholds are defined based on vehicle type, model specifications and manufacturer standards, and are usually provided by official data when the vehicle is registered. Confirmation of the vehicle type enables the system to select appropriate standards to determine whether the vehicle is overloaded. On this basis, combined with the previously collected axle weight detection data set, dual overload and overlimit detection of local axle weight and overall weight is performed. If the axle load exceeds the preset threshold, it is determined to be overloaded and a warning is generated.

[0033] In addition to axle weight detection, it is also necessary to check whether the vehicle's cargo is overloaded, which is usually done through a 3D point cloud scanning data set. Based on the vehicle type, the system will determine the corresponding cargo volume overload threshold when the vehicle is not modified. This standard is usually closely related to the vehicle's total length, width, height and designed cargo space. The 3D point cloud scanning data set is provided by LiDAR, which is used to scan the three-dimensional form of the vehicle and its cargo, and accurately measure the volume of the cargo carried by the vehicle. When the scanning data indicates that the cargo volume exceeds the specified threshold, the system will determine that the volume is overloaded. By combining 3D point cloud scanning data and the set volume overload threshold, the system can accurately determine whether the cargo is overloaded, further ensuring the comprehensiveness and accuracy of the detection.

[0034] P30: If the vehicle to be inspected has been reported for vehicle modification, the modification report features of the vehicle to be inspected are extracted from the cloud-based vehicle modification report database, and the vehicle's own weight is combined with the 3D point cloud scanning data set to identify unauthorized re-modification, and generate an unauthorized re-modification judgment result.

[0035] Furthermore, step P30 of the embodiment of the present application further includes: P31: Map and segment the modified parts of the modification report features to generate multiple modification parameters of the multiple modified parts; P32: Segment and model the point cloud data of the 3D point cloud scanning data set based on the multiple modified parts to generate multiple local models; P33: Use the multiple local models and the multiple modification parameters to identify the unauthorized modification of the vehicle's own weight to generate the unauthorized modification judgment result.

[0036] It should be understood that when the vehicle to be inspected is confirmed to have been reported for vehicle modification in the cloud-based vehicle modification reporting database, the vehicle's modification reporting features are extracted and combined with the 3D point cloud scanning data set to identify the unauthorized re-modification of the vehicle's deadweight. The purpose of this step is to determine whether there is any unauthorized modification after the vehicle is reported for modification, such as the removal of part of the modification or the unreported change in deadweight, which may affect the vehicle's overweight detection results.

[0037] First, the extracted modification report features are analyzed and mapped and segmented. The modification report features include detailed parameters and part information of the vehicle modification, usually including the weight, size, wheelbase, cargo capacity, etc. after the modification. These features are obtained based on the official report information submitted when the vehicle is modified. Through mapping and segmentation, the reported modified parts are divided to generate multiple modification parameters related to the modification, such as the weight change and material strength change of different parts after modification. This information provides basic data for the subsequent identification of unauthorized modifications.

[0038] Next, based on the modification parameters of these multiple modified parts, the previously collected 3D point cloud scanning data set is further processed to perform segmentation and modeling of the point cloud data. The 3D point cloud scanning data set is provided by sensors such as LiDAR, which can accurately capture the three-dimensional spatial information of the vehicle and its cargo. In this step, the scanned three-dimensional point cloud data is segmented according to the mapping features of the modified parts to generate multiple local models. Each local model represents the three-dimensional shape of a part of the vehicle, such as the body, frame, cargo box, etc., and each local model can be compared with the corresponding modification features.

[0039] Finally, by comparing and analyzing these multiple local models and multiple modification parameters, the identification of unauthorized modification of the vehicle's own weight is performed. For example, if the vehicle's own weight is found to have changed in the local model of a certain part (for example, the weight is reduced due to the removal of the part after modification), this may indicate that the vehicle has been unauthorizedly modified without reporting. During this process, the system will compare the modification report features with the actual 3D model data to determine whether there are inconsistencies, especially changes in key parameters such as weight and volume, and generate unauthorized modification judgment results, indicating whether the vehicle has been unauthorizedly modified.

[0040] Through the processing of the above steps, it is possible to fully identify whether there is any unauthorized modification after the vehicle is reported for modification, further enhancing the system's monitoring ability of vehicle modification behavior and providing strong support for subsequent overload detection.

[0041] Furthermore, step P33 of the embodiment of the present application also includes: P33-1: Perform adaptation analysis and fusion on the multiple modification parameters and the multiple local models to generate a fusion modification adaptation index; P33-2: If the fusion modification adaptation index is greater than the preset adaptation index, mark the vehicle to be inspected as unauthorized re-modification and generate the unauthorized re-modification judgment result.

[0042] Optionally, after completing the preliminary analysis of multiple local models and multiple modification parameters, the embodiments of the present application further perform adaptation analysis and fusion processing on these data, and determine whether there is any unreported private modification through in-depth analysis of the modification characteristics and vehicle models.

[0043] First, the extracted multiple modification parameters are analyzed for matching with the corresponding multiple local models. The core of the matching analysis is to compare the matching degree between the parameters of each modified part and the actual model of the vehicle. For example, if the size, weight or other key parameters of a modified part change, the system will analyze whether these changes match the actual 3D model. This process requires combining the modification parameters (such as the weight increase or decrease after modification, size changes, etc.) with the actual data of the relevant parts in each local model to conduct accurate matching analysis and evaluate whether the vehicle has been privately dismantled and modified.

[0044] Then, through the adaptation analysis results, the data of various modification parameters and local models are integrated to generate a fusion modification adaptation index. This index combines the changes in the modified parts of the vehicle and the adaptability of the local model, and quantifies whether the vehicle meets the modification reporting standards. The fusion modification adaptation index not only considers the physical changes on the outside of the vehicle, but also involves structural changes, such as the strength of the frame and body, load capacity and other factors. Through this fusion process, a comprehensive and accurate evaluation index can be obtained.

[0045] After the fusion modification adaptation index is generated, the index is compared with the preset adaptation index threshold. If the fusion modification adaptation index is greater than the threshold, it means that the vehicle may have been privately modified after the modification was reported. At this time, the vehicle can be marked as privately modified and a privately modified judgment result can be generated. This judgment result will clearly indicate whether the vehicle has unreported private modification behavior, thereby affecting the result of overload and overlimit detection.

[0046] The preset adaptation index threshold is determined by historical data, experimental analysis and modification standards, which can ensure the accuracy of the system in identifying unauthorized modifications. If the fusion index exceeds the threshold, it means that some modified parts of the vehicle have changed significantly and exceeded the normal modification range, so it can be confirmed that the vehicle may have been unauthorizedly modified.

[0047] Furthermore, step P33-1 of the embodiment of the present application also includes: P33-11: Extract multiple structural modification parameters from the multiple modification parameters, perform adaptation analysis on the multiple local models and the multiple structural modification parameters, and generate multiple local fitness indicators; P33-12: Extract multiple modification weight gain parameters from the multiple modification parameters, and configure multiple modification influence weights for the multiple modification parts; P33-13: Perform weighted fusion of the multiple local fitness indicators with the multiple modification influence weights to generate the fused modification adaptation index.

[0048] In a possible embodiment of the present application, a detailed adaptation analysis can be further performed on multiple local models of the modified vehicle to generate accurate fusion modification adaptation indicators to effectively identify whether the vehicle has been privately modified.

[0049] First, multiple structural modification parameters are extracted from multiple modification parameters. These parameters mainly describe the structural changes of the vehicle, such as frame reinforcement, chassis changes, or cargo compartment volume adjustments. For these modification parameters, the system performs adaptation analysis on multiple local models to check the matching degree between the modified parts of the vehicle and the pre-reported design parameters. This process evaluates the compliance of each modified part one by one by comparing the actual 3D point cloud scan data with the reported information. Through this adaptation analysis, multiple local fitness indicators are generated, which reflect whether the modification of each part meets the vehicle reporting requirements.

[0050] Next, multiple modification weight gain parameters are extracted. These weight gain parameters involve the weight added after the vehicle is modified, such as the additional weight caused by adding reinforced steel plates or changing the chassis configuration. In order to quantify the impact of modifications to different parts on the vehicle's own weight, the system assigns a modification impact weight to each modified part. This step evaluates the actual impact of each part modification on the change in self-weight through historical data analysis, while taking into account the differences in the impact of different modified parts on vehicle performance and safety, thereby assigning different weights to different parts. Certain key components, such as suspension system or frame modifications, usually have a greater impact on the vehicle's load capacity and stability, so they have higher weights.

[0051] Finally, multiple local fitness indicators are weighted and fused with the modification influence weights configured above to generate a fused modification fitness indicator. The core purpose of weighted fusion is to reasonably quantify the fitness of each modified part to ensure that the modification of important parts has a greater impact on the final fitness indicator. The fusion formula calculates all local fitness indicators by weight to obtain the final comprehensive result, which accurately reflects the compliance of the entire vehicle modification. If the fusion index is lower than the preset standard, it indicates that the vehicle may have been privately modified, which can be further marked and the privately modified judgment result can be generated as the basis for overload and overlimit detection.

[0052] Through the precise execution of the above steps, the vehicle modification can be comprehensively and meticulously detected and analyzed, providing an accurate basis for the subsequent overload and over-limit identification.

[0053] P40: Perform a dual overload and overlimit detection of the local axle weight and the overall weight based on the unauthorized re-modification judgment result and the axle weight detection data set to generate a first detection result.

[0054] Furthermore, step P40 of the embodiment of the present application also includes: P41: Extract each local axle weight data based on the axle weight detection data set; P42: Add and calculate the each local axle weight data to generate the overall weight data; P43: If the unauthorized re-modification judgment result does not include the unauthorized re-modification mark, determine the local axle weight over-limit threshold and the overall weight over-limit threshold according to the modification reporting characteristics; P44: Use the local axle weight over-limit threshold and the overall weight over-limit threshold to make over-limit judgments on the each local axle weight data and the overall weight data to generate the first detection result.

[0055] It should be understood that based on the unauthorized re-modification judgment result and the axle weight detection data set, dual overload and overlimit detection of local axle weight and overall weight is performed to generate a first detection result to ensure the accuracy and reliability of the detection.

[0056] First, extract each local axle weight data from the axle weight detection data set. This data is obtained through embedded axle weight sensors (such as quartz sensors, piezoelectric sensors or fiber optic sensors) and reflects the weight of each axle of the vehicle. By obtaining these local axle weight data, the system can analyze in detail whether each axle is overloaded, laying the foundation for subsequent overload and overlimit judgment.

[0057] Next, the extracted partial axle weight data are added up to obtain the overall weight data of the vehicle. The purpose of this operation is to aggregate the weight data of each axle and calculate the total weight of the vehicle, so as to detect the weight overload of the vehicle. The overall weight data can help detect whether there is a risk of overloading the entire vehicle, especially after the vehicle is modified, which may cause overloading due to the reduction of its own weight.

[0058] Furthermore, based on the previous private modification judgment result, if the result does not contain the private modification mark, the local axle weight overlimit threshold and the overall weight overlimit threshold are determined based on the modification report characteristics. This means that when the vehicle has not been privately modified, the system will set the corresponding overload and overlimit detection standards based on the modification report information (such as the report of weight increase or weight reduction). These standards are determined comprehensively through historical data and regulatory requirements to ensure that the overload detection of each vehicle is in line with the actual situation.

[0059] On the contrary, if the unauthorized modification mark is included, it means that the vehicle may have been unauthorizedly modified after the modification was reported, such as reducing the vehicle's own weight to cover up the fact of overloading. In this case, it is not necessary to rely on the vehicle's reported parameters, and the vehicle can be directly detected for overloading and unauthorized modification and warned, prompting relevant departments to intervene.

[0060] Finally, the local axle weight over-limit threshold and the overall weight over-limit threshold are used to judge the over-limit of each extracted local axle weight data and the overall weight data. By comparing the actual measured axle weight and overall weight with the set over-limit threshold, it is judged whether there is an overload. If the weight of an axle exceeds its threshold, or the weight of the whole vehicle exceeds the overall threshold, the vehicle will be marked as overloaded and the first detection result will be generated.

[0061] Through the implementation of the above steps, it is possible to effectively combine the vehicle modification report and accurately perform dual tests of local axle weight and overall weight, ensuring that the vehicle's overload detection results are more accurate, thereby improving the efficiency and reliability of overload and over-limit detection.

[0062] P50: Perform volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result.

[0063] Specifically, based on the 3D point cloud scanning data set, volume overload and overlimit detection is performed, and a second detection result is generated. The core of this process is to determine whether the vehicle has a volume overload problem by accurately analyzing the three-dimensional data of the vehicle.

[0064] Specifically, first, the 3D point cloud scanning data set is obtained through LiDAR or other 3D scanning equipment, which can provide accurate 3D point cloud data of the vehicle and its cargo area. These point cloud data contain the spatial information of the vehicle and its cargo, such as the shape of the vehicle body, the boundaries of the carriage, the height and width of the cargo stacking, etc. Through these data, the system can intuitively obtain the volume information of the vehicle and cargo, and then determine whether it meets the volume requirements for transportation.

[0065] Next, based on these 3D point cloud scan data, the system will analyze the data through point cloud processing algorithms to extract the shape and size of the cargo area. Usually, this process involves segmenting and modeling the point cloud data to determine the specific shape of the vehicle cargo compartment and the volume of the objects loaded in it. By performing geometric modeling and volume calculation on the point cloud data, the system can accurately estimate the total volume of the cargo.

[0066] When the cargo volume information is obtained, the system will compare it with the preset cargo volume overlimit threshold, which is based on the vehicle type, purpose, and transportation laws and regulations. If the volume of the cargo exceeds the threshold, the vehicle is judged to have a volume overload problem, and a second test result is generated. This result will indicate whether the vehicle violates the volume overload limit, ensuring the comprehensiveness of overload detection, not only relying on weight detection, but also taking into account the compliance of the cargo volume. It provides a more detailed and three-dimensional assessment for overload and overlimit detection, further improving the accuracy and reliability of detection.

[0067] P60: Mark the vehicle to be inspected as overloaded and over-limit based on the first detection result and the second detection result.

[0068] Optionally, the aforementioned first detection result and the second detection result are combined to mark the vehicle to be inspected as overloaded or over-limited, so as to clarify whether there is an overload or over-limit situation, which is convenient for further processing measures.

[0069] First, the first test result comes from the aforementioned double overload and overlimit test based on the axle weight test data set. This result covers the overload of the vehicle's local axle weight and overall weight. Specifically, the system will determine whether it exceeds the preset weight threshold based on the axle weight data and overall weight data of the vehicle to be tested. If it is overweight, the system will mark it as "overloaded".

[0070] The second test result comes from the 3D point cloud scanning data set. By testing the volume of the vehicle's cargo, it is determined whether there is a volume overload problem. This test can identify whether the size of the cargo exceeds the prescribed volume limit, thereby determining whether there is a volume overload. This step not only considers the weight of the vehicle, but also the volume of the cargo, providing a more comprehensive overload judgment.

[0071] The two test results are analyzed comprehensively. If one or both of the first test result or the second test result indicate that the vehicle is overloaded or over-limited, the vehicle to be inspected is marked as overloaded or over-limited, and relevant marking information is generated. The generation of this mark is an automated process based on the previous data collection and analysis results, without the need for manual intervention, which greatly improves the efficiency and accuracy of the inspection. By combining two different detection methods - axle weight detection and volume detection, it is possible to more accurately determine whether the vehicle is overloaded, avoiding misjudgment that may be caused by a single indicator, improving the comprehensiveness and accuracy of the inspection, and ensuring the effective identification and management of overload problems.

[0072] In summary, the embodiments of the present application have at least the following technical effects: This application uses multi-source sensors to detect the license plate information, axle weight data and 3D point cloud data of the vehicle to be inspected, and combines it with the cloud-based vehicle modification report database to determine whether the modification has been performed. If the modification report exists, the 3D point cloud data is further combined to identify the unauthorized modification and generate a judgment result. Subsequently, local and overall overload detection is performed using the axle weight data, and volume overload detection is performed using 3D point cloud data. Finally, two test results are generated, and the vehicle is marked as overloaded and over-limited to ensure accurate identification of modified vehicles and overload conditions.

[0073] The technical effect of improving detection efficiency and accurately identifying unauthorized modifications of vehicles through non-stop detection and modification identification, thereby ensuring the accuracy of overload detection, is achieved.

[0074] Embodiment 2, based on the same inventive concept as the non-stop overload and overlimit detection method in the above embodiment, Figure 2 As shown, the present application provides a non-stop overload and overrun detection system, and the system and method embodiments in the present application are based on the same inventive concept. The system includes: The multi-source sensing detection module 11 is used to generate license plate information, axle weight detection data set and 3D point cloud scanning data set through preset multi-source sensor detection when the vehicle to be inspected drives into a preset overload and overlimit detection area.

[0075] The vehicle modification information matching module 12 is used to compare the license plate information in the cloud-based vehicle modification reporting database to determine whether the vehicle to be inspected has been reported for vehicle modification.

[0076] The unauthorized re-modification identification module 13 is used to extract the modification report features of the vehicle to be inspected from the cloud-based vehicle modification report database if the vehicle to be inspected has been reported for vehicle modification, and to identify the unauthorized re-modification of the vehicle's own weight in combination with the 3D point cloud scanning data set to generate an unauthorized re-modification judgment result.

[0077] The double overload and overlimit detection module 14 is used to perform double overload and overlimit detection of local axle weight and overall weight according to the unauthorized refitting judgment result and the axle weight detection data set to generate a first detection result.

[0078] The volume overload and overlimit detection module 15 is used to perform volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result.

[0079] An overload and over-limit marking module 16 is used to mark the vehicle to be inspected as overloaded and over-limited based on the first detection result and the second detection result.

[0080] Furthermore, the multi-source sensing detection module 11 is also used to perform the following steps: The multi-source sensor includes an intelligent camera, a lidar and an embedded axle weight sensor, and the embedded axle weight sensor includes one or more of a quartz sensor, a piezoelectric sensor and a fiber grating sensor. The specific configuration steps of the embedded axle weight sensor include: collecting the road surface flatness of the preset overload and overlimit detection area; if the road surface flatness is less than or equal to a preset flatness threshold, constructing the embedded axle weight sensor with a quartz sensor and a piezoelectric sensor, wherein the preset flatness threshold is the road surface flatness with an error less than the error threshold obtained by analyzing the axle weight measurement error through historical data.

[0081] Furthermore, the multi-source sensing detection module 11 is also used to perform the following steps: If the road surface flatness is greater than a preset flatness threshold, the embedded axle weight sensor is constructed using a quartz sensor, a piezoelectric sensor, and a fiber grating sensor, and an error compensation network based on tire contact stress is configured; wherein the error compensation network is constructed by using historical tire contact stress data and historical axle measurement data to learn the relationship between tire contact stress and axle load.

[0082] Furthermore, the vehicle modification information matching module 12 is further configured to perform the following steps: If the vehicle to be inspected has not been reported for vehicle modification, the vehicle type is matched in the vehicle registration information database according to the license plate information; the local axle weight overlimit threshold and the overall weight overlimit threshold corresponding to the unmodified state are determined based on the vehicle type, and the dual overload and overlimit detection of the local axle weight and the overall weight is performed in combination with the axle weight detection data set; the cargo volume overlimit threshold corresponding to the unmodified state is determined based on the vehicle type, and the volume overload and overlimit detection is performed in combination with the 3D point cloud scanning data set.

[0083] Furthermore, the unauthorized refitting identification module 13 is further used to perform the following steps: The modification report features are mapped and segmented into modified parts to generate multiple modification parameters for the multiple modified parts; the 3D point cloud scanning data set is segmented and modeled based on the multiple modified parts to generate multiple local models; the vehicle's own weight unauthorized re-modification is identified using the multiple local models and the multiple modification parameters to generate the unauthorized re-modification judgment result.

[0084] Furthermore, the unauthorized refitting identification module 13 is further used to perform the following steps: The multiple modification parameters and the multiple local models are adapted and analyzed and integrated to generate an integrated modification adaptation index; if the integrated modification adaptation index is greater than a preset adaptation index, the vehicle to be inspected is marked as unauthorized re-modification to generate the unauthorized re-modification judgment result.

[0085] Furthermore, the unauthorized refitting identification module 13 is further used to perform the following steps: Extract multiple structural modification parameters from the multiple modification parameters, perform adaptation analysis on the multiple local models and the multiple structural modification parameters, and generate multiple local fitness indexes; extract multiple modification weight gain parameters from the multiple modification parameters, and configure multiple modification influence weights for the multiple modification parts; perform weighted fusion on the multiple local fitness indexes with the multiple modification influence weights to generate the fused modification adaptation index.

[0086] Furthermore, the dual overload and overlimit detection module 14 is also used to perform the following steps: Based on the axle weight detection data set, each local axle weight data is extracted; the each local axle weight data is added and calculated to generate the overall weight data; if the unauthorized re-modification judgment result does not include the unauthorized re-modification mark, the local axle weight over-limit threshold and the overall weight over-limit threshold are determined according to the modification reporting characteristics; the each local axle weight data and the overall weight data are judged to be over-limited by using the local axle weight over-limit threshold and the overall weight over-limit threshold to generate the first detection result.

[0087] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0089] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A non-stop overload and overlimit detection method, characterized in that: include: When the vehicle to be inspected drives to the preset overload and over-limit detection area, the preset multi-source sensor detection generates license plate information, axle weight detection data set and 3D point cloud scanning data set; Compare the license plate information in the cloud-based vehicle modification registration database to determine whether the vehicle to be inspected has been registered for vehicle modification; If the vehicle to be inspected has been reported for vehicle modification, the modification report features of the vehicle to be inspected are extracted from the cloud-based vehicle modification report database, and the vehicle self-weight unauthorized re-modification is identified in combination with the 3D point cloud scanning data set to generate an unauthorized re-modification judgment result; Performing a dual overload and overlimit detection of the local axle weight and the overall weight according to the unauthorized re-modification judgment result and the axle weight detection data set to generate a first detection result; Performing volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result; The vehicle to be inspected is marked as overloaded and over-limit based on the first detection result and the second detection result.

2. The non-stop overload and overlimit detection method according to claim 1, characterized in that: If the vehicle to be inspected has been reported for vehicle modification, the modification report features of the vehicle to be inspected are extracted from the cloud-based vehicle modification report database, and the vehicle self-weight unauthorized re-modification is identified in combination with the 3D point cloud scanning data set to generate an unauthorized re-modification judgment result, including: Mapping and segmenting the modification report features according to the modification parts to generate multiple modification parameters for the multiple modification parts; Based on the multiple modified parts, segment and model the point cloud data of the 3D point cloud scanning data set to generate multiple local models; The multiple local models and the multiple modification parameters are used to identify the unauthorized modification of the vehicle's own weight, and generate the unauthorized modification judgment result.

3. The non-stop overload and overlimit detection method according to claim 2, characterized in that: Using the multiple local models to identify the unauthorized re-modification of the vehicle's own weight, and generating the unauthorized re-modification judgment result, includes: Performing adaptation analysis and fusing the multiple modification parameters and the multiple local models to generate a fused modification adaptation index; If the fused modification adaptation index is greater than the preset adaptation index, the vehicle to be inspected is marked as unauthorized re-modification, and the unauthorized re-modification judgment result is generated.

4. The non-stop overload and overlimit detection method according to claim 3, characterized in that: Performing adaptation analysis and fusing the multiple modification parameters and the multiple local models to generate a fusion modification adaptation index includes: extracting a plurality of structural modification parameters from the plurality of modification parameters, performing adaptation analysis on the plurality of local models and the plurality of structural modification parameters, and generating a plurality of local adaptation indexes; Extracting a plurality of modification weight gain parameters from the plurality of modification parameters, and configuring a plurality of modification influence weights for the plurality of modification parts; The multiple local fitness indicators are weightedly fused using the multiple modification influence weights to generate the fused modification adaptation indicator.

5. The non-stop overload and overlimit detection method according to claim 1, characterized in that: Performing dual overload and overlimit detection of local axle weight and overall weight according to the unauthorized re-modification judgment result and the axle weight detection data set to generate a first detection result, including: Extracting each local axle weight data based on the axle weight detection data set; Adding and calculating the local axle weight data to generate overall weight data; If the unauthorized re-modification judgment result does not include the unauthorized re-modification mark, determining the local axle weight over-limit threshold and the overall weight over-limit threshold according to the modification report characteristics; The local axle weight over-limit threshold and the overall weight over-limit threshold are used to perform over-limit judgment on each local axle weight data and the overall weight data to generate the first detection result.

6. The non-stop overload and overlimit detection method according to claim 1, characterized in that: Determining whether the vehicle to be inspected has been reported for vehicle modification also includes: If the vehicle to be inspected has not been reported for vehicle modification, the vehicle type is matched in the vehicle registration information database according to the license plate information; Determine the local axle weight overlimit threshold and the overall weight overlimit threshold corresponding to the vehicle type when not modified, and perform dual overload and overlimit detection of the local axle weight and the overall weight in combination with the axle weight detection data set; The corresponding cargo volume overlimit threshold when the vehicle is not modified is determined based on the vehicle type, and volume overload and overlimit detection is performed in combination with the 3D point cloud scanning data set.

7. The non-stop overload and overlimit detection method according to claim 1, characterized in that: The multi-source sensor includes a smart camera, a laser radar and an embedded axle load sensor, and the embedded axle load sensor includes one or more of a quartz sensor, a piezoelectric sensor and a fiber grating sensor. The specific configuration steps of the embedded axle load sensor include: Collecting the road surface smoothness of the preset overload and overlimit detection area; If the road surface flatness is less than or equal to a preset flatness threshold, the embedded axle weight sensor is constructed using a quartz sensor and a piezoelectric sensor, wherein the preset flatness threshold is the road surface flatness with an error less than the error threshold obtained by analyzing the axle weight measurement error through historical data.

8. The non-stop overload and overlimit detection method according to claim 7, characterized in that: Also includes: If the road surface flatness is greater than a preset flatness threshold, constructing the embedded axle load sensor with a quartz sensor, a piezoelectric sensor and a fiber grating sensor, and configuring an error compensation network based on tire contact stress; The error compensation network is constructed by learning the relationship between tire contact stress and axle load using historical tire contact stress data and historical axle measurement data.

9. Non-stop overload and overlimit detection system, characterized by: The system comprises: A multi-source sensor detection module, which is used to generate license plate information, axle weight detection data set and 3D point cloud scanning data set through preset multi-source sensor detection when the vehicle to be inspected drives into a preset overload and overlimit detection area; A vehicle modification information matching module, wherein the vehicle modification information matching module is used to compare the license plate information with the cloud vehicle modification reporting database to determine whether the vehicle to be inspected has been reported for vehicle modification; An unauthorized re-modification identification module is used to extract the modification report features of the vehicle to be inspected from the cloud-based vehicle modification report database if the vehicle to be inspected has been reported for vehicle modification, and to identify the unauthorized re-modification of the vehicle's own weight in combination with the 3D point cloud scanning data set to generate an unauthorized re-modification judgment result; A double overload and overlimit detection module, the double overload and overlimit detection module is used to perform a double overload and overlimit detection of the local axle weight and the overall weight according to the unauthorized refitting judgment result and the axle weight detection data set to generate a first detection result; A volume overload and overlimit detection module, wherein the volume overload and overlimit detection module is used to perform volume overload and overlimit detection based on the 3D point cloud scanning data set to generate a second detection result; An overload and overlimit marking module is used to mark the vehicle to be inspected as overloaded and overlimit based on the first detection result and the second detection result.

Images