A truck scale electronic fence and vehicle position centering monitoring system and method

Abstract

The application discloses a kind of truck scale electronic fence and vehicle position centering monitoring system and method.The present application proposes closed-loop monitoring system by video monitoring module, truck scale measurement module and trusted evidence module.Video monitoring module is monitored by multimodal perception and dynamic electronic fence real-time vehicle position and intrusion behavior, generates vehicle position deviation information and warning signal;Truck scale measurement module receives deviation information, combines weighing data to carry out dynamic compensation or determine weighing validity, and outputs measurement result;Trusted evidence module binds warning signal, measurement result and positioning data, calibration log and stores evidence, and forms tamper-evident evidence chain.The present application realizes high-precision vehicle positioning, dynamic error compensation and whole-process trusted evidence, significantly improves the fairness of fuel into plant measurement, anti-fraud ability and automation level.

Description

Technical Field

[0001] This application relates to the field of remote measurement technology, and in particular to a safety system and method for monitoring the idle operation of equipment in a sample preparation room and linking it with an electronic fence. Background Technology

[0002] In the production and operation of thermal power plants, fuel acceptance upon arrival is a core component of the fuel management system. Fuel metering upon arrival directly impacts cost control and production safety. Currently, the industry widely uses truck scales (weighbridges) as the core equipment for fuel metering upon arrival. For vehicle supervision and data management in the truck scale metering process, existing technologies primarily employ a combination of manual supervision and basic electronic monitoring. Specifically, operators visually inspect whether vehicles are parked in the central area of ​​the truck scale weighing platform, manually record weighing data, and verify vehicle information. Some power plants also use cameras to monitor vehicle positions, supplemented by infrared beam detectors to determine if vehicles are fully on the scale. Simultaneously, weighing data, video monitoring, and transportation documents are stored in different systems, with metering processes and data traceability achieved through manual secondary verification and cross-system retrieval.

[0003] However, existing fuel metering solutions still face numerous technical challenges in practical applications: vehicle centering relies on manual experience and lacks high-precision automated positioning methods; electronic fences and positioning technologies are not accurate enough to detect covert fraudulent activities such as edge weighing and repeated weighing deviations; multi-source data, including weighing data, video footage, and vehicle information, are stored in different systems, lacking real-time linkage and cross-verification mechanisms, resulting in poor data coordination and low audit traceability efficiency; furthermore, manual guidance of vehicles for weighing adjustments is time-consuming, has low scheduling efficiency, and lacks automated and reliable evidence chains in the event of weighing disputes, leading to weak emergency response capabilities. These problems collectively result in existing metering processes failing to meet the refined and digitalized requirements of thermal power plant fuel management in terms of positioning accuracy, fraud prevention reliability, data fusion and coordination, and process automation. Summary of the Invention

[0004] To address the technical challenges in the existing fuel metering process at thermal power plants, such as large errors in manually judging vehicle centering deviations and difficulty in identifying violations like edge-weighting, as well as the inability to cross-verify data from multiple independent systems, this application provides a vehicle weighbridge electronic fence and vehicle position centering monitoring system and method.

[0005] To achieve the above objectives, this application provides the following technical solution: This application provides a vehicle scale electronic fence and vehicle position centering monitoring system, including: The video monitoring module is used to monitor the vehicle's position in real time and identify vehicle intrusion behavior when the vehicle enters the weighbridge area, and generate vehicle position deviation information and warning signals. The weighbridge metering module is communicatively connected to the monitoring module. It is used to receive the vehicle position offset information, perform dynamic compensation or determine the weighing validity based on the vehicle position offset, and output the metering result. The trusted evidence storage module is connected to both the monitoring module and the truck scale metering module. It is used to acquire the warning signals generated by the monitoring module and the metering results output by the truck scale metering module, and to bind and store the vehicle positioning data, weighing data and calibration logs.

[0006] Furthermore, the video surveillance module includes: The multimodal sensing module is used to acquire visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the truck scale area. The edge computing module is used to run a multi-target detection model based on the data collected by the multimodal perception module, output the target category, location coordinates and movement speed, and generate a dynamic electronic fence; The intelligent alarm module is used to determine a valid intrusion and trigger a tiered alarm when the centroid of a target enters the dynamic electronic fence area and stays there for more than a preset threshold.

[0007] Furthermore, the edge computing module includes: The laser calibration unit is used to obtain the physical dimensions of the truck scale and, in conjunction with the camera intrinsic parameter matrix, establish the perspective projection mapping relationship from the three-dimensional coordinate system to the two-dimensional image. The dynamic correction unit is used to periodically trigger the edge detection algorithm to identify the actual edge contour of the truck scale, dynamically adjust the virtual boundary of the electronic fence, and compensate for equipment deformation errors. The intrusion determination unit is used to detect the three-dimensional spatial intersection state between the target bounding box and the dynamic electronic fence in real time, and to perform hierarchical determination logic according to the target category.

[0008] Furthermore, the intelligent alarm module includes: A tiered response unit is used to execute differentiated responses based on the category of the intrusion target: If a non-vehicle target intrudes, a Level 1 alarm will be triggered, outputting an audible and visual warning as well as a platform pop-up notification; If a vehicle illegally enters the weighbridge, a level two alarm will be triggered, freezing the weighing process, activating the camera to record a close-up of the vehicle, and closing the entrance gate of the weighbridge. The anti-vandalism unit is used to embed the core algorithm into the FPGA chip and add physical mask detection function to the camera to identify occlusion or spraying damage.

[0009] Furthermore, the truck scale measuring module includes: A high-precision vision unit is used to acquire full-angle contour images of the vehicle and extract the overall vehicle mask and independent bounding boxes of the front and rear of the vehicle through an instance segmentation model. A laser positioning auxiliary unit is used to project a bright baseline along the edge of the truck scale to mark the physical boundary; The dynamic calibration engine is used to construct a three-dimensional spatial coordinate system based on the actual size of the truck scale, and to map the physical size to the pixel space of the video screen through perspective transformation to generate a virtual outline. The offset quantization analysis module is used to calculate the proportion of non-overlapping areas between the overall vehicle mask and the virtual outline, as well as the lateral distance difference between the center points of the vehicle front and rear boundary frames and the center line of the weighbridge, and to generate the vehicle position offset.

[0010] Furthermore, the truck scale measuring module also includes: The dynamic compensation module is used to adjust the confidence level of the weighing sensor data according to the vehicle offset rate. When the offset rate exceeds the first preset threshold, the compensation algorithm is activated to eliminate the weight deviation caused by the position tilt. The multi-condition linkage module is used to determine the weighing status based on the offset rate range: If the offset rate is less than or equal to the second preset threshold, it is determined to be in compliance status, and the weighing data is directly output. If the offset rate is greater than the second preset threshold and less than or equal to the third preset threshold, it is determined to be a warning state, the calibrated data is to be reviewed and a guidance prompt is triggered; If the offset rate is greater than the third preset threshold or wheel pressing is detected, it is determined to be in a failure state, the current measurement value is locked and the manual review process is initiated.

[0011] Furthermore, the trusted evidence storage module includes: The data encapsulation unit is used to package vehicle positioning screenshots, weighing curves, calibration logs, and operator biometric hash values ​​into indivisible data units and encrypt them using national cryptographic algorithms. The blockchain evidence storage unit is used to write the hash value of the data unit into the consortium blockchain node in real time and simultaneously generate an evidence storage certificate containing timestamp, geographical location and device code. The tamper-proof unit is used to embed invisible digital watermarks in video evidence and to set up a physical tamper-proof design in the critical data storage medium, which triggers self-destruction when unauthorized access is made.

[0012] Furthermore, the video monitoring module and the truck scale measuring module achieve clock synchronization through a time-sensitive network, ensuring millisecond-level alignment between the weight data and the vehicle positioning screenshot.

[0013] This application also provides a method for monitoring the electronic fence and vehicle position centering of a truck scale, used in the aforementioned electronic fence and vehicle position centering monitoring device, comprising the following steps: Using the video monitoring module, the vehicle position is monitored in real time when the vehicle enters the weighbridge area and the vehicle intrusion behavior is identified, generating vehicle position deviation information and warning signals. The weighbridge module receives vehicle position offset information generated by the video monitoring module, performs dynamic compensation or determines the validity of weighing based on the vehicle position offset, and outputs the measurement result. Using the trusted evidence storage module, the warning signal generated by the video monitoring module and the measurement result output by the truck scale measurement module are obtained, and the vehicle positioning data, weighing data and calibration log are bound and stored together.

[0014] Furthermore, the generation of vehicle position offset information and alarm signals specifically includes: Collect visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the truck scale area; The multi-target detection model is run based on the collected data, and the target category, location coordinates and movement speed are output, and a dynamic electronic fence is generated. When the centroid of a target enters the dynamic electronic fence area and stays there for more than a preset threshold, it is determined to be a valid intrusion, and a graded alarm is triggered according to the category of the intruding target.

[0015] Compared with the prior art, the beneficial effects of this application are as follows: This application provides an intelligent monitoring device for fuel delivery metering. It features a video monitoring module to monitor vehicle location in real time and identify intrusion, a weighbridge module to dynamically compensate for weighing data based on vehicle position offset or to determine weighing validity, and a trusted evidence storage module to bind and store vehicle positioning data, weighing data, and calibration logs, forming a closed loop of monitoring, metering, and evidence storage. This device solves the problems of large errors in manual vehicle centering determination, difficulty in identifying fraudulent behaviors such as edge-weighted weighing, and the inability to cross-verify isolated data from multiple systems in existing technologies. It achieves high-precision vehicle positioning, dynamic compensation to eliminate metering errors caused by position deviations, and ensures data immutability through blockchain and national cryptographic algorithms, significantly improving the fairness, anti-fraud capabilities, and data credibility of fuel delivery metering.

[0016] This application provides an intelligent monitoring method for fuel intake metering. It utilizes a video monitoring module to generate vehicle position deviation information and warning signals; a weighbridge module to dynamically compensate for or determine weighing validity based on the deviation and output metering results; and a trusted evidence storage module to bind and store vehicle positioning data, weighing data, and calibration logs, achieving fully automated and intelligent metering control. This method upgrades traditional manual vehicle centering determination to real-time quantitative analysis using machine vision, passive video review to proactive hierarchical alarms and process freezing, and distributed storage to integrated blockchain evidence storage. This significantly reduces vehicle centering time, greatly improves the efficiency of intercepting fraudulent activities, and significantly reduces controversial metering events. It also meets judicial-level evidence collection requirements, providing efficient, fair, and reliable technical support for fuel intake metering in thermal power plants. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the electronic fence and vehicle position centering monitoring system for truck scales in this application; Figure 2 This is a flowchart of the electronic fence and vehicle position centering monitoring method for truck scales in this application; Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In one embodiment of this application, such as Figure 1 As shown, an electronic fence and vehicle position centering monitoring device for a truck scale is provided. The system includes: a video monitoring module, a truck scale measurement module, and a trusted evidence storage module. The trusted evidence storage module is communicatively connected to both the video monitoring module and the truck scale measurement module.

[0021] Specifically, the video monitoring module is used to monitor the vehicle's position in real time and identify intrusion behavior when a vehicle enters the weighbridge area, generating vehicle position deviation information and warning signals. The video monitoring module includes a multimodal perception module, an edge computing module, and an intelligent alarm module. The multimodal perception module is used to collect visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the weighbridge area. As one embodiment, the multimodal perception module includes: visible light and thermal imaging dual-mode cameras (multispectral vision devices) deployed around the weighbridge; visible light is used for high-resolution detail capture, and thermal imaging is used for personnel and object identification at night or in inclement weather; and millimeter-wave radar added in the visual blind spot for real-time scanning of dynamic targets in the weighbridge area, fusing with camera data to improve detection reliability. In addition, a laser contour calibration device is installed along the edge of the weighbridge to dynamically project high-precision electronic fence boundary lines, assisting drivers and supervisors in intuitively identifying restricted areas.

[0022] The edge computing module is used to run a multi-target detection model based on the data collected by the multimodal perception module, output the target category, location coordinates, and movement speed, and generate a dynamic electronic fence. As one embodiment, the edge computing module is an embedded AI analysis terminal equipped with a multi-target detection model (an improved version of YOLOv8), supporting parallel processing of video streams and radar point cloud data to achieve classification and recognition of people, vehicles, and objects, and tracking of their movement trajectories. Specifically, the edge computing module further includes: a laser calibration unit, a dynamic correction unit, and an intrusion determination unit. The laser calibration unit is used to obtain the physical dimensions of the truck scale and, in conjunction with the camera intrinsic parameter matrix, establish a perspective projection mapping relationship from the three-dimensional coordinate system to the two-dimensional image. The dynamic correction unit is used to periodically trigger the edge detection algorithm, identify the actual edge contour of the truck scale, dynamically adjust the virtual boundary of the electronic fence, and compensate for equipment deformation errors. Edge detection (Canny algorithm + Hough transform) is triggered every 1-5 seconds to identify the actual edge contour of the truck scale, dynamically adjust the virtual boundary of the electronic fence, and compensate for deformation errors caused by equipment vibration or thermal expansion and contraction. The intrusion determination unit is used to detect the three-dimensional spatial intersection state between the target bounding box and the dynamic electronic fence in real time, and execute hierarchical determination logic according to the target category. Specifically, the three-dimensional spatial intersection state between the target bounding box and the electronic fence is detected in real time. If the centroid of the target enters the fence area and stays for more than 2 seconds, it is determined to be a valid intrusion. For intrusion targets, a distinction is made between instantaneous crossing (such as birds flying) and continuous stay (such as illegally placing heavy objects), and a trajectory prediction algorithm (Kalman filter) is used to determine whether it is an intentional intrusion behavior.

[0023] The intelligent alarm module is used to determine a valid intrusion and trigger a tiered alarm when the centroid of a target enters the dynamic electronic fence area and stays there for more than a preset threshold. Specifically, the intelligent alarm module includes a tiered response unit and an anti-vandalism unit. The tiered response unit is used to execute differentiated responses based on the type of intrusion target. If a non-vehicle target intrudes, a level one alarm is triggered, outputting audible and visual warnings and pop-up prompts on relevant system platforms, linking factory broadcasts and emergency lighting equipment, activating cameras to record personnel information, and analyzing the specific target type, such as human or non-human targets. If a vehicle illegally enters, a level two alarm is triggered, freezing the weighing process, activating cameras to record close-ups of wheels and license plates, and closing the weighbridge entrance gate. The three-dimensional alarm output includes: directional sound wave speakers and LED warning screens installed around the weighbridge, with voice announcements of the violation type and a screen displaying a schematic diagram of the intrusion target's location; support for pushing captured images, real-time location, and target type of the intrusion target to inspection personnel via a mobile APP, and activation of drones for aerial inspection of key areas. The anti-vandalism unit is used to embed the core algorithm into the FPGA chip and add physical mask detection function to the camera to identify occlusion or vandalism. Even in the event of a cyberattack or system crash, basic protection functions can still be maintained through hardware-level electronic fences.

[0024] Specifically, the weighbridge metering module receives vehicle position offset information generated by the video monitoring module, dynamically compensates for the weighing data based on the vehicle position offset, or determines the validity of the weighing, and outputs the metering result. The weighbridge metering module includes: a high-precision vision unit, a laser positioning auxiliary unit, a weighing sensor group, an edge intelligent computing layer, a dynamic calibration engine, an offset quantization analysis module, a dynamic compensation module, and a multi-condition linkage module. The high-precision vision unit acquires full-angle contour images of the vehicle and extracts the overall vehicle mask and independent bounding boxes for the front and rear of the vehicle through an instance segmentation model. As one embodiment, industrial-grade multispectral cameras are deployed on both sides and the top of the weighbridge, supporting 4K resolution and HDR imaging, covering the full-angle contour of the vehicle. A wide-angle lens captures the overall position, while a telephoto lens focuses on details of the front / rear of the vehicle (license plate, wheels). An improved YOLOv8-seg model is employed, adding attention mechanism layers for the front (license plate, headlights) and rear (taillights, bumper) of the vehicle. This outputs a complete vehicle mask and independent bounding boxes for the front and rear. Morphological filtering eliminates interference such as shadows and water accumulation, accurately extracting the vehicle's contour pixel set. The laser positioning auxiliary unit projects a bright baseline along the edge of the weighbridge to calibrate the physical boundaries. A laser scanning device is installed along the edge of the weighbridge to project a bright baseline in real time, assisting the vision system in quickly calibrating the physical boundaries.

[0025] The weighing sensor group integrates a high-precision weighing module, synchronized with the vision system clock, achieving millisecond-level alignment of weight data and image frames. The edge intelligent computing layer is equipped with a customized YOLOv8-seg model, enabling parallel computation of vehicle segmentation, front / rear vehicle detection, and contour matching, with a processing latency of ≤300ms. The dynamic calibration engine is used to construct a three-dimensional spatial coordinate system based on the actual dimensions of the truck scale, mapping the physical dimensions to the pixel space of the video image through perspective transformation to generate a virtual outline. A laser scanning device is used to acquire the physical boundary point cloud data of the truck scale, and a virtual outline matching the camera's viewpoint is generated through a calibration algorithm. Contour edge detection (Canny algorithm + sub-pixel interpolation) is triggered every 1-5 seconds to dynamically correct boundary offsets caused by thermal expansion and contraction or mechanical deformation. The offset quantization analysis module is used to calculate the proportion of non-overlapping areas between the overall vehicle mask and the virtual outline, as well as the lateral distance difference between the center points of the front and rear boundary frames and the center line of the truck scale, generating the vehicle position offset. Specifically, the intersection and union regions of the vehicle mask and the weighbridge outline are calculated. The percentage of non-overlapping areas is calculated using pixel differences: Offset rate = (Union area - Intersection area) / Union area × 100%. If the percentage of non-overlapping areas on one side is greater than 5%, it is considered a risk of wheel edge pressing. The coordinates of the center points of the vehicle's front and rear bounding boxes are extracted, and the lateral distance difference (Δd) between them and the weighbridge centerline is calculated. The area ratio of the front and rear masks is compared simultaneously. If it exceeds the range of 0.9 to 1.1 or Δd is greater than 10 pixels, it is considered a center of gravity shift. The dynamic compensation module is used to adjust the confidence level of the weighing sensor data according to the vehicle offset rate. When the offset rate exceeds a first preset threshold, the compensation algorithm is activated to eliminate weight deviation caused by position tilt. The confidence level of the weighing sensor data is adjusted according to the vehicle offset rate. When the offset rate is greater than 2%, the compensation algorithm is activated to eliminate weight deviation caused by position tilt. The multi-condition linkage module is used to determine the weighing status based on the offset rate range: if the offset rate is less than or equal to the second preset threshold (e.g., 1%), it is determined to be compliant, and the weighing data is directly output, while simultaneously saving a screenshot of the vehicle's positioning; if the offset rate is greater than the second preset threshold but less than or equal to the third preset threshold (e.g., 1% < offset rate ≤ 3%), it is determined to be in a warning state, the calibration data is set to await review, and a guidance prompt is triggered (a green arrow is displayed on the LED screen indicating the adjustment direction, and a voice announcement is made: "Please park in the center"); if the offset rate is greater than the third preset threshold (e.g., > 3%) or wheel pressing against the edge is detected, it is determined to be in a failure state, the current measurement value is locked, and a manual review process is initiated (a red warning light flashes, the barrier gate automatically closes to prevent the vehicle from leaving, and close-up images of the front, rear, and wheels are captured simultaneously, immediately interrupting the upload of measurement data to the enterprise ERP system). When a cheating event is triggered, the intrusion event is bound to the weighing record and the vehicle's RFID tag, marking high-risk transport vehicles and restricting their subsequent access permissions.

[0026] Specifically, the trusted evidence storage module is used to acquire the warning signals generated by the video monitoring module and the measurement results output by the truck scale measuring module, and binds and stores the vehicle positioning data, weighing data, and calibration logs as evidence. The trusted evidence storage module includes: a data encapsulation unit, a blockchain evidence storage unit, and an anti-tampering unit.

[0027] The data encapsulation unit packages vehicle location screenshots, weighing curves, calibration logs, and operator biometric hash values ​​into indivisible data units, which are then encrypted using national cryptographic algorithms. The vehicle location screenshots include sub-pixel-level edge annotations, and the operator biometrics include vein fingerprint hash values, all encrypted using the national cryptographic SM4 algorithm. The blockchain evidence storage unit writes the hash values ​​of these data units to the consortium blockchain node in real time, simultaneously generating an evidence storage certificate containing a timestamp, geographical location, and device code. The entire process from data generation to blockchain evidence storage takes less than 2 seconds, supporting judicial institutions in verifying the integrity of the original data using public keys. The anti-tampering unit embeds an invisible digital watermark (DWT-SVD algorithm) into the video evidence and incorporates a physical dismantling-resistant design in the critical data storage medium, triggering self-destruction upon unauthorized access. The critical data storage medium employs a physical dismantling-resistant design, triggering chip self-destruction upon any unauthorized access; the embedded invisible digital watermark in the video evidence distorts the watermark upon tampering.

[0028] The technical solution of this application will be explained below with reference to a specific application scenario.

[0029] At a thermal power plant, a coal truck drove into the weighbridge area during the fuel metering process.

[0030] As a vehicle enters, the laser contour calibration device automatically projects a bright electronic fence boundary line, while millimeter-wave radar simultaneously scans the vehicle's contour. A multispectral camera activates both wide-angle and telephoto dual-mode shooting to capture details of the front, rear, and wheels. An embedded AI terminal runs an improved YOLOv8-seg model to segment the vehicle mask and extract independent bounding boxes for the front and rear. The dynamic calibration engine uses a perspective transformation algorithm to map the physical boundaries scanned by the laser onto the video feed, generating a calibrated virtual contour box for the weighbridge. The percentage of non-overlapping pixels (offset rate) between the front / rear bounding boxes and the virtual contour box is calculated. If the offset rate on one side is greater than 5%, it is determined that the wheels are pressing against the edge. The difference in lateral distance between the center points of the front and rear of the vehicle and the center line of the weighbridge (Δd > 10 pixels) is used to determine if the vehicle's center of gravity is shifted, triggering an audible and visual warning (LED arrows guide adjustment). If the vehicle fails to adjust as prompted and continues to press against the edge for more than 5 seconds, the gate automatically locks, a close-up of the wheels pressing against the edge is captured, and the weighing process is frozen. Meanwhile, the FPGA chip embeds the core algorithm, so even if the system is attacked by network outage, it can still maintain the electronic fence boundary through laser projection to prevent people from covering the camera or tampering with the calibration parameters.

[0031] The weighing sensor group and vision system achieve millisecond-level clock synchronization via TSN, ensuring a strict correspondence between weight data and vehicle positioning screenshots. When the mask area ratio of the vehicle's front and rear exceeds the range of 0.9 to 1.1, the vehicle is judged to be tilted and a dynamic compensation algorithm is activated. The fluctuation value of the weighing data, combined with the offset rate (>2%), triggers a weight correction model to automatically eliminate errors caused by position deviations. In compliance status (offset rate ≤1%), the weight data is directly output, and the vehicle centering screenshot and calibration log are saved. In warning status (1% < offset rate ≤3%), the data is marked for review, adjustment instructions are broadcast via voice, and the weighing waiting time is extended. In failure status (offset rate >3%), the current measurement value is frozen, manual review is initiated, and the driver's identity is recorded (RFID or license plate recognition). If the same vehicle is detected to have three consecutive critical offsets (e.g., fluctuations of 1.8% to 2%), its subsequent entry permissions are automatically restricted, and historical data is pushed to the risk control platform.

[0032] Vehicle location screenshots (including subpixel-level edge annotations), weighing curves, calibration logs, and operator biometric data (vein fingerprint hash values) are packaged into indivisible data units and encrypted using the national cryptographic algorithm SM4. The hash values ​​of these data units are written to the consortium blockchain node in real time, simultaneously generating a certificate of evidence containing timestamps, geographical locations, and device codes, supporting judicial institutions in verifying the integrity of the original data via public keys. Key data storage media employs a physical dismantling-resistant design; any unauthorized access triggers chip self-destruction. Video evidence is embedded with an invisible digital watermark (DWT-SVD algorithm); tampering with it will distort the watermark. Through the event timeline backtracking function, the weighing process records of any batch of fuel (including vehicle adjustment times, compensation coefficients, and operator operation logs) can be retrieved with a single click, meeting the requirements for penetrating auditing.

[0033] This embodiment reduces the vehicle centering error from ±10cm (traditionally determined manually) to ±0.5cm, improving the efficiency of intercepting fraudulent behavior by 300%; reduces disputed measurement events by 90% and shortens the settlement cycle by 35%; achieves zero disputes in judicial evidence collection, and meets the anti-fraud certification requirements of GB / T7723-2017.

[0034] In summary, the electronic fence and vehicle position centering monitoring device for truck scales provided in this application, through the construction of an interactive system architecture consisting of a video monitoring module, a truck scale measurement module, and a trusted evidence storage module, and utilizing the core control logic of edge computing, dynamic compensation, and blockchain evidence storage, achieves high-precision monitoring of vehicle position, dynamic and reliable determination of weighing data, and tamper-proof evidence storage throughout the entire process. This solves the problems of low positioning accuracy, weak anti-fraud capabilities, data silos, and poor automation levels in the prior art, and has significant economic and social benefits.

[0035] Example 2 In one embodiment of this application, such as Figure 2 As shown, in conjunction with the above-mentioned electronic fence and vehicle position centering monitoring system for truck scales, this application also provides a method for monitoring the electronic fence and vehicle position centering of truck scales, including the following steps: Using the video monitoring module, the vehicle's position is monitored in real time when the vehicle enters the weighbridge area, and the vehicle's intrusion behavior is identified, generating vehicle position deviation information and warning signals.

[0036] Specifically, visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the weighbridge area are collected; a multi-target detection model is run based on the collected data to output the target category, location coordinates, and movement speed, and a dynamic electronic fence is generated; when the target's centroid enters the dynamic electronic fence area and stays for more than a preset threshold, it is determined to be a valid intrusion, and a graded alarm is triggered according to the category of the intruding target.

[0037] The weighbridge module receives vehicle position offset information generated by the video monitoring module, performs dynamic compensation or determines the validity of weighing based on the vehicle position offset, and outputs the measurement result.

[0038] Specifically, the system acquires a full-angle contour image of the vehicle and extracts the overall vehicle mask and independent bounding boxes for the front and rear of the vehicle using an instance segmentation model. A three-dimensional spatial coordinate system is constructed based on the actual dimensions of the weighbridge, and a virtual contour box is generated through perspective transformation. The proportion of non-overlapping areas between the overall vehicle mask and the virtual contour box, as well as the lateral distance difference between the center points of the front and rear bounding boxes and the center line of the weighbridge, are calculated to generate the vehicle position offset. The confidence level of the weighing sensor data is adjusted based on the vehicle offset rate: if the offset rate is less than or equal to a first preset threshold (e.g., 1%), it is considered compliant, and the weighing data is directly output; if the offset rate is greater than the first preset threshold but less than or equal to a second preset threshold (e.g., 1% < offset rate ≤ 3%), it is considered a warning state, the calibrated data is set to await review, and a guidance prompt is triggered; if the offset rate is greater than the second preset threshold (e.g., > 3%) or wheel pressing on the edge is detected, it is considered a failure state, the current measurement value is locked, and a manual review process is initiated; when the offset rate exceeds a third preset threshold (e.g., 2%), a compensation algorithm is activated to automatically eliminate measurement errors caused by position deviation.

[0039] Using the trusted evidence storage module, the warning signal generated by the video monitoring module and the measurement result output by the truck scale measurement module are obtained, and the vehicle positioning data, weighing data and calibration log are bound and stored together.

[0040] Specifically, vehicle location screenshots, weighing curves, calibration logs, and operator biometric hash values ​​are packaged into indivisible data units and encrypted using national cryptographic algorithms. The hash values ​​of these data units are written to the consortium blockchain nodes in real time, and a certificate of evidence containing timestamps, geographical locations, and device codes is generated synchronously. Invisible digital watermarks are embedded in video evidence, and anti-physical dismantling designs are set in key data storage media.

[0041] Example 3 In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method steps described in Embodiment 2 above.

[0042] Figure 3 An internal structural diagram of a computer device is shown in one embodiment. This computer device may specifically be a terminal or a server. Figure 3 As shown, the computer device includes a processor, memory, network interface, display, camera, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse. Those skilled in the art will understand that… Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computing devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0043] Example 4 In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the method steps described in Embodiment 2 above. The computer-readable storage medium can be any medium capable of storing data, including but not limited to: USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0044] Example 5 In one embodiment, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method steps described in Embodiment 2 above. The computer program product may be provided in the form of a software installation package, firmware, etc., and is suitable for loading into the computer device described in Embodiment 3 above.

[0045] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0046] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A vehicle weighbridge electronic fence and vehicle position centering monitoring system, characterized in that, include: The video monitoring module is used to monitor the vehicle's position in real time and identify vehicle intrusion behavior when the vehicle enters the weighbridge area, and generate vehicle position deviation information and warning signals. The weighbridge metering module is communicatively connected to the monitoring module. It is used to receive the vehicle position offset information, perform dynamic compensation or determine the weighing validity based on the vehicle position offset, and output the metering result. The trusted evidence storage module is connected to both the monitoring module and the truck scale metering module. It is used to acquire the warning signals generated by the monitoring module and the metering results output by the truck scale metering module, and to bind and store the vehicle positioning data, weighing data and calibration logs.

2. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 1, characterized in that, The video surveillance module includes: The multimodal sensing module is used to acquire visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the truck scale area. The edge computing module is used to run a multi-target detection model based on the data collected by the multimodal perception module, output the target category, location coordinates and movement speed, and generate a dynamic electronic fence; The intelligent alarm module is used to determine a valid intrusion and trigger a tiered alarm when the centroid of a target enters the dynamic electronic fence area and stays there for more than a preset threshold.

3. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 2, characterized in that, The edge computing module includes: The laser calibration unit is used to obtain the physical dimensions of the truck scale and, in conjunction with the camera intrinsic parameter matrix, establish the perspective projection mapping relationship from the three-dimensional coordinate system to the two-dimensional image. The dynamic correction unit is used to periodically trigger the edge detection algorithm to identify the actual edge contour of the truck scale, dynamically adjust the virtual boundary of the electronic fence, and compensate for equipment deformation errors. The intrusion determination unit is used to detect the three-dimensional spatial intersection state between the target bounding box and the dynamic electronic fence in real time, and to perform hierarchical determination logic according to the target category.

4. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 2, characterized in that, The intelligent alarm module includes: A tiered response unit is used to execute differentiated responses based on the category of the intrusion target: If a non-vehicle target intrudes, a Level 1 alarm will be triggered, outputting an audible and visual warning as well as a platform pop-up notification; If a vehicle illegally enters the weighbridge, a level two alarm will be triggered, freezing the weighing process, activating the camera to record a close-up of the vehicle, and closing the entrance gate of the weighbridge. The anti-vandalism unit is used to embed the core algorithm into the FPGA chip and add physical mask detection function to the camera to identify occlusion or spraying damage.

5. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 1, characterized in that, The truck scale measuring module includes: A high-precision vision unit is used to acquire full-angle contour images of the vehicle and extract the overall vehicle mask and independent bounding boxes of the front and rear of the vehicle through an instance segmentation model. A laser positioning auxiliary unit is used to project a bright baseline along the edge of the truck scale to mark the physical boundary; The dynamic calibration engine is used to construct a three-dimensional spatial coordinate system based on the actual size of the truck scale, and to map the physical size to the pixel space of the video screen through perspective transformation to generate a virtual outline. The offset quantization analysis module is used to calculate the proportion of non-overlapping areas between the overall vehicle mask and the virtual outline, as well as the lateral distance difference between the center points of the vehicle front and rear boundary frames and the center line of the weighbridge, and to generate the vehicle position offset.

6. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 5, characterized in that, The truck scale measuring module also includes: The dynamic compensation module is used to adjust the confidence level of the weighing sensor data according to the vehicle offset rate. When the offset rate exceeds the first preset threshold, the compensation algorithm is activated to eliminate the weight deviation caused by the position tilt. The multi-condition linkage module is used to determine the weighing status based on the offset rate range: If the offset rate is less than or equal to the second preset threshold, it is determined to be in compliance status, and the weighing data is directly output. If the offset rate is greater than the second preset threshold and less than or equal to the third preset threshold, it is determined to be a warning state, the calibrated data is to be reviewed and a guidance prompt is triggered; If the offset rate is greater than the third preset threshold or wheel pressing is detected, it is determined to be in a failure state, the current measurement value is locked and the manual review process is initiated.

7. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 1, characterized in that, The trusted evidence storage module includes: The data encapsulation unit is used to package vehicle positioning screenshots, weighing curves, calibration logs, and operator biometric hash values ​​into indivisible data units and encrypt them using national cryptographic algorithms. The blockchain evidence storage unit is used to write the hash value of the data unit into the consortium blockchain node in real time and simultaneously generate an evidence storage certificate containing timestamp, geographical location and device code. The tamper-proof unit is used to embed invisible digital watermarks in video evidence and to set up a physical tamper-proof design in the critical data storage medium, which triggers self-destruction when unauthorized access is made.

8. The electronic fence and vehicle position centering monitoring system for truck scales according to claim 1, characterized in that, The video monitoring module and the truck scale measuring module are synchronized via a time-sensitive network to ensure millisecond-level alignment of weight data with vehicle location screenshots.

9. A method for monitoring the electronic fence and vehicle position centering of a truck scale, characterized in that, The electronic fence and vehicle position centering monitoring device for a truck scale as described in claim 1 includes the following steps: Using the video monitoring module, the vehicle position is monitored in real time when the vehicle enters the weighbridge area and the vehicle intrusion behavior is identified, generating vehicle position deviation information and warning signals. The weighbridge module receives vehicle position offset information generated by the video monitoring module, performs dynamic compensation or determines the validity of weighing based on the vehicle position offset, and outputs the measurement result. Using the trusted evidence storage module, the warning signal generated by the video monitoring module and the measurement result output by the truck scale measurement module are obtained, and the vehicle positioning data, weighing data and calibration log are bound and stored together.

10. The method for monitoring the electronic fence and vehicle position centering of a truck scale according to claim 9, characterized in that, The generation of vehicle position offset information and alarm signals specifically includes: Collect visible light images, thermal imaging temperature distribution maps, and millimeter-wave radar point cloud data of the truck scale area; The multi-target detection model is run based on the collected data, and the target category, location coordinates and movement speed are output, and a dynamic electronic fence is generated. When the centroid of a target enters the dynamic electronic fence area and stays there for more than a preset threshold, it is determined to be a valid intrusion, and a graded alarm is triggered according to the category of the intruding target.

Images