ETC license plate number information comparison system

Abstract

The invention provides an ETC license plate number information comparison system, and relates to the technical field of intelligent transportation, and the system comprises a dual-mode sensing layer which is disposed on a toll lane and is provided with an ETC antenna module and an image collection and license plate recognition module; the ETC antenna module is used for communicating with a vehicle-mounted OBU to obtain ETC transaction data; the image acquisition and license plate recognition module is used for capturing a vehicle image and performing license plate recognition; the edge calculation layer is deployed with a heterogeneous calculation platform and is used for running a license plate recognition algorithm in real time; the application service layer is used for executing a lane control instruction, recording a transaction log, managing an alarm event and carrying out system configuration according to the comparison result; and a hardware-level time synchronization unit, through hardware-level time synchronization and edge calculation, information comparison and risk judgment are completed at the moment of vehicle passing, barrier gate control, alarm and other interception measures are executed immediately, and a traditional passive mode of post-auditing is changed.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to an ETC license plate number information comparison system. Background Technology

[0002] The Electronic Toll Collection (ETC) system for highways has been widely used. It completes transactions through wireless communication between the on-board unit (OBU) and the roadside unit (RSU), which greatly improves traffic efficiency. However, traditional ETC systems primarily rely on vehicle information (such as license plate number and vehicle model) stored in the OBU for identity authentication, which poses certain security risks. For example, criminals may use illegal OBUs or impersonate others' license plate OBUs to evade tolls or engage in illegal activities. Currently, some toll stations use independent video license plate recognition systems to assist in inspection, but ETC transaction data and video recognition data are usually processed independently in the backend and compared manually, which presents the following problems: Poor real-time performance: There is a delay in comparison between manual or back-end systems, making it impossible to complete risk assessment and take interception measures the instant the vehicle passes through; Insufficient accuracy: In complex scenarios such as nighttime, inclement weather, damaged license plates, or steep tilt, the robustness of a single video recognition algorithm is limited, which can easily lead to misjudgment or missed judgment. The system has low coupling: ETC transactions and video recognition lack hardware-level guarantees in terms of time synchronization and data association, resulting in low confidence in the comparison results; Lack of proactive defense: The system is mostly based on post-event audits and cannot provide real-time early warning and coordinated control for behaviors such as card swapping, card switching, and OBU anomalies. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and design a system that can compare ETC transaction information with real-time vehicle information in real time, accurately and automatically, so as to improve the security and reliability of the toll collection system.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: An ETC license plate number information comparison system includes interconnected communication links: The dual-mode perception layer, deployed in the toll lane, is equipped with an ETC antenna module and an image acquisition and license plate recognition module; the ETC antenna module is used to communicate with the on-board unit (OBU) to obtain ETC transaction data; the image acquisition and license plate recognition module is used to capture vehicle images and perform license plate recognition. The edge computing layer is equipped with a heterogeneous computing platform, which is used to run the license plate recognition algorithm in real time and compare the license plate information in the ETC transaction data with the license plate recognition results based on a hierarchical comparison strategy. The application service layer is used to execute lane control commands, record transaction logs, manage alarm events, and perform system configuration based on the comparison results. In addition, a hardware-level time synchronization unit is provided to achieve time synchronization between the ETC antenna module and the image acquisition and license plate recognition module based on the Precise Clock Protocol (PTP); wherein the timestamp error after synchronization is no greater than 1 millisecond.

[0005] Furthermore, the ETC antenna module supports multi-protocol compatible communication based on software-defined radio (SDR) and uses adaptive frequency hopping (FHSS) technology for anti-interference communication in the 5.7GHz to 5.9GHz frequency band; The image acquisition and license plate recognition module integrates a visible light camera and an infrared camera, adopts dynamic exposure control, and is equipped with a 3D structured light projector to assist in image acquisition. The visible light camera and the infrared camera work together to recognize license plates under different lighting conditions.

[0006] Furthermore, the heterogeneous computing platform of the edge computing layer adopts an architecture combining CPU and GPU, or an architecture combining CPU and FPGA. The heterogeneous computing platform is configured to assign video decoding, license plate recognition, and data comparison tasks to corresponding dedicated computing cores for execution; and to use the Tensor Core of the GPU or the parallel computing unit of the FPGA to accelerate the inference of the license plate recognition model. The license plate recognition model is subjected to INT8 quantization and compression processing.

[0007] Furthermore, the hierarchical comparison strategy includes: Level 1 comparison: Perform a complete string match between the license plate number in the ETC transaction data and the license plate number identified by the video; Secondary comparison: When the primary comparison is successful, further consistency verification is performed on the vehicle color and / or model information; Level 3 verification: Initiated when Level 1 or Level 2 verification fails, performing at least one of the following verifications: a) Verify whether the difference between the ETC transaction timestamp and the vehicle image capture timestamp is within a preset allowable time window; b) Retrieve image data from cameras in adjacent lanes to cross-verify the vehicle's driving trajectory.

[0008] Furthermore, the application service layer includes an anomaly alarm system, configured as follows: In response to the output of the hierarchical comparison strategy, an alarm operation of the corresponding level is executed; Among them, the first-level alarm corresponds to the license plate number mismatch, and the operation includes controlling the gate to close, activating the audible and visual alarm and notifying the on-duty personnel; A level 2 alarm corresponds to a license plate number that matches but the vehicle color or model does not. The operation includes recording an anomaly log, pushing information to the monitoring center, and activating the cameras in adjacent lanes for coordinated tracking. A level 3 alarm corresponds to an abnormal OBU device status reported by the ETC antenna module. The operation includes marking the corresponding vehicle as a subject of interest and triggering a secondary verification process in its subsequent passage events.

[0009] Furthermore, the system also includes a security hardening module, which is configured to perform the following operations: Secure data transmission: ETC transaction data is encrypted using the SM4 national cryptographic algorithm, and video stream data is transmitted with channel encryption using the TLS protocol; Node authentication: A two-way TLS authentication mechanism based on digital certificates is implemented between the edge computing layer and the application service layer; Access control and auditing: A role-based access control model is used to manage user operation permissions in the system and to generate complete operation audit logs for all critical system operations.

[0010] Furthermore, the application service layer provides a remote configuration management interface, which supports visual configuration of license plate recognition confidence threshold, hierarchical comparison rules, and special access list; The system supports hot deployment updates of configuration rules. The response time for configuration updates to take effect within the application service layer is shorter than a preset time threshold. It also supports rollback operations of configuration versions and historical version tracing functions.

[0011] Furthermore, the edge computing layer adopts a pipelined processing architecture to perform pipelined parallel processing on the three stages of vehicle image capture, license plate recognition, and comparison of ETC and video data. The edge computing layer is also configured with a data preloading mechanism, which is configured to control the image acquisition and license plate recognition module to start the image pre-capture process before the ETC antenna module detects that a vehicle has entered the transaction area and triggers the transaction.

[0012] Furthermore, the image acquisition and license plate recognition module is installed in front of the ETC antenna module along the vehicle's driving direction, and maintains a preset physical distance from the ETC antenna module. The installation height and optical axis tilt angle of the image acquisition and license plate recognition module are optimized so that its core imaging area covers the vehicle transaction area of ​​the ETC antenna module.

[0013] Compared with the prior art, the beneficial effects of the present invention are: It achieves real-time proactive prevention and control: through hardware-level time synchronization and edge computing, information comparison and risk assessment are completed the instant a vehicle passes through, and interception measures such as gate control and alarms are immediately executed, changing the traditional passive mode of post-event auditing.

[0014] Improved accuracy and robustness of identification: By integrating ETC transaction data with video recognition results and using a hierarchical comparison strategy to perform multi-dimensional cross-verification of license plate, vehicle type, timestamp, and driving trajectory, it can effectively cope with complex scenarios and accurately identify behaviors such as license plate cloning and OBU anomalies.

[0015] A deeply collaborative integrated system was built: from hardware synchronization and installation layout to edge computing, it ensures a high degree of consistency of data in time, space and logic, and solves the problems of low system coupling and low comparison confidence in traditional methods.

[0016] It ensures system security and manageability: System security is ensured through data transmission encryption, two-way authentication and a robust access control mechanism, and remote visual configuration and hot updates are supported, making system operation and maintenance flexible and reliable. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] Example 1, by Figure 1 As can be seen, the ETC license plate number information comparison system disclosed in this invention mainly consists of a dual-mode perception layer, an edge computing layer, an application service layer, and a hardware-level time synchronization unit that are interconnected. Each layer works together to achieve accurate comparison of ETC license plate number information and related functions. Specifically, dual-mode sensing layer Deployment location and module composition: Deployed in the toll lane, equipped with an ETC antenna module and an image acquisition and license plate recognition module.

[0020] The ETC antenna module is responsible for communicating with the on-board unit (OBU) to obtain ETC transaction data. This module supports multi-protocol compatible communication based on software-defined radio (SDR), which can adapt to ETC devices with different standards and protocols, enhancing system compatibility. At the same time, it adopts adaptive frequency hopping (FHSS) technology to perform anti-interference communication in the 5.7GHz to 5.9GHz frequency band, effectively avoiding external signal interference and ensuring the stability and reliability of communication.

[0021] Image acquisition and license plate recognition module functions: It is mainly used to capture vehicle images and perform license plate recognition. This module integrates a visible light camera and an infrared camera, and adopts dynamic exposure control technology, which can automatically adjust the exposure parameters according to different lighting conditions to ensure image quality. In addition, it is also equipped with a 3D structured light projector to assist in image acquisition. By projecting structured light, it obtains the depth information of the vehicle, further improving the accuracy and completeness of image acquisition. The visible light camera and the infrared camera work together to accurately recognize license plates under different lighting conditions (such as strong light during the day and weak light at night), improving the system's adaptability and recognition accuracy.

[0022] Edge computing layer Deployment and Functionality: A heterogeneous computing platform is deployed to run the license plate recognition algorithm in real time and compare the license plate information in ETC transaction data with the license plate recognition results based on a hierarchical comparison strategy. The heterogeneous computing platform can give full play to the advantages of different computing cores and improve computing efficiency and processing capabilities.

[0023] Heterogeneous computing platform architecture: It can adopt an architecture combining CPU and GPU or an architecture combining CPU and FPGA. This architecture design can allocate video decoding, license plate recognition and data comparison tasks to the corresponding dedicated computing cores for execution according to different task requirements. Among them, the license plate recognition model is processed by INT8 quantization compression, which reduces the computational load and storage space of the model while ensuring a certain recognition accuracy, thereby further improving the performance of the system.

[0024] Application service layer Function Overview: Based on the comparison results, it executes lane control commands, records transaction logs, manages alarm events, and performs system configuration. It is the core management layer of the system, responsible for coordinating and controlling the operation of the entire system.

[0025] Anomaly Alarm System: The application service layer includes an anomaly alarm system, configured to execute alarm operations of the corresponding level in response to the output results of the hierarchical comparison strategy.

[0026] Level 1 Alarm: This corresponds to a situation where the license plate number does not match. The operation includes controlling the gate to close and prevent the vehicle from passing; activating the audible and visual alarm to alert on-site staff; and notifying the on-duty personnel so that the abnormal situation can be handled in a timely manner.

[0027] Level 2 Alarm: This corresponds to situations where the license plate number matches but the vehicle color or model is inconsistent. The actions include recording an anomaly log to provide a basis for subsequent analysis and tracing; pushing information to the monitoring center so that management personnel can be informed of the anomaly in a timely manner; and activating adjacent lane cameras for linkage tracking to further confirm vehicle information.

[0028] Level 3 alarm: This corresponds to an abnormal OBU device status reported by the ETC antenna module. The operation includes marking the corresponding vehicle as a subject of interest and triggering a secondary verification process in its subsequent passage events to ensure the accuracy and safety of vehicle passage.

[0029] Hardware-level time synchronization unit The ETC antenna module and the image acquisition and license plate recognition module are synchronized using a Precise Time Protocol (PTP) to ensure time consistency between the two modules. The timestamp error after synchronization is no more than 1 millisecond, providing an accurate time reference for subsequent data comparison and avoiding comparison errors caused by time differences.

[0030] Example 2, as Figure 1 As shown, based on the detailed description of Embodiment 1 Further explanation of the edge computing layer Heterogeneous computing platform task allocation and accelerated inference: The heterogeneous computing platform is configured to allocate video decoding, license plate recognition, and data comparison tasks to corresponding dedicated computing cores for execution; for example, the CPU is responsible for the overall system control and scheduling, the GPU utilizes its powerful parallel computing capabilities for inference calculations of the license plate recognition model, and the FPGA can perform optimized processing for specific data processing tasks. Utilizing the GPU's Tensor Cores or the FPGA's parallel computing units to accelerate the inference of the license plate recognition model can significantly improve the speed of license plate recognition and meet the requirements of real-time processing.

[0031] Pipeline Processing Architecture: The edge computing layer adopts a pipelined processing architecture, which parallelizes the three stages of vehicle image capture, license plate recognition, and comparison of ETC and video data. This architecture design allows each processing stage to run simultaneously, reducing data processing waiting time and improving the overall processing efficiency and throughput of the system.

[0032] Data preloading mechanism: The edge computing layer is also equipped with a data preloading mechanism. Before the ETC antenna module detects a vehicle entering the transaction area and triggers the transaction, it controls the image acquisition and license plate recognition module to start the image pre-capture process. By acquiring vehicle images in advance, it prepares for subsequent license plate recognition and data comparison, further shortening the system response time and improving processing efficiency.

[0033] Detailed Explanation of the Hierarchical Comparison Strategy Level 1 comparison: Perform a complete string match between the license plate number in the ETC transaction data and the license plate number identified by the video. This is the most basic comparison method, quickly determining whether the license plate numbers match by directly comparing the strings.

[0034] Secondary comparison: When the primary comparison is successful, the consistency of vehicle color and / or model information is further verified; by comprehensively comparing information such as license plate number, vehicle color and model, the accuracy and reliability of the comparison are improved, and misjudgment caused by the same license plate number but inconsistent other vehicle information is avoided.

[0035] Level 3 comparison: Initiated when Level 1 or Level 2 comparison fails, performing at least the following verifications as described in this embodiment: Verify whether the difference between the ETC transaction timestamp and the vehicle image capture timestamp is within a preset allowable time window: By comparing the difference between the two timestamps, determine whether the ETC transaction and the vehicle image capture occurred within the same time period, avoiding comparison errors caused by excessive time differences.

[0036] Accessing image data from cameras in adjacent lanes allows for cross-verification of vehicle trajectories: By analyzing image data from cameras in adjacent lanes, the vehicle's trajectory can be understood, further confirming the vehicle's identity and traffic status, thus improving the accuracy and reliability of the comparison.

[0037] Application service layer further explains Security Hardening Module: The system also includes a security hardening module, configured to perform the following operations: Secure data transmission: ETC transaction data is encrypted using the SM4 national cryptographic algorithm to ensure the confidentiality and integrity of the transaction data during transmission; video stream data is transmitted with channel encryption using the TLS protocol to prevent video data from being stolen or tampered with during transmission.

[0038] Node authentication: A two-way TLS authentication mechanism based on digital certificates is implemented between the edge computing layer and the application service layer. By verifying the identity certificates of both parties, the legitimacy of the communicating parties is ensured, unauthorized nodes are prevented from accessing the system, and the security of the system is guaranteed.

[0039] Access Control and Auditing: A role-based access control model is adopted to manage user operation permissions in the system. Different operation permissions are assigned according to the user's role to ensure that users can only access data and functions within their authorized scope. Complete operation audit logs are generated for all critical system operations, recording information such as the time, user, and content of the operation, which facilitates subsequent auditing and traceability.

[0040] Remote Configuration Management Interface: The application service layer provides a remote configuration management interface, which supports visual configuration of license plate recognition confidence thresholds, hierarchical comparison rules, and special access lists. Users can easily adjust and configure system parameters through this interface without on-site operation, improving system management efficiency. The system supports hot deployment updates of configuration rules, with the response time for configuration updates to take effect within the application service layer being shorter than a preset time threshold. It also supports configuration version rollback and historical version tracing. When configuration rules need to be updated, real-time updates can be achieved without stopping system operation. At the same time, it is easy to roll back to previous configuration versions and trace historical configuration information, ensuring system stability and maintainability.

[0041] Dual-mode perception layer installation: Image acquisition and license plate recognition module installation location: The image acquisition and license plate recognition module is installed in front of the ETC antenna module along the vehicle's driving direction, and maintains a preset physical distance from the ETC antenna module; this installation location ensures that the image acquisition and license plate recognition module can capture the vehicle image in advance before the vehicle enters the ETC transaction area, providing sufficient time for subsequent comparison and processing.

[0042] Installation height and optical axis tilt angle optimization: The installation height and optical axis tilt angle of the image acquisition and license plate recognition module have been optimized to ensure that its core imaging area covers the vehicle transaction area of ​​the ETC antenna module. By optimizing the installation height and tilt angle, the quality and accuracy of image acquisition are ensured, enabling the license plate recognition module to clearly capture the vehicle's license plate information and improving the overall performance of the system.

[0043] In summary, the ETC license plate number information comparison system of the present invention achieves accurate comparison, efficient processing and security of ETC license plate number information through the collaborative work of each layer and the comprehensive application of multiple technologies. It has high accuracy, reliability and adaptability and can meet the needs of modern highway toll collection systems.

[0044] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0045] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An ETC license plate number information comparison system, characterized in that, Including interconnected communication connections: The dual-mode perception layer, deployed in the toll lane, is equipped with an ETC antenna module and an image acquisition and license plate recognition module; the ETC antenna module is used to communicate with the on-board unit (OBU) to obtain ETC transaction data; the image acquisition and license plate recognition module is used to capture vehicle images and perform license plate recognition. The edge computing layer is equipped with a heterogeneous computing platform, which is used to run the license plate recognition algorithm in real time and compare the license plate information in the ETC transaction data with the license plate recognition results based on a hierarchical comparison strategy. The application service layer is used to execute lane control commands, record transaction logs, manage alarm events, and perform system configuration based on the comparison results. In addition, a hardware-level time synchronization unit is provided to achieve time synchronization between the ETC antenna module and the image acquisition and license plate recognition module based on the Precise Clock Protocol (PTP); wherein the timestamp error after synchronization is no greater than 1 millisecond.

2. The ETC license plate number information comparison system according to claim 1, characterized in that: The ETC antenna module supports multi-protocol compatible communication based on software-defined radio (SDR) and uses adaptive frequency hopping (FHSS) technology for anti-interference communication in the 5.7GHz to 5.9GHz frequency band. The image acquisition and license plate recognition module integrates a visible light camera and an infrared camera, adopts dynamic exposure control, and is equipped with a 3D structured light projector to assist in image acquisition. The visible light camera and the infrared camera work together to recognize license plates under different lighting conditions.

3. The ETC license plate number information comparison system according to claim 1, characterized in that: The heterogeneous computing platform of the edge computing layer adopts an architecture combining CPU and GPU, or an architecture combining CPU and FPGA. The heterogeneous computing platform is configured to assign video decoding, license plate recognition, and data comparison tasks to corresponding dedicated computing cores for execution; and to use the Tensor Core of the GPU or the parallel computing unit of the FPGA to accelerate the inference of the license plate recognition model. The license plate recognition model is subjected to INT8 quantization and compression processing.

4. The ETC license plate number information comparison system according to claim 1, characterized in that: The hierarchical comparison strategy includes: Level 1 comparison: Perform a complete string match between the license plate number in the ETC transaction data and the license plate number identified by the video; Secondary comparison: When the primary comparison is successful, further consistency verification is performed on the vehicle color and / or model information; Level 3 verification: Initiated when Level 1 or Level 2 verification fails, performing at least one of the following verifications: a) Verify whether the difference between the ETC transaction timestamp and the vehicle image capture timestamp is within a preset allowable time window; b) Retrieve image data from cameras in adjacent lanes to cross-verify the vehicle's trajectory.

5. An ETC license plate number information comparison system according to claim 1 or 4, characterized in that: The application service layer includes an anomaly alarm system, configured as follows: In response to the output of the hierarchical comparison strategy, an alarm operation of the corresponding level is executed; Among them, the first-level alarm corresponds to the license plate number mismatch, and the operation includes controlling the gate to close, activating the audible and visual alarm and notifying the on-duty personnel; A level 2 alarm corresponds to a license plate number that matches but the vehicle color or model does not. The operation includes recording an anomaly log, pushing information to the monitoring center, and activating the cameras in adjacent lanes for coordinated tracking. A level 3 alarm corresponds to an abnormal OBU device status reported by the ETC antenna module. The operation includes marking the corresponding vehicle as a subject of interest and triggering a secondary verification process in its subsequent passage events.

6. The ETC license plate number information comparison system according to claim 1, characterized in that: The system also includes a security hardening module, which is configured to perform the following operations: (1) Secure data transmission: ETC transaction data is encrypted using the SM4 national cryptographic algorithm, and video stream data is transmitted with channel encryption using the TLS protocol; (2) Node identity authentication: A two-way TLS identity authentication mechanism based on digital certificates is implemented between the edge computing layer and the application service layer; (3) Access control and auditing: The role-based access control model is used to manage the user operation permissions of the system and generate complete operation audit logs for all critical system operations.

7. The ETC license plate number information comparison system according to claim 1, characterized in that: The application service layer provides a remote configuration management interface, which supports visual configuration of license plate recognition confidence threshold, hierarchical comparison rules, and special access list. The system supports hot deployment updates of configuration rules. The response time for configuration updates to take effect within the application service layer is shorter than a preset time threshold. It also supports rollback operations of configuration versions and historical version tracing functions.

8. The ETC license plate number information comparison system according to claim 1, characterized in that: The edge computing layer adopts a pipelined processing architecture, which performs pipelined parallel processing on the three stages of vehicle image capture, license plate recognition, and comparison of ETC and video data. The edge computing layer is also configured with a data preloading mechanism, which is configured to control the image acquisition and license plate recognition module to start the image pre-capture process before the ETC antenna module detects that a vehicle has entered the transaction area and triggers the transaction.

9. The ETC license plate number information comparison system according to claim 1, characterized in that: The image acquisition and license plate recognition module is installed in front of the ETC antenna module along the vehicle's driving direction, and maintains a preset physical distance from the ETC antenna module. The installation height and optical axis tilt angle of the image acquisition and license plate recognition module are optimized so that its core imaging area covers the vehicle transaction area of ​​the ETC antenna module.

Images