Abnormal detection and instant release processing system applied to ETC (Electronic Toll Collection) transaction

Abstract

The invention discloses an abnormal detection and instant release processing system applied to ETC transaction, relates to the technical field of ETC transaction, and particularly relates to an abnormal detection and instant release processing system with dynamic environment adaptability. The problems that the license plate recognition efficiency is reduced and the real-time performance is insufficient due to microwave signal reflection, lane interaction interference and external environment influence in a multi-lane environment of a toll plaza can be effectively solved, the system obtains vehicle information through cooperation of a license plate recognition module and an ETC information reading module, and matching verification is carried out through a comparison module. Meanwhile, the transaction verification module is combined to verify the validity of the completed transaction, the anomaly detection module specially identifies abnormal vehicles not passing through the entrance, and the instant release module is combined to quickly release the vehicles which confirm that the transaction is completed, so that the problems of congestion and wrong fee deduction caused by wrong transactions are reduced; therefore, the transaction accuracy, the real-time performance and the passing efficiency of the charging system are obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ETC transactions, and in particular to a system for detecting anomalies in ETC transactions and instantly releasing the transaction. Background Art

[0002] With the implementation of a nationwide expressway network, ETC (Electronic Toll Collection System) has played an important role in improving traffic efficiency. However, in toll plazas, due to the reflection and interference of microwave signals, ETC equipment may make erroneous transactions with antennas in non-target lanes. Such erroneous transactions not only hinder vehicle traffic, but may also cause subsequent management chaos. Therefore, an effective technical solution is needed to achieve rapid detection and immediate processing of ETC transaction anomalies through the comprehensive use of license plate recognition and transaction verification mechanisms, so as to ensure the stability of the toll collection system and traffic efficiency.

[0003] The prior art has the following deficiencies: In a multi-lane toll plaza environment, due to the complex reflection of microwave signals and lane interaction interference, even if the license plate recognition and ETC transaction information can be compared to a certain extent, external factors such as light, rain or vehicle occlusion may still cause the license plate recognition efficiency to decrease, thus failing to meet real-time requirements. In addition, when an abnormal transaction occurs in a non-target lane, the system lacks a fast verification mechanism to determine and release the vehicle, which may cause congestion or even continuous problems of erroneous deductions. These problems require the design of an abnormal detection and immediate release processing system with dynamic environmental adaptability to minimize the impact of erroneous transactions. Summary of the invention

[0004] The purpose of the present invention is to provide a system for detecting anomalies in ETC transactions and instantly releasing the transaction, so as to solve the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solution: applied to an ETC transaction anomaly detection and instant release processing system, including a license plate recognition module, an ETC information reading module, a comparison module, a transaction verification module, an anomaly detection module and an instant release module; A license plate recognition module is used to obtain the vehicle's license plate information; ETC information reading module, used to obtain the license plate information in the vehicle ETC device; A comparison module, used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module, and allow the transaction if they match; The transaction verification module is used to verify the vehicle that has been traded after the transaction is completed to verify the validity of the transaction; The anomaly detection module is used to detect vehicles that have not passed through the entrance, obtain their ETC information, and determine whether they have completed transactions at the current toll plaza within the preset time; The instant release module is used to release the vehicle immediately when it is detected that the vehicle has completed the transaction within the preset time.

[0006] Preferably, the license plate recognition module includes: a high-resolution camera for capturing the license plate image of the vehicle; an optical character recognition unit for analyzing the license plate image and extracting character information; and a data processing unit for sending the extracted license plate information to the comparison module.

[0007] Preferably, the ETC information reading module includes: a 5.8 GHz microwave antenna for acquiring a license plate and related transaction information from the vehicle's ETC device; a data decoder for decoding received ETC device information; and a cache unit for temporarily storing the decoded ETC information for subsequent processing.

[0008] Preferably, a comparison module is used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module, and if they match, the transaction is allowed; there are two strings S1 and S2, with lengths of n and m respectively, and a two-dimensional matrix D with a size of (n+1)×(m+1) is constructed, and initialized: ;The initial state represents the editing cost from the empty string to the target string; Fill the matrix D using the recursive formula: ;in: ,Finally, the value of the matrix D[n][m] is the edit distance between S1 and S2; The character similarity is calculated by normalization formula: ;in, is the value of the matrix D[n][m], HAK is the character deviation value, ranging from [0,1].

[0009] Preferably, the license plate image is converted into a grayscale image, all license plate images are scaled to the same resolution, the color histogram of the license plate image is calculated, and the license plate image is divided into RGB or HSV channels; the histogram is calculated for each channel , each histogram is normalized: ; i represents the segmented interval of the pixel value, n is the number of intervals of the histogram, and the outer frame and character edge of the license plate are extracted using Canny edge detection; the contour feature vector Fs is calculated for the detected contour, the image is segmented, and the coordinates of each character are extracted; the character position feature vector is constructed , represents the center coordinates of each character, and calculates the license plate image similarity anomaly value of the two images. The expression is: ; Where GHA is the license plate image similarity outlier.

[0010] Preferably, the character deviation values ​​and license plate image similarity anomaly values ​​are converted into feature vectors, and the feature vectors are used as inputs of the machine learning model. The machine learning model predicts the matching degree value label of the license plate information for each group of feature vectors as the prediction target, and minimizes the sum of prediction errors for the matching degree value labels of all license plate information as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The matching degree value of the license plate information is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0011] Preferably, the obtained matching degree value of the license plate information is compared with a gradient matching threshold, the gradient matching threshold includes a first matching threshold and a second matching threshold, and the first matching threshold is less than the second matching threshold, and the matching degree value of the license plate information is compared with the first matching threshold and the second matching threshold respectively; If the matching degree value of the license plate information is greater than the second matching threshold, it means that the matching degree of the license plate information is high, and the license plate information is marked as a complete match, and the transaction is allowed; If the matching degree value of the license plate information is greater than or equal to the first matching threshold and less than or equal to the second matching threshold, it means that the matching degree of the license plate information is moderate, and the license plate information is marked as an incomplete match and needs further verification; If the matching degree value of the license plate information is less than the first matching threshold, it means that the matching degree of the license plate information is low, the license plate information is marked as unmatched, and the transaction is rejected.

[0012] Preferably, in the anomaly detection module, a license plate recognition module or an ETC information reading module at the entrance is used to capture the timestamp Tentry of vehicle entry; for an exit transaction, the timestamp Texit of vehicle departure is recorded; Calculate the time a vehicle spends in a toll plaza: ; Compare residence time Whether it is within a reasonable range, the maximum allowed stay time is set to Tmax; if the vehicle stays for more than Tmax or less than the reasonable time threshold Tmin, it is marked as an abnormal vehicle; Define the path set Pvalid, including the matching relationship between the entry site Sentry and the exit site Sexit: (Sentry, Sexit)∈Pvalid; if the vehicle path is not in Pvalid, it is marked as a path abnormality; Set a reasonable stay time range: Tmin≤Tstay≤Tmax; if it exceeds this range, it will be marked as abnormal.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention solves the problem of reduced license plate recognition efficiency due to microwave signal reflection, lane interaction interference and external environment in the prior art by designing a set of abnormality detection and instant release processing systems with strong dynamic environment adaptability, and at the same time makes up for the lack of fast verification and abnormality processing mechanisms in a multi-lane environment. The system integrates license plate recognition module, ETC information reading module, comparison module, transaction verification module, abnormality detection module and instant release module, which can efficiently complete vehicle information collection, comparison verification, abnormality detection and release operations, significantly improve the transaction accuracy and traffic efficiency of the toll collection system, and reduce congestion and erroneous deductions caused by erroneous transactions.

[0014] 2. The present invention further improves the reliability and intelligence level of the system by optimizing the detailed design. The polynomial regression model is used in combination with the character deviation value and the license plate image similarity anomaly value to train the machine learning model to achieve efficient prediction of the degree of license plate information matching; the matching degree is graded and decided by the gradient matching threshold to ensure the accuracy of the transaction; abnormal vehicles are judged based on reasonable time intervals and path rules, which enhances the accuracy and real-time performance of anomaly detection. The present invention is not only applicable to the existing ETC charging system, but also has good scalability and can be widely used in other scenarios in the field of intelligent transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] For examples, see Figure 1 As shown, the system for detecting and immediately releasing ETC transactions according to this embodiment includes a license plate recognition module, an ETC information reading module, a comparison module, a transaction verification module, an anomaly detection module and an immediate release module; A license plate recognition module is used to obtain the vehicle's license plate information; ETC information reading module, used to obtain the license plate information in the vehicle ETC device; A comparison module, used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module, and allow the transaction if they match; The transaction verification module is used to verify the vehicle that has been traded after the transaction is completed to verify the validity of the transaction; The anomaly detection module is used to detect vehicles that have not passed through the entrance, obtain their ETC information, and determine whether they have completed transactions at the current toll plaza within the preset time; The instant release module is used to release the vehicle immediately when it is detected that the vehicle has completed the transaction within the preset time.

[0019] The license plate recognition module includes: a high-resolution camera for capturing the vehicle's license plate image; an optical character recognition (OCR) unit for analyzing the license plate image and extracting character information; and a data processing unit for sending the extracted license plate information to the comparison module.

[0020] High-resolution camera: used to capture the license plate image of the vehicle. The camera has a wide dynamic range (WDR) and infrared enhancement function to ensure that the license plate information can be clearly captured under various lighting conditions (such as day, night, strong light, shadow or backlight). The camera is installed at the entrance of the toll station or above the ETC lane, and the angle is adjusted to ensure that the field of view covers the target vehicle. The resolution is not less than 1080P and supports high-definition video capture; the frame rate is ≥30fps to ensure fast capture of license plate images; equipped with auto focus function to meet the shooting needs of vehicles at different distances.

[0021] Optical Character Recognition (OCR) Unit: Analyzes license plate images captured by high-resolution cameras and extracts character information from the images as processable data. Supports recognition of multiple license plate formats, including standards from different countries or regions; built-in anti-interference algorithms to handle character defects caused by stains, reflections, occlusions, etc.; provides real-time recognition capabilities with a processing delay of less than 100 milliseconds. Optimized features: The OCR unit combines image preprocessing techniques (such as edge enhancement, noise filtering, and contrast enhancement) to improve recognition accuracy, especially in harsh environmental conditions.

[0022] Data processing unit: Formats the license plate character information extracted by the OCR unit and sends it to the comparison module. Binds the character information with additional information such as timestamp and camera number to ensure data traceability; verifies and encodes the extracted information to ensure data integrity and transmission security. The data processing unit and other modules in the system (such as the ETC information reading module and the comparison module) realize data interaction through a unified communication protocol (such as TCP / IP or CAN bus protocol).

[0023] The high-resolution camera captures the license plate image of the vehicle in real time and sends it to the OCR unit. The OCR unit extracts characters from the image and completes and optimizes incomplete or ambiguous characters. The data processing unit formats the extracted license plate character information and adds metadata such as timestamps, and sends the processed information to the comparison module for subsequent matching operations.

[0024] The ETC information reading module includes: a 5.8GHz microwave antenna for obtaining license plates and related transaction information from the vehicle's ETC device; a data decoder for decoding received ETC device information; and a cache unit for temporarily storing decoded ETC information for subsequent processing.

[0025] 5.8GHz microwave antenna: Through the 5.8GHz microwave communication protocol, it can communicate with the vehicle ETC device in two directions to obtain license plate information and related transaction data (such as vehicle type, user account information, device number, etc.). It supports short-distance high-speed communication, and the communication distance is generally 1-15 meters; it provides a highly directional antenna design to reduce signal interference from adjacent lanes; it has an automatic frequency adjustment function to adapt to frequency interference in a multi-channel environment. The antenna is equipped with a signal booster to ensure that the ETC device signal can still be stably received in harsh environments (such as rainy, foggy or toll plazas with severe signal reflection).

[0026] Data decoder: Decodes the encrypted data packets received by the 5.8GHz microwave antenna to extract the vehicle's ETC information. The original data packets obtained by the microwave antenna are usually sent in an encrypted format. The data decoder decrypts the data packets through the built-in algorithm to extract key information, such as license plate numbers, transaction records, and user account status. The decoded information is checked for integrity and validity to filter out erroneous or incomplete data packets. Supports multiple ETC protocol standards (such as the GB / T20851 Chinese ETC standard); has a high-performance processor to ensure that the decoding delay is less than 10 milliseconds to meet real-time requirements; is equipped with an anomaly detection mechanism to trigger an alarm when decoding fails or the signal is interrupted.

[0027] Cache unit: used to temporarily store decoded ETC information to support subsequent comparison and processing. Data writing: decoded ETC information is temporarily stored in the cache unit before being transmitted to other modules; data management: cache data is managed through the first-in-first-out (FIFO) algorithm to ensure the accuracy of the transmission order; data cleaning: after the data is successfully transmitted or processed, the cache is automatically cleaned to release storage space. High-speed storage chips are used to support high-frequency read and write operations; equipped with power-off protection function to ensure that data is not lost in the event of sudden failure; support dynamic cache expansion function to adjust cache capacity according to traffic load.

[0028] When a vehicle enters the ETC lane, the 5.8GHz microwave antenna detects the vehicle's ETC device and acquires its data packet through the microwave communication protocol.

[0029] The data decoder receives the data packets sent by the antenna, decrypts them and extracts key data such as the license plate number, transaction information, and device number. The decoded data is temporarily stored in the cache unit, and metadata such as timestamp and lane number are attached for further processing by the comparison module and transaction verification module. If duplicate transactions or abnormal signals are detected, the cache unit can also retain some data for system log recording or manual verification.

[0030] A comparison module is used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module. If they match, the transaction is allowed. The matching module includes: based on the edit distance or similarity index, fault-tolerant processing of partial character deviations in the license plate information caused by contamination, occlusion or recognition errors, and judging whether the license plates belong to the same vehicle by analyzing the overall features of the license plate image (such as color, font and arrangement).

[0031] Suppose there are two strings S1 (license plate recognition result) and S2 (license plate information in the ETC device), with lengths of n and m respectively. Construct a two-dimensional matrix D of size (n+1)×(m+1), initialized: ; The initial state represents the editing cost from the empty string to the target string.

[0032] Fill the matrix D using the recursive formula: ;in: Finally, the value of the matrix D[n][m] is the edit distance between S1 and S2.

[0033] The character similarity is calculated by normalization formula: ;in, is the value of the matrix D[n][m], HAK is the character deviation value, ranging from [0,1]. The closer the value is to 1, the more similar the strings are.

[0034] Convert the license plate image to grayscale to remove the interference of color information. Scale all license plate images to the same resolution (e.g. 100×30 pixels) to ensure consistency in feature extraction. Reduce image noise through Gaussian blur or median filtering.

[0035] Calculate the color histogram of the license plate image, divide the license plate image into RGB or HSV channels; calculate the histogram for each channel , each histogram is normalized: ; i represents the segmented interval of pixel values, n is the number of intervals of the histogram, and Canny edge detection is used to extract the outer border and character edge of the license plate; the contour feature vector Fs is calculated for the detected contour, such as contour area, perimeter, aspect ratio, etc.

[0036] Segment the image into characters and extract the coordinates of each character; construct the character position feature vector , represents the center coordinates of each character, and calculates the license plate image similarity anomaly value of the two images. The expression is: ; Where GHA is the license plate image similarity outlier.

[0037] According to the license plate image similarity anomaly value, the similarity anomaly value is defined as: E=1−GHA; when E>Threshold (Threshold is the anomaly threshold, such as 0.3), the license plate image is judged to be abnormal.

[0038] The character deviation values ​​and license plate image similarity anomaly values ​​are converted into feature vectors, and the feature vectors are used as inputs of the machine learning model. The machine learning model predicts the matching degree value label of the license plate information for each set of feature vectors as the prediction target, and minimizes the sum of prediction errors of the matching degree value labels of all license plate information as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The matching degree value of the license plate information is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0039] The method for obtaining the matching degree value of the license plate information is as follows: from the comprehensive feature vector training data of the trained machine learning model, the corresponding function expression is obtained: ; In the formula, is the output function of the model, HAK is the character deviation value, GHA is the license plate image similarity anomaly value, is the matching degree value of the license plate information.

[0040] Compare the obtained matching degree value of the license plate information with the gradient matching threshold, the gradient matching threshold includes a first matching threshold and a second matching threshold, and the first matching threshold is less than the second matching threshold, and compare the matching degree value of the license plate information with the first matching threshold and the second matching threshold respectively; If the matching degree value of the license plate information is greater than the second matching threshold, it means that the matching degree of the license plate information is high, and the license plate information is marked as a complete match, and the transaction is allowed; If the matching degree value of the license plate information is greater than or equal to the first matching threshold and less than or equal to the second matching threshold, it means that the matching degree of the license plate information is moderate, and the license plate information is marked as an incomplete match and needs further verification; If the matching degree value of the license plate information is less than the first matching threshold, it means that the matching degree of the license plate information is low, the license plate information is marked as unmatched, and the transaction is rejected.

[0041] The transaction verification module is used to verify the vehicle involved in the transaction after the transaction is completed to verify the validity of the transaction.

[0042] The composition of the transaction verification module: stores the historical transaction information of the vehicle, including license plate number, transaction time, entry / exit station information, transaction amount, etc. Technical features: supports fast query by license plate number, ETC equipment number and other conditions; the data storage period can be flexibly set according to system requirements (such as 30 days, 90 days).

[0043] Real-time transaction data interface: obtain real-time transaction data from the current toll station for processing by the verification module. Data source: including license plate recognition module, ETC information reading module and comparison module, etc. Verification logic unit: verify and compare transaction data according to preset rules and logical algorithms to determine whether the transaction is valid. Verification dimension: Time verification: check whether the transaction is completed within a reasonable time range (for example, whether the exit transaction time is later than the entry time). Path verification: verify whether the vehicle passes through a legal toll path; check the rationality of the transaction in combination with the path generation algorithm. Amount verification: check whether the transaction amount is consistent with the charging standard corresponding to the path. Repeat verification: determine whether the vehicle repeats the transaction in a short period of time (to prevent repeated deductions).

[0044] Abnormal marking module: when verification fails or abnormalities are found, abnormal marking is generated and the information is recorded in the system log for subsequent analysis and manual intervention. Abnormal types: missing entry: no entry transaction record is detected; repeated transaction: the time interval between transaction records is too short; abnormal path: the vehicle's entry and exit do not comply with the legal path rules.

[0045] Feedback unit: Feedback the verification results to the system's instant release module or other related modules to determine whether the vehicle is allowed to pass or needs further processing. Receive vehicle transaction information (such as license plate number, transaction time, entry / exit information) from the real-time transaction data interface. Call the historical transaction database to query the vehicle's historical transaction records by license plate number or ETC device number.

[0046] Verify the validity of the transaction: Time verification: verify whether the transaction time is compliant; Path verification: check whether the entry and exit sites are within the legal path; Amount verification: compare whether the transaction amount is consistent with the path charging standard; Repeat verification: check whether the vehicle passes the same charging point repeatedly in a short period of time. If the verification passes, a "transaction valid" mark is generated, allowing the vehicle to continue to pass; if the verification fails, an abnormal mark is generated and sent to the abnormal mark module for recording, and feedback is given to the instant release module to prompt manual intervention or refuse passage. The verification result is returned to the system, and its feedback content may include: verification status (valid / invalid); abnormal type (if any).

[0047] The anomaly detection module further includes: a time monitoring unit for recording the time after the vehicle enters the toll plaza; an anomaly rule library for storing rules related to vehicles not following the prescribed path or not trading at the entrance; and a detection logic unit for comparing the vehicle behavior with the rules to determine whether there is an anomaly.

[0048] The time monitoring unit is mainly used to record and track the time a vehicle spends in a toll plaza, thereby providing a time basis for anomaly detection. When a vehicle enters a toll plaza, the system automatically captures the timestamp and monitors the length of time it spends in different areas.

[0049] Use the license plate recognition module or ETC information reading module at the entrance to capture the timestamp Tentry of the vehicle entering; for exit transactions, record the timestamp Texit of the vehicle leaving.

[0050] Calculate the time a vehicle spends in a toll plaza: ; Compare residence time Is it within a reasonable range, for example, the maximum allowable residence time is Tmax.

[0051] If the vehicle's stay time exceeds Tmax or is less than the reasonable time threshold Tmin, it is marked as an abnormal vehicle.

[0052] The abnormal rule library stores a series of predefined rules to detect whether a vehicle violates traffic regulations. The rules cover various scenarios such as path abnormalities, transaction abnormalities, and dwell time abnormalities.

[0053] Define the legal path set Pvalid, such as the matching relationship between the entry site Sentry and the exit site Sexit: (Sentry, Sexit)∈Pvalid; if the vehicle path is not in Pvalid, it is marked as a path exception.

[0054] Trading rules include: Rule 1: The vehicle must have an import transaction record. If there is no import record and an export transaction is carried out directly, it will be marked as abnormal.

[0055] Rule 2: If a vehicle repeats transactions within a short period of time (e.g., the entry time interval is less than 5 minutes), it is marked as abnormal.

[0056] Set a reasonable stay time interval: Tmin≤Tstay≤Tmax; if it exceeds this range, it will be marked as abnormal. Special rules are formulated for special vehicles (such as emergency vehicles and VIP vehicles), such as allowing free passage or detouring specific routes. The abnormal rule library supports dynamic updates, and rules can be added, modified or deleted according to actual needs; rules are synchronized with the upper-level management platform of the toll system regularly to ensure real-time and effectiveness.

[0057] The detection logic unit compares the vehicle behavior data with the rules in the abnormal rule library to determine whether the vehicle has an abnormal situation. It combines the time information provided by the time monitoring unit and the real-time data of other modules to comprehensively analyze the vehicle behavior.

[0058] Receive the vehicle's timestamp, path information (entry point, exit point), transaction record, and other relevant information.

[0059] Check whether the rules in the anomaly rule base are satisfied one by one: Path anomaly detection: Check whether the vehicle path meets Pvalid; Transaction anomaly detection: Verify whether there are missing or duplicate transactions for the vehicle; Time anomaly detection: compare whether the vehicle stay time is within the range of Tmin and Tmax.

[0060] If a vehicle violates any rule, an abnormal flag Flagabnormal is generated and the abnormal type is recorded. The abnormal flag is fed back to the system log and the instant release module. For vehicles that trigger abnormal rules multiple times, their abnormal behavior is recorded and reported to the management system; support the generation of abnormal statistical reports for manual verification and decision-making reference.

[0061] The instant release module is used to release the vehicle immediately when it is detected that the vehicle has completed the transaction within the preset time.

[0062] Release control unit: receives the transaction verification results from the detection module and controls the release equipment at the toll plaza. Implementation method: Control signal interface: communicates with the gate or signal light control system through standard protocols (such as RS-485 or CAN bus); Release operation: after confirming that the transaction is completed and there are no abnormalities, sends a release signal to open the gate or switch the green light.

[0063] Multi-channel processing unit: In a multi-lane environment, it identifies and distributes release signals to ensure that vehicles pass through designated lanes accurately. Lane binding: associates the vehicle's transaction information with the corresponding lane number; multi-threaded processing: parallel processing of vehicle release operations in different lanes to avoid interference and delay.

[0064] Exception handling unit: handles incomplete transactions due to exceptions or errors, and provides support for subsequent processing. Exception prompt: informs the driver of transaction abnormalities through LED display screen or voice prompt device; manual operation: prompts the toll station administrator to manually check and release abnormal vehicles. Workflow of the instant release module Receive the detection results of the anomaly detection module or the transaction verification module, including the following information: vehicle transaction status (valid / invalid); license plate number or ETC device number; detection timestamp and lane number.

[0065] Confirm whether the vehicle completes the transaction within the preset time. The verification rules include: whether there is a valid transaction record; whether the transaction time is within a reasonable range; whether the license plate information or ETC device information matches.

[0066] Based on the transaction verification results, the following decisions are made: If the transaction is completed without any exceptions, a release signal is triggered; if the transaction is abnormal (such as duplicate transactions, path abnormalities, etc.), the exception handling process is triggered.

[0067] Control gates or signal lights: Normal release: open the gate or switch the green light to allow vehicles to pass; abnormal handling: keep the gate closed or switch the red light, and prompt manual verification. The release results and related records are stored in the system log for subsequent analysis and management.

[0068] In this embodiment, efficient management and exception handling of electronic toll collection (ETC) transactions are achieved by integrating multiple modules. The license plate recognition module works in conjunction with the ETC information reading module to obtain the vehicle's license plate information and ETC device information respectively. The comparison module matches the two to confirm the legitimacy of the transaction; the transaction verification module further verifies the completed transaction to ensure that the transaction is valid; the anomaly detection module specifically handles vehicles that have not passed the entrance but attempted to exit the transaction, and makes anomaly judgments based on preset time and transaction rules; the instant release module quickly releases the vehicle after confirming that the transaction is completed based on the transaction verification and anomaly detection results, ensuring the safety and efficiency of the toll collection system.

[0069] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0070] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0071] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. Applied to ETC transaction anomaly detection and instant release processing system, characterized by: It includes license plate recognition module, ETC information reading module, comparison module, transaction verification module, anomaly detection module and instant release module; A license plate recognition module is used to obtain the vehicle's license plate information; ETC information reading module, used to obtain the license plate information in the vehicle ETC device; A comparison module, used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module, and allow the transaction if they match; The transaction verification module is used to verify the vehicle that has been traded after the transaction is completed to verify the validity of the transaction; The anomaly detection module is used to detect vehicles that have not passed through the entrance, obtain their ETC information, and determine whether they have completed transactions at the current toll plaza within the preset time; The instant release module is used to release the vehicle immediately when it is detected that the vehicle has completed the transaction within the preset time.

2. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 1 is characterized in that: The license plate recognition module includes: a high-resolution camera for capturing the vehicle's license plate image; an optical character recognition unit for analyzing the license plate image and extracting character information; and a data processing unit for sending the extracted license plate information to the comparison module.

3. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 1 is characterized in that: The ETC information reading module includes: a 5.8GHz microwave antenna for obtaining license plates and related transaction information from the vehicle's ETC device; a data decoder for decoding received ETC device information; and a cache unit for temporarily storing decoded ETC information for subsequent processing.

4. The system for detecting and processing abnormalities in ETC transactions and immediately releasing traffic according to claim 1 is characterized in that: The comparison module is used to compare the license plate information obtained by the license plate recognition module with the license plate information obtained by the ETC information reading module. If they match, the transaction is allowed. It is assumed that there are two strings S1 and S2, with lengths of n and m respectively, and a two-dimensional matrix D with a size of (n+1)×(m+1) is constructed and initialized: ;The initial state represents the editing cost from the empty string to the target string; Fill the matrix D using the recursive formula: ;in: ,Finally, the value of the matrix D[n][m] is the edit distance between S1 and S2; The character similarity is calculated by normalization formula: ;in, is the value of the matrix D[n][m], HAK is the character deviation value, ranging from [0,1].

5. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 4 is characterized in that: Convert the license plate image into grayscale, scale all license plate images to the same resolution, calculate the color histogram of the license plate image, and separate the license plate image into RGB or HSV channels; Compute the histogram for each channel , each histogram is normalized: ; i represents the segmented interval of the pixel value, n is the number of intervals in the histogram, and the outer frame and character edge of the license plate are extracted using Canny edge detection; the contour feature vector Fs is calculated for the detected contour, the image is segmented, and the coordinates of each character are extracted; the character position feature vector Fs is constructed. , represents the center coordinates of each character, and calculates the license plate image similarity anomaly value of the two images. The expression is: ; Where GHA is the license plate image similarity outlier.

6. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 5 is characterized in that: The character deviation values ​​and license plate image similarity anomaly values ​​are converted into feature vectors, and the feature vectors are used as inputs of the machine learning model. The machine learning model predicts the matching degree value label of the license plate information for each set of feature vectors as the prediction target, and minimizes the sum of prediction errors of the matching degree value labels of all license plate information as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The matching degree value of the license plate information is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

7. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 6 is characterized in that: Compare the obtained matching degree value of the license plate information with the gradient matching threshold, the gradient matching threshold includes a first matching threshold and a second matching threshold, and the first matching threshold is less than the second matching threshold, and compare the matching degree value of the license plate information with the first matching threshold and the second matching threshold respectively; If the matching degree value of the license plate information is greater than the second matching threshold, it means that the matching degree of the license plate information is high, and the license plate information is marked as a complete match, and the transaction is allowed; If the matching degree value of the license plate information is greater than or equal to the first matching threshold and less than or equal to the second matching threshold, it means that the matching degree of the license plate information is moderate, and the license plate information is marked as an incomplete match and needs further verification; If the matching degree value of the license plate information is less than the first matching threshold, it means that the matching degree of the license plate information is low, the license plate information is marked as unmatched, and the transaction is rejected.

8. The system for detecting abnormalities in ETC transactions and immediately releasing traffic according to claim 1 is characterized in that: In the anomaly detection module, the license plate recognition module or ETC information reading module at the entrance is used to capture the timestamp Tentry of vehicle entry; for exit transactions, the timestamp Texit of vehicle departure is recorded; Calculate the time a vehicle spends in a toll plaza: ; Compare residence time Is it within a reasonable range? Set the maximum allowed residence time as Tmax; If the vehicle stays for longer than Tmax or less than the reasonable time threshold Tmin, it is marked as an abnormal vehicle; Define the path set Pvalid, including the matching relationship between the entry site Sentry and the exit site Sexit: (Sentry, Sexit)∈Pvalid; if the vehicle path is not in Pvalid, it is marked as a path abnormality; Set a reasonable stay time range: Tmin≤Tstay≤Tmax; if it exceeds this range, it will be marked as abnormal.

Images