A vehicle intelligent identification and passing management method and system for unmanned toll station

Abstract

The application provides a vehicle intelligent identification and passing management method and system of unmanned toll station. The method comprises the following steps: obtaining license plate image and vehicle shape feature data of passing vehicle, performing image preprocessing according to the license plate image to obtain license plate recognition data, performing matching processing according to the license plate recognition data and the vehicle shape feature data through a preset passing vehicle account binding database to obtain a matching account corresponding to the passing vehicle, deducting toll according to the matching account through a preset Internet of Things communication protocol, and performing passing management according to toll deduction state feature data. Through multi-feature fusion identification, dual-mode Internet of Things communication, multi-scene abnormal grading disposal and offline emergency passing management, the vehicle identification accuracy, passing efficiency, passing abnormal processing capacity and passing management stability are improved, so that the vehicle intelligent identification and passing management of unmanned toll station are realized.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and more specifically, to a method and system for intelligent vehicle identification and traffic management at unmanned toll stations. Background Technology

[0002] Unmanned toll stations can effectively improve traffic efficiency and promote the rapid development of intelligent transportation technology. Traditional unmanned toll stations achieve intelligent identification, toll deduction, and barrier lifting through license plate recognition and ETC assistance. However, traditional technologies have poor anti-interference capabilities and insufficient accuracy in license plate recognition when the license plate is damaged, deformed, or cloned. They also lack auxiliary verification measures, which can easily lead to matching errors. Traditional technologies use centralized cloud processing, which takes a long time for data interaction and transmission, and can easily lead to abnormal vehicle congestion. Moreover, data processing and handling rely on the network and lack offline processing capabilities. When the network fluctuates, the system cannot perform normal traffic management, affecting the operational stability of unmanned toll stations. Therefore, there is an urgent need for an intelligent vehicle identification and traffic management method for unmanned toll stations that can achieve high anti-interference capabilities, accurate identification, low latency, hierarchical handling in all scenarios, and ensure stable offline operation.

[0003] Effective technical solutions are urgently needed to address the above problems. Summary of the Invention

[0004] The purpose of this application is to provide a vehicle intelligent identification and traffic management method and system for unmanned toll stations. Through multi-feature fusion identification, dual-mode IoT communication, multi-scenario anomaly graded handling, and offline emergency traffic management, it improves the accuracy of vehicle identification, traffic efficiency, traffic anomaly handling capability, and traffic management stability, thereby realizing vehicle intelligent identification and traffic management for unmanned toll stations.

[0005] Firstly, this application provides a method for intelligent vehicle identification and traffic management at unmanned toll stations, including the following steps:

[0006] Acquire license plate images and vehicle shape feature data of passing vehicles, perform image preprocessing based on the license plate images, and obtain license plate recognition data;

[0007] Based on the license plate recognition data and vehicle appearance feature data, a matching process is performed through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle.

[0008] Toll fees are deducted from the matched account via a preset IoT communication protocol, and toll management is performed based on toll fee deduction status feature data.

[0009] Optionally, in the vehicle intelligent recognition and traffic management method for unmanned toll stations described in this application, the step of acquiring the license plate image and vehicle shape feature data of the passing vehicle, and performing image preprocessing based on the license plate image to obtain license plate recognition data includes:

[0010] Acquire license plate images and vehicle shape feature data of passing vehicles. The vehicle shape feature data includes vehicle body color feature data, vehicle type feature data, and window feature data.

[0011] Image preprocessing is performed on the license plate image to obtain license plate recognition data.

[0012] Optionally, in the vehicle intelligent recognition and traffic management method for unmanned toll stations described in this application, the step of performing image preprocessing based on the license plate image to obtain license plate recognition data includes:

[0013] Obtain image quality assessment parameter data for license plate images, including mean brightness, contrast, noise intensity, and saturation;

[0014] The average brightness, contrast, noise intensity, and saturation are input into a preset image acquisition environment type evaluation model for processing to obtain image acquisition environment category feature data;

[0015] Based on the image acquisition environment category feature data, a preset environment category and image enhancement processing mapping table is queried to obtain the image enhancement algorithm, and image enhancement processing is performed to obtain an enhanced license plate image;

[0016] The enhanced license plate image is processed to obtain license plate recognition data.

[0017] Optionally, in the vehicle intelligent identification and traffic management method for unmanned toll stations described in this application, the step of matching the license plate recognition data and vehicle shape feature data through a preset vehicle account binding database to obtain the matching account corresponding to the passing vehicle includes:

[0018] Based on the license plate recognition data and vehicle shape feature data, a matching process is performed through a preset vehicle account binding database to obtain the license plate recognition matching degree and vehicle shape feature matching degree.

[0019] The vehicle exterior feature matching degree includes body color matching degree, vehicle type matching degree, and window feature matching degree;

[0020] Based on the image acquisition environment category feature data, a preset acquisition environment and weight value mapping table is queried to obtain feature weight values, including license plate recognition weight value, vehicle body color matching weight value, vehicle type matching weight value and window feature matching weight value;

[0021] The license plate recognition matching degree, body color matching degree, vehicle type matching degree and window feature matching degree are weighted and summed with the corresponding license plate recognition weight value, body color matching weight value, vehicle type matching weight value and window feature matching weight value to obtain the account matching evaluation index.

[0022] The account matching evaluation indices are sorted in descending order, and the account with the highest account matching evaluation index is determined as the matching account corresponding to the passing vehicle.

[0023] Optionally, in the vehicle intelligent identification and traffic management method for unmanned toll stations described in this application, the step of deducting tolls according to the matched account via a preset Internet of Things communication protocol and performing traffic management based on toll deduction status feature data includes:

[0024] Based on the matched account, obtain the corresponding account balance and preset deduction rules at the preset edge computing node;

[0025] Based on the account balance and preset deduction rules, toll fees are deducted through a preset Internet of Things communication protocol to obtain toll deduction status feature data, including whether the deduction is completed or not.

[0026] If the toll deduction is not completed, the toll deduction status data will be transmitted to the toll gate control system via a preset IoT communication protocol, and permission will be granted.

[0027] If the payment has not been completed, passage will be prohibited.

[0028] Optionally, the vehicle intelligent identification and traffic management method for unmanned toll stations described in this application further includes:

[0029] Acquire passage perception data of passing vehicles, including network connection status characteristic data, barrier gate operation status characteristic data, and passage visual characteristic data;

[0030] The license plate recognition data, toll deduction status feature data, network connection status feature data, barrier gate operation status feature data, and passage visual feature data are input into a preset passage anomaly type evaluation model for processing to obtain passage anomaly type feature data.

[0031] Based on the aforementioned abnormality type feature data, a preset abnormality type and abnormality level mapping table is queried to obtain the abnormality level, including mild abnormality, moderate abnormality, or severe abnormality.

[0032] The database of preset anomaly handling methods is queried based on the severity of the anomaly (mild, moderate, or severe) to obtain the corresponding anomaly handling method.

[0033] Optionally, the vehicle intelligent identification and traffic management method for unmanned toll stations described in this application further includes:

[0034] If the network connection status feature data is in a disconnected state, then user matching is performed based on the license plate recognition data and vehicle appearance feature data through a preset high-frequency traffic offline record database to obtain the user matching status.

[0035] If the user matching status is "match successful", then pre-deduction will be performed;

[0036] If the user matching status is "matching failed", then the temporary access identity binding is activated, and pre-deduction is executed according to the preset deduction rules.

[0037] Optionally, the vehicle intelligent identification and traffic management method for unmanned toll stations described in this application further includes:

[0038] If the prepayment fails, the system will retrieve the user's historical credit assessment data, including the toll payment success rate and the number of toll evasion records.

[0039] The credit rating of the user is determined by processing the toll payment success rate and the number of toll evasion records, including excellent credit, good credit, or poor credit.

[0040] If the credit rating is excellent or good, then the credit pass response will be activated;

[0041] If the credit score is poor, a temporary pass identity will be activated and pre-deducted according to the preset deduction rules.

[0042] Secondly, this application provides a vehicle intelligent identification and passage management system for unmanned toll stations. The system includes a memory and a processor. The memory includes a program for a vehicle intelligent identification and passage management method for unmanned toll stations. When the program for the vehicle intelligent identification and passage management method for unmanned toll stations is executed by the processor, it implements the following steps:

[0043] Acquire license plate images and vehicle shape feature data of passing vehicles, perform image preprocessing based on the license plate images, and obtain license plate recognition data;

[0044] Based on the license plate recognition data and vehicle appearance feature data, a matching process is performed through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle.

[0045] Toll fees are deducted from the matched account via a preset IoT communication protocol, and toll management is performed based on toll fee deduction status feature data.

[0046] As can be seen from the above, the vehicle intelligent identification and traffic management method and system for unmanned toll stations provided in this application improves the accuracy of vehicle identification, traffic efficiency, traffic anomaly handling capability and traffic management stability through multi-feature fusion identification, dual-mode IoT communication, multi-scenario anomaly hierarchical handling and offline emergency traffic management, thereby realizing vehicle intelligent identification and traffic management for unmanned toll stations.

[0047] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application 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.

[0049] Figure 1 A flowchart illustrating a vehicle intelligent identification and traffic management method for an unmanned toll station, as provided in this application embodiment;

[0050] Figure 2 A flowchart illustrating the acquisition of license plate recognition data in a vehicle intelligent identification and traffic management method for an unmanned toll station, as provided in an embodiment of this application.

[0051] Figure 3 A flowchart illustrating the process of obtaining the matching account corresponding to a passing vehicle in a vehicle intelligent identification and traffic management method for an unmanned toll station, as provided in an embodiment of this application.

[0052] Figure 4 A high-level flowchart of a vehicle intelligent identification and traffic management method for an unmanned toll station provided in the application embodiment. Detailed Implementation

[0053] 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 a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0054] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0055] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a vehicle intelligent identification and traffic management method for an unmanned toll station, as described in some embodiments of this application. This vehicle intelligent identification and traffic management method for an unmanned toll station is used in terminal devices, such as computers and mobile terminals. The vehicle intelligent identification and traffic management method for an unmanned toll station includes the following steps:

[0056] S11. Obtain the license plate image and vehicle shape feature data of the passing vehicles, perform image preprocessing based on the license plate image, and obtain license plate recognition data;

[0057] S12. Based on the license plate recognition data and vehicle shape feature data, a matching process is performed through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle.

[0058] S13. Deduct toll fees according to the matched account through a preset IoT communication protocol, and perform toll management based on toll fee deduction status feature data.

[0059] It should be noted that, in order to achieve accurate identification of passing vehicles and stable intelligent management of unmanned toll stations, firstly, multi-dimensional accurate identification is assisted by license plate and vehicle shape characteristics. Then, quantitative evaluation and accurate matching are performed to obtain the accurate matching account corresponding to the passing vehicle, so as to facilitate toll deduction and timely passage. This embodiment adopts the 5G and LoRa dual-mode IoT communication protocol to form a point-to-point real-time communication link between the traffic monitoring camera, edge computing node, traffic payment system and traffic gate controller, with high data interaction and transmission efficiency.

[0060] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the process of obtaining license plate recognition data in a vehicle intelligent identification and traffic management method for an unmanned toll station, as described in some embodiments of this application. According to embodiments of the present invention, the step of obtaining license plate images and vehicle shape feature data of passing vehicles, and performing image preprocessing based on the license plate images to obtain license plate recognition data, includes:

[0061] S21. Obtain license plate images and vehicle shape feature data of passing vehicles, wherein the vehicle shape feature data includes vehicle body color feature data, vehicle type feature data and window feature data.

[0062] S22. Perform image preprocessing based on the license plate image to obtain license plate recognition data.

[0063] It should be noted that, in order to achieve multi-feature cross-fusion verification, while acquiring license plate images, multi-dimensional auxiliary features including vehicle body color, vehicle type, and window features are also acquired. Vehicle type features include compact SUV subtype under SUV vehicle classification, and window features include the number of windows, window aspect ratio, and window area. The acquired license plate images are preprocessed with image enhancement based on the acquisition environment to facilitate accurate license plate recognition.

[0064] According to an embodiment of the present invention, the step of performing image preprocessing based on the license plate image to obtain license plate recognition data includes:

[0065] Obtain image quality assessment parameter data for license plate images, including mean brightness, contrast, noise intensity, and saturation;

[0066] The average brightness, contrast, noise intensity, and saturation are input into a preset image acquisition environment type evaluation model for processing to obtain image acquisition environment category feature data;

[0067] Based on the image acquisition environment category feature data, a preset environment category and image enhancement processing mapping table is queried to obtain the image enhancement algorithm, and image enhancement processing is performed to obtain an enhanced license plate image;

[0068] The enhanced license plate image is processed to obtain license plate recognition data.

[0069] It should be noted that, in order to improve the accuracy of license plate recognition, the average brightness, contrast, noise intensity, and saturation of the acquired license plate image are used to identify the corresponding image acquisition environment type evaluation model through pre-training. The image acquisition environment type is identified by the average brightness, contrast, noise intensity, and saturation of the acquired license plate image. The corresponding image acquisition environment type feature data is obtained. The image acquisition environment type is represented by a unique identifier. Based on the determined image acquisition environment type, the preset environment type and image enhancement processing mapping table is queried to obtain the image enhancement algorithm. For example, in the backlight environment, dynamic exposure compensation and shadow detail restoration are performed. In the rainy environment, rain streaks are removed and noise suppression is performed to obtain an enhanced license plate image with enhanced local features. The license plate recognition data is obtained through precise ROI positioning and feature refinement extraction. The preset image acquisition environment type evaluation model is obtained by training with a large number of historical samples of average brightness, contrast, noise intensity, and saturation, as well as the corresponding image acquisition environment type feature data. The preset environment type and image enhancement processing mapping table is constructed by those skilled in the art through analysis of historical image enhancement samples and can be dynamically adjusted according to specific applications.

[0070] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating the process of obtaining a matching account corresponding to a passing vehicle in a vehicle intelligent identification and traffic management method for an unmanned toll station, according to some embodiments of this application. According to an embodiment of the present invention, the step of obtaining a matching account corresponding to a passing vehicle by matching the license plate recognition data and vehicle appearance feature data through a preset passing vehicle account binding database includes:

[0071] S31. Based on the license plate recognition data and vehicle shape feature data, perform matching processing through a preset vehicle account binding database to obtain the license plate recognition matching degree and the vehicle shape feature matching degree.

[0072] S32, The vehicle exterior feature matching degree includes body color matching degree, vehicle type matching degree, and window feature matching degree;

[0073] S33. Query the preset acquisition environment and weight value mapping table according to the image acquisition environment category feature data to obtain feature weight values, including license plate recognition weight value, body color matching weight value, vehicle type matching weight value and window feature matching weight value;

[0074] S34. The license plate recognition matching degree, body color matching degree, vehicle type matching degree and window feature matching degree are weighted and summed with the corresponding license plate recognition weight value, body color matching weight value, vehicle type matching weight value and window feature matching weight value to obtain the account matching evaluation index.

[0075] S35. Sort the account matching evaluation index in descending order, and determine the account with the largest account matching evaluation index as the matching account corresponding to the passing vehicle.

[0076] It should be noted that, in order to improve the accurate matching of passing vehicles and payment accounts, a pre-defined database of passing vehicle account bindings is obtained by passing users through the system by pre-registering their accounts, which include license plates, vehicle body features, and payment accounts. Based on the identified license plate recognition data and vehicle appearance feature data, a matching degree is obtained. Different weight values ​​are assigned to different features according to different collection environments, and a weighted summation process is performed to obtain an account matching evaluation index, thereby achieving quantitative evaluation. The account with the highest account matching evaluation index is then determined as the matching account corresponding to the passing vehicle. For example, in license plate recognition, D is identified as O, while vehicle body color, vehicle type, and window features are used for auxiliary recognition to accurately match the corresponding passing account, avoiding recognition failures caused by static binding of license plates and accounts.

[0077] According to an embodiment of the present invention, the step of deducting toll fees based on the matched account via a preset Internet of Things communication protocol, and performing toll management based on toll fee deduction status feature data, includes:

[0078] Based on the matched account, obtain the corresponding account balance and preset deduction rules at the preset edge computing node;

[0079] Based on the account balance and preset deduction rules, toll fees are deducted through a preset Internet of Things communication protocol to obtain toll deduction status feature data, including whether the deduction is completed or not.

[0080] If the toll deduction is not completed, the toll deduction status data will be transmitted to the toll gate control system via a preset IoT communication protocol, and permission will be granted.

[0081] If the payment has not been completed, passage will be prohibited.

[0082] It should be noted that after completing the identification of passing vehicles and account matching, in order to improve traffic efficiency and reduce vehicle dwell time, this embodiment adopts the 5G and LoRa dual-mode IoT communication protocol. The edge computing node caches the balance of the matched account and the preset deduction rules. After the identification and matching are completed, a local pre-deduction request is directly initiated to the payment system without waiting for cloud instructions. The deduction result is fed back to the gate control module in real time to perform gate management. The cloud backend is used for payment data verification and anomaly handling, but does not participate in real-time decision-making.

[0083] According to an embodiment of the present invention, it further includes:

[0084] Acquire passage perception data of passing vehicles, including network connection status characteristic data, barrier gate operation status characteristic data, and passage visual characteristic data;

[0085] The license plate recognition data, toll deduction status feature data, network connection status feature data, barrier gate operation status feature data, and passage visual feature data are input into a preset passage anomaly type evaluation model for processing to obtain passage anomaly type feature data.

[0086] Based on the aforementioned abnormality type feature data, a preset abnormality type and abnormality level mapping table is queried to obtain the abnormality level, including mild abnormality, moderate abnormality, or severe abnormality.

[0087] The database of preset anomaly handling methods is queried based on the severity of the anomaly (mild, moderate, or severe) to obtain the corresponding anomaly handling method.

[0088] It should be noted that, in order to enable timely intelligent handling of anomalies at unmanned toll stations, the following steps are taken: First, based on predetermined license plate recognition data, toll deduction status data, network connection status data, barrier gate operation status data, and visual traffic characteristics, a pre-trained, preset anomaly type evaluation model identifies anomaly type features. Toll deduction status includes completed or incomplete deduction; network connection status includes normal or disconnected; barrier gate operation status includes raised or not raised; visual traffic characteristics include vehicle collision, following / bumping, and vehicle fire; and anomaly types include payment anomaly, passage anomaly, or emergency anomaly. Each anomaly type feature is represented by a unique identifier. Then, based on the determined anomaly type feature data, a preset anomaly type-anomaly level mapping table is queried to obtain the anomaly level, and a preset anomaly handling method database is queried to obtain the corresponding anomaly handling method. The method involves several scenarios. For example, if a payment anomaly is classified as a minor anomaly, an LED screen display and voice prompts are triggered. If a vehicle without a license plate is identified as a medium anomaly, the edge computing node generates a temporary access ID based on the collected vehicle body color, vehicle type, and window features, triggering an LED screen display and voice prompts. If a vehicle fire is identified as a severe anomaly, an audible and visual alarm is activated, an LED screen display is triggered, and the visual characteristics of the passage are pushed to maintenance personnel for collaborative handling. The preset passage anomaly type assessment model is trained using a large amount of historical sample data, including license plate recognition data, toll deduction status feature data, network connection status feature data, gate operation status feature data, and passage visual feature data, as well as corresponding passage anomaly type feature data. The preset anomaly type and anomaly level mapping table and the preset anomaly handling method database are constructed by those skilled in the art based on historical cases and can be dynamically adjusted.

[0089] According to an embodiment of the present invention, it further includes:

[0090] If the network connection status feature data is in a disconnected state, then user matching is performed based on the license plate recognition data and vehicle appearance feature data through a preset high-frequency traffic offline record database to obtain the user matching status.

[0091] If the user matching status is "match successful", then pre-deduction will be performed;

[0092] If the user matching status is "matching failed", then the temporary access identity binding is activated, and pre-deduction is executed according to the preset deduction rules.

[0093] It should be noted that, to avoid the inability to pass through unattended toll stations due to network fluctuations or cloud failures, the following measures are taken: First, during normal passage, a preset high-frequency offline passage record database is built based on the passage frequency of high-frequency users. The network connection status is identified through heartbeat detection and communication timeout detection. When a disconnection is detected, user matching is performed based on license plate recognition data and vehicle appearance feature data through the preset high-frequency offline passage record database. If a user is matched in the database, pre-deduction is performed directly. If no user is matched in the database, a temporary passage ID is generated based on the collected vehicle body color, vehicle type, and window features, and pre-deduction is performed according to the preset deduction rules (such as no preferential rate). After the network is reconnected, the data is checked. Data without conflicts is directly written to the cloud database, and abnormal data is pushed to maintenance personnel for manual processing.

[0094] According to an embodiment of the present invention, it further includes:

[0095] If the user matching status is missing matching information, then obtain the historical credit assessment data of the user, including the toll payment success rate and the number of toll evasion records;

[0096] The credit rating of the user is determined by processing the toll payment success rate and the number of toll evasion records, including excellent credit, good credit, or poor credit.

[0097] If the credit rating is excellent or good, then the credit pass response will be activated;

[0098] If the credit score is poor, a temporary pass identity will be activated and pre-deducted according to the preset deduction rules.

[0099] It should be noted that in offline mode, when matching users, matching information may be missing. To improve passage efficiency and reduce waiting time, the credit rating of users is first assessed based on the payment success rate and the number of evasion records. For example, a payment success rate of 100% and 0 evasion records are considered excellent credit; a payment success rate of 90%-100% and less than 1 evasion record are considered good credit; and vice versa. The specific credit assessment can be dynamically adjusted. Users with excellent or good credit will be granted a credit passage response and a fee payment reminder will be sent within a preset time after passage. However, for users with poor credit, a temporary passage identity will be activated and pre-deducted according to the preset deduction rules.

[0100] Please refer to Figure 4 , Figure 4 This is a high-level flowchart of a vehicle intelligent identification and traffic management method for an unmanned toll station according to some embodiments of this application.

[0101] This invention also discloses a vehicle intelligent identification and passage management system for unmanned toll stations, including a memory and a processor. The memory includes a method program for vehicle intelligent identification and passage management at unmanned toll stations. When the processor executes the method program, the following steps are implemented:

[0102] Acquire license plate images and vehicle shape feature data of passing vehicles, perform image preprocessing based on the license plate images, and obtain license plate recognition data;

[0103] Based on the license plate recognition data and vehicle appearance feature data, a matching process is performed through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle.

[0104] Toll fees are deducted from the matched account via a preset IoT communication protocol, and toll management is performed based on toll fee deduction status feature data.

[0105] It should be noted that, in order to achieve accurate identification of passing vehicles and stable intelligent management of unmanned toll stations, firstly, multi-dimensional accurate identification is assisted by license plate and vehicle shape characteristics. Then, quantitative evaluation and accurate matching are performed to obtain the accurate matching account corresponding to the passing vehicle, so as to facilitate toll deduction and timely passage. This embodiment adopts the 5G and LoRa dual-mode IoT communication protocol to form a point-to-point real-time communication link between the traffic monitoring camera, edge computing node, traffic payment system and traffic gate controller, with high data interaction and transmission efficiency.

[0106] According to an embodiment of the present invention, the step of acquiring the license plate image and vehicle shape feature data of a passing vehicle, and performing image preprocessing based on the license plate image to obtain license plate recognition data includes:

[0107] Acquire license plate images and vehicle shape feature data of passing vehicles. The vehicle shape feature data includes vehicle body color feature data, vehicle type feature data, and window feature data.

[0108] Image preprocessing is performed on the license plate image to obtain license plate recognition data.

[0109] It should be noted that, in order to achieve multi-feature cross-fusion verification, while acquiring license plate images, multi-dimensional auxiliary features including vehicle body color, vehicle type, and window features are also acquired. Vehicle type features include compact SUV subtype under SUV vehicle classification, and window features include the number of windows, window aspect ratio, and window area. The acquired license plate images are preprocessed with image enhancement based on the acquisition environment to facilitate accurate license plate recognition.

[0110] According to an embodiment of the present invention, the step of performing image preprocessing based on the license plate image to obtain license plate recognition data includes:

[0111] Obtain image quality assessment parameter data for license plate images, including mean brightness, contrast, noise intensity, and saturation;

[0112] The average brightness, contrast, noise intensity, and saturation are input into a preset image acquisition environment type evaluation model for processing to obtain image acquisition environment category feature data;

[0113] Based on the image acquisition environment category feature data, a preset environment category and image enhancement processing mapping table is queried to obtain the image enhancement algorithm, and image enhancement processing is performed to obtain an enhanced license plate image;

[0114] The enhanced license plate image is processed to obtain license plate recognition data.

[0115] It should be noted that, in order to improve the accuracy of license plate recognition, the average brightness, contrast, noise intensity, and saturation of the acquired license plate image are used to identify the corresponding image acquisition environment type evaluation model through pre-training. The image acquisition environment type is identified by the average brightness, contrast, noise intensity, and saturation of the acquired license plate image. The corresponding image acquisition environment type feature data is obtained. The image acquisition environment type is represented by a unique identifier. Based on the determined image acquisition environment type, the preset environment type and image enhancement processing mapping table is queried to obtain the image enhancement algorithm. For example, in the backlight environment, dynamic exposure compensation and shadow detail restoration are performed. In the rainy environment, rain streaks are removed and noise suppression is performed to obtain an enhanced license plate image with enhanced local features. The license plate recognition data is obtained through precise ROI positioning and feature refinement extraction. The preset image acquisition environment type evaluation model is obtained by training with a large number of historical samples of average brightness, contrast, noise intensity, and saturation, as well as the corresponding image acquisition environment type feature data. The preset environment type and image enhancement processing mapping table is constructed by those skilled in the art through analysis of historical image enhancement samples and can be dynamically adjusted according to specific applications.

[0116] According to an embodiment of the present invention, the step of matching the license plate recognition data and vehicle appearance feature data through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle includes:

[0117] Based on the license plate recognition data and vehicle shape feature data, a matching process is performed through a preset vehicle account binding database to obtain the license plate recognition matching degree and vehicle shape feature matching degree.

[0118] The vehicle exterior feature matching degree includes body color matching degree, vehicle type matching degree, and window feature matching degree;

[0119] Based on the image acquisition environment category feature data, a preset acquisition environment and weight value mapping table is queried to obtain feature weight values, including license plate recognition weight value, vehicle body color matching weight value, vehicle type matching weight value and window feature matching weight value;

[0120] The license plate recognition matching degree, body color matching degree, vehicle type matching degree and window feature matching degree are weighted and summed with the corresponding license plate recognition weight value, body color matching weight value, vehicle type matching weight value and window feature matching weight value to obtain the account matching evaluation index.

[0121] The account matching evaluation indices are sorted in descending order, and the account with the highest account matching evaluation index is determined as the matching account corresponding to the passing vehicle.

[0122] It should be noted that, in order to improve the accurate matching of passing vehicles and payment accounts, a pre-defined database of passing vehicle account bindings is obtained by passing users through the system by pre-registering their accounts, which include license plates, vehicle body features, and payment accounts. Based on the identified license plate recognition data and vehicle appearance feature data, a matching degree is obtained. Different weight values ​​are assigned to different features according to different collection environments, and a weighted summation process is performed to obtain an account matching evaluation index, thereby achieving quantitative evaluation. The account with the highest account matching evaluation index is then determined as the matching account corresponding to the passing vehicle. For example, in license plate recognition, D is identified as O, while vehicle body color, vehicle type, and window features are used for auxiliary recognition to accurately match the corresponding passing account, avoiding recognition failures caused by static binding of license plates and accounts.

[0123] According to an embodiment of the present invention, the step of deducting toll fees based on the matched account via a preset Internet of Things communication protocol, and performing toll management based on toll fee deduction status feature data, includes:

[0124] Based on the matched account, obtain the corresponding account balance and preset deduction rules at the preset edge computing node;

[0125] Based on the account balance and preset deduction rules, toll fees are deducted through a preset Internet of Things communication protocol to obtain toll deduction status feature data, including whether the deduction is completed or not.

[0126] If the toll deduction is not completed, the toll deduction status data will be transmitted to the toll gate control system via a preset IoT communication protocol, and permission will be granted.

[0127] If the payment has not been completed, passage will be prohibited.

[0128] It should be noted that after completing the identification of passing vehicles and account matching, in order to improve traffic efficiency and reduce vehicle dwell time, this embodiment adopts the 5G and LoRa dual-mode IoT communication protocol. The edge computing node caches the balance of the matched account and the preset deduction rules. After the identification and matching are completed, a local pre-deduction request is directly initiated to the payment system without waiting for cloud instructions. The deduction result is fed back to the gate control module in real time to perform gate management. The cloud backend is used for payment data verification and anomaly handling, but does not participate in real-time decision-making.

[0129] According to an embodiment of the present invention, it further includes:

[0130] Acquire passage perception data of passing vehicles, including network connection status characteristic data, barrier gate operation status characteristic data, and passage visual characteristic data;

[0131] The license plate recognition data, toll deduction status feature data, network connection status feature data, barrier gate operation status feature data, and passage visual feature data are input into a preset passage anomaly type evaluation model for processing to obtain passage anomaly type feature data.

[0132] Based on the aforementioned abnormality type feature data, a preset abnormality type and abnormality level mapping table is queried to obtain the abnormality level, including mild abnormality, moderate abnormality, or severe abnormality.

[0133] The database of preset anomaly handling methods is queried based on the severity of the anomaly (mild, moderate, or severe) to obtain the corresponding anomaly handling method.

[0134] It should be noted that, in order to enable timely intelligent handling of anomalies at unmanned toll stations, the following steps are taken: First, based on predetermined license plate recognition data, toll deduction status data, network connection status data, barrier gate operation status data, and visual traffic characteristics, a pre-trained, preset anomaly type evaluation model identifies anomaly type features. Toll deduction status includes completed or incomplete deduction; network connection status includes normal or disconnected; barrier gate operation status includes raised or not raised; visual traffic characteristics include vehicle collision, following / bumping, and vehicle fire; and anomaly types include payment anomaly, passage anomaly, or emergency anomaly. Each anomaly type feature is represented by a unique identifier. Then, based on the determined anomaly type feature data, a preset anomaly type-anomaly level mapping table is queried to obtain the anomaly level, and a preset anomaly handling method database is queried to obtain the corresponding anomaly handling method. The method involves several scenarios. For example, if a payment anomaly is classified as a minor anomaly, an LED screen display and voice prompts are triggered. If a vehicle without a license plate is identified as a medium anomaly, the edge computing node generates a temporary access ID based on the collected vehicle body color, vehicle type, and window features, triggering an LED screen display and voice prompts. If a vehicle fire is identified as a severe anomaly, an audible and visual alarm is activated, an LED screen display is triggered, and the visual characteristics of the passage are pushed to maintenance personnel for collaborative handling. The preset passage anomaly type assessment model is trained using a large amount of historical sample data, including license plate recognition data, toll deduction status feature data, network connection status feature data, gate operation status feature data, and passage visual feature data, as well as corresponding passage anomaly type feature data. The preset anomaly type and anomaly level mapping table and the preset anomaly handling method database are constructed by those skilled in the art based on historical cases and can be dynamically adjusted.

[0135] According to an embodiment of the present invention, it further includes:

[0136] If the network connection status feature data is in a disconnected state, then user matching is performed based on the license plate recognition data and vehicle appearance feature data through a preset high-frequency traffic offline record database to obtain the user matching status.

[0137] If the user matching status is "match successful", then pre-deduction will be performed;

[0138] If the user matching status is "matching failed", then the temporary access identity binding is activated, and pre-deduction is executed according to the preset deduction rules.

[0139] It should be noted that, to avoid the inability to pass through unattended toll stations due to network fluctuations or cloud failures, the following measures are taken: First, during normal passage, a preset high-frequency offline passage record database is built based on the passage frequency of high-frequency users. The network connection status is identified through heartbeat detection and communication timeout detection. When a disconnection is detected, user matching is performed based on license plate recognition data and vehicle appearance feature data through the preset high-frequency offline passage record database. If a user is matched in the database, pre-deduction is performed directly. If no user is matched in the database, a temporary passage ID is generated based on the collected vehicle body color, vehicle type, and window features, and pre-deduction is performed according to the preset deduction rules (such as no preferential rate). After the network is reconnected, the data is checked. Data without conflicts is directly written to the cloud database, and abnormal data is pushed to maintenance personnel for manual processing.

[0140] According to an embodiment of the present invention, it further includes:

[0141] If the user matching status is missing matching information, then obtain the historical credit assessment data of the user, including the toll payment success rate and the number of toll evasion records;

[0142] The credit rating of the user is determined by processing the toll payment success rate and the number of toll evasion records, including excellent credit, good credit, or poor credit.

[0143] If the credit rating is excellent or good, then the credit pass response will be activated;

[0144] If the credit score is poor, a temporary pass identity will be activated and pre-deducted according to the preset deduction rules.

[0145] It should be noted that in offline mode, when matching users, matching information may be missing. To improve passage efficiency and reduce waiting time, the credit rating of users is first assessed based on the payment success rate and the number of evasion records. For example, a payment success rate of 100% and 0 evasion records are considered excellent credit; a payment success rate of 90%-100% and less than 1 evasion record are considered good credit; and vice versa. The specific credit assessment can be dynamically adjusted. Users with excellent or good credit will be granted a credit passage response and a fee payment reminder will be sent within a preset time after passage. However, for users with poor credit, a temporary passage identity will be activated and pre-deducted according to the preset deduction rules.

[0146] This invention discloses a vehicle intelligent identification and traffic management method and system for unmanned toll stations. By using multi-feature fusion identification, dual-mode IoT communication, multi-scenario anomaly graded handling, and offline emergency traffic management, it improves the accuracy of vehicle identification, traffic efficiency, traffic anomaly handling capability, and traffic management stability, thereby realizing vehicle intelligent identification and traffic management for unmanned toll stations.

[0147] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0148] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0149] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0150] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0151] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for intelligent vehicle identification and traffic management at an unmanned toll station, characterized in that, Includes the following steps: Acquire license plate images and vehicle shape feature data of passing vehicles, perform image preprocessing based on the license plate images, and obtain license plate recognition data; Based on the license plate recognition data and vehicle appearance feature data, a matching process is performed through a preset vehicle account binding database to obtain the matching account corresponding to the vehicle. Toll fees are deducted from the matched account via a preset IoT communication protocol, and toll management is performed based on toll fee deduction status feature data. The process of acquiring license plate images and vehicle shape feature data of passing vehicles, and performing image preprocessing based on the license plate images to obtain license plate recognition data includes: acquiring license plate images and vehicle shape feature data of passing vehicles, wherein the vehicle shape feature data includes vehicle body color feature data, vehicle type feature data, and window feature data; and performing image preprocessing based on the license plate images to obtain license plate recognition data. The step of preprocessing the license plate image to obtain license plate recognition data includes: acquiring image quality assessment parameter data of the license plate image, including average brightness, contrast, noise intensity, and saturation; inputting the average brightness, contrast, noise intensity, and saturation into a preset image acquisition environment type evaluation model for processing to obtain image acquisition environment category feature data; querying a preset environment category and image enhancement processing mapping table based on the image acquisition environment category feature data to obtain an image enhancement algorithm, and performing image enhancement processing to obtain an enhanced license plate image; and performing recognition processing based on the enhanced license plate image to obtain license plate recognition data. The step of matching the license plate recognition data and vehicle appearance feature data through a preset vehicle account binding database to obtain the matching account corresponding to the passing vehicle includes: matching the license plate recognition data and vehicle appearance feature data through a preset vehicle account binding database to obtain a license plate recognition matching degree and a vehicle appearance feature matching degree; the vehicle appearance feature matching degree includes a body color matching degree, a vehicle type matching degree, and a window feature matching degree; querying a preset acquisition environment and weight value mapping table based on the image acquisition environment category feature data to obtain feature weight values, including a license plate recognition weight value, a body color matching weight value, a vehicle type matching weight value, and a window feature matching weight value; performing a weighted summation of the license plate recognition matching degree, body color matching degree, vehicle type matching degree, and window feature matching degree with the corresponding license plate recognition weight value, body color matching weight value, vehicle type matching weight value, and window feature matching weight value to obtain an account matching evaluation index; sorting the account matching evaluation index in descending order, and determining the account with the largest account matching evaluation index as the matching account corresponding to the passing vehicle. The vehicle intelligent identification and passage management method of the unmanned toll station specifically adopts the 5G and LoRa dual-mode Internet of Things communication protocol. The edge computing node caches and matches the account balance and preset deduction rules. After identification and matching, it initiates a local pre-deduction request to the payment system and feeds back the deduction result to the passage barrier control system in real time to execute the gate management. The intelligent vehicle identification and traffic management method for unmanned toll stations also acquires traffic perception data of passing vehicles, including network connection status feature data, barrier gate operation status feature data, and traffic visual feature data; the license plate recognition data, toll deduction status feature data, network connection status feature data, barrier gate operation status feature data, and traffic visual feature data are input into a preset traffic anomaly type evaluation model for processing to obtain traffic anomaly type feature data; based on the traffic anomaly type feature data, a preset anomaly type and anomaly level mapping table is queried to obtain the traffic anomaly level, including mild anomaly, moderate anomaly, or severe anomaly; based on the mild, moderate, or severe anomaly, a preset anomaly handling method database is queried to obtain the corresponding anomaly handling method; The vehicle intelligent identification and passage management method for the unmanned toll station further includes: if the network connection status feature data is in a disconnected state, then user matching is performed through a preset high-frequency offline passage record database based on the license plate recognition data and vehicle appearance feature data to obtain the user matching status; if the user matching status is successful, then pre-deduction is performed; if the user matching status is unsuccessful, then temporary passage identity binding is activated, and pre-deduction is performed according to preset deduction rules; if pre-deduction fails, then historical credit assessment data of the passing user is obtained, including passage payment success rate and number of toll evasion records; the passing user's credit rating is obtained based on the passage payment success rate and number of toll evasion records, including excellent credit, good credit, or poor credit; if the credit rating is excellent or good, then credit passage response is activated; if the credit rating is poor, then temporary passage identity binding is activated, and pre-deduction is performed according to preset deduction rules.

2. The vehicle intelligent identification and traffic management method for unmanned toll stations according to claim 1, characterized in that, The step of deducting tolls according to the matched account via a preset IoT communication protocol and performing toll management based on toll deduction status feature data includes: Based on the matched account, obtain the corresponding account balance and preset deduction rules at the preset edge computing node; Based on the account balance and preset deduction rules, toll fees are deducted through a preset Internet of Things communication protocol to obtain toll deduction status feature data, including whether the deduction is completed or not. If the toll deduction is not completed, the toll deduction status data will be transmitted to the toll gate control system via a preset IoT communication protocol, and permission will be granted. If the payment has not been completed, passage will be prohibited.

3. A vehicle intelligent identification and passage management system for unmanned toll stations, characterized in that, The device includes a memory and a processor. The memory contains a program for a vehicle intelligent identification and traffic management method for unmanned toll stations. When the program for the vehicle intelligent identification and traffic management method for unmanned toll stations is executed by the processor, it implements the steps of a vehicle intelligent identification and traffic management method for unmanned toll stations as described in any one of claims 1 to 2.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a vehicle intelligent identification and passage management method program for unmanned toll stations. When the vehicle intelligent identification and passage management method program for unmanned toll stations is executed by a processor, it implements the steps of a vehicle intelligent identification and passage management method for unmanned toll stations as described in any one of claims 1 to 2.

Images