Electronic information-based highway toll collection system

Abstract

This invention relates to the field of highway toll collection and discloses a highway toll collection system based on electronic information. It includes: an entrance roadside controller, multiple gantry roadside controllers along the route, an exit vehicle roadside controller, and a cloud control platform. The cloud control platform includes an event standardization module, a candidate path generation module, a dual-chain evidence construction module, a conflict classification module, a supplementary data collection strategy distribution module, a chain state update module, and a tolling control module. The system uniformly models entrance passage events, gantry passage events, exit passage events, and supplementary passage events, constructing identity consistency chains and path continuity chains respectively. It then executes corresponding supplementary data collection operations based on conflict categories and completes tolling processing based on the updated dual-chain state. This system improves the stability of vehicle identity verification and passage path confirmation, increases the accuracy of toll collection results, and reduces the review pressure under abnormal situations.

Description

Technical Field

[0001] This invention relates to the field of highway toll collection, and more specifically to a highway toll collection system based on electronic information. Background Technology

[0002] Existing highway toll collection systems typically collect data such as electronic tag information, license plate recognition information, vehicle type information, and travel time information when vehicles enter toll station entrances, pass through gantries along the route, and exit toll lanes. The backend platform then uses this data to reconstruct the vehicle's travel path and process the tolls based on the road network topology. To improve the automation of toll collection, some systems also perform centralized comparison of entrance, gantry, and exit data to assist in anomaly detection and toll auditing.

[0003] However, during actual vehicle passage, factors such as weak reading of electronic tags, obscured or damaged license plates, deviations in vehicle type recognition, fluctuations in wheel axle detection, and missing data in some gantry structures can easily disrupt the correspondence between entrance data, gantry data, and exit data. This makes it difficult to maintain a stable relationship between vehicle identity and passage path simultaneously, resulting in unreliable toll collection criteria. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a highway toll collection system based on electronic information, thereby resolving the technical problems existing in the prior art.

[0005] The above-mentioned technical objective of the present invention is achieved through the following technical solution: An electronic information-based highway toll collection system includes: Entrance roadside controller, multiple gantry roadside controllers along the route, exit vehicle roadside controller, and cloud control platform; The entrance roadside controller is used to collect entrance passage events when vehicles enter the highway. The roadside controller for the gantry along the route is used to collect gantry passage events when a vehicle passes through the corresponding gantry along the route. The exit vehicle roadside controller is used to collect exit passage events when vehicles leave the highway, and executes corresponding supplementary collection operations to generate supplementary passage events when it receives a supplementary collection control command. The cloud control platform is communicatively connected to the entrance roadside controller, the gantry roadside controller along the route, and the exit vehicle roadside controller. The cloud control platform includes an event standardization module, a candidate path generation module, a dual-chain evidence construction module, a conflict classification module, a supplementary collection strategy distribution module, a chain status update module, and a billing control module. The event standardization module is used to convert entrance access events, gantry access events, exit access events, and supplementary collection access events into access event records with a unified field structure; The candidate path generation module is used to generate candidate travel paths based on the road network topology and the temporal relationship of each travel event record. The dual-chain evidence construction module is used to construct an identity consistency chain and a path continuity chain for each candidate passage path. The identity consistency chain is used to characterize the identity association relationship of the same vehicle between different passage event records, and the path continuity chain is used to characterize the path association relationship of the vehicle between the entrance station, the gantries along the line, and the exit station. The conflict classification module is used to determine the corresponding conflict category based on the verification results of the identity consistency chain and the path continuity chain; The supplementary sampling strategy issuing module is used to issue corresponding supplementary sampling control commands to the exit vehicle roadside controller and the gantry roadside controller along the route according to the conflict category. The chain state update module is used to update the identity consistency chain and the path continuity chain based on the supplementary passage event; The billing control module is used to determine the billing status of the corresponding vehicle based on the updated identity consistency chain and path continuity chain and to perform the corresponding billing processing.

[0006] Preferably, the entrance roadside controller is connected to the entrance electronic tag reader / writer, the entrance license plate capture device, the entrance vehicle model detection device, and the entrance wheel axle detection device, respectively, to collect entrance passage events including entrance station identification, entrance time information, license plate recognition information, electronic tag recognition information, vehicle model detection information, wheel axle detection information, and entrance image information; The roadside controllers for the gantries along the route are connected to the gantry electronic tag reading and writing equipment and the gantry capture equipment, respectively, to collect gantry passage events including gantry identification, gantry time information, license plate recognition information, electronic tag recognition information, lane markings and gantry image information; The exit roadside controller is connected to the exit electronic tag reader / writer, the exit panoramic capture device, the exit close-up capture device, the exit wheel axle retesting device, and the exit vehicle model retesting device to collect exit passage events including exit station signage, exit time information, license plate recognition information, electronic tag recognition information, vehicle model detection information, wheel axle detection information, exit image information, and toll lane signage.

[0007] Preferably, the unified field structure of the passage event record includes an event source field, a station field, a lane field, an event time field, a license plate field, an electronic tag field, a vehicle type field, a wheel axle field, an image feature index field, and a conflict marker field.

[0008] Preferably, the candidate route generation module generates and filters multiple feasible travel routes corresponding to the same vehicle based on the road network topology connection relationship between the entrance station, the gantries along the route and the exit station, as well as the event time sequence of each travel event record and the preset travel time constraint.

[0009] Preferably, when constructing the identity consistency chain, the dual-chain evidence construction module uses the license plate field and electronic tag field in the passage event record as the main association fields, and the vehicle model field, wheel axle field and image feature index field as auxiliary verification fields. When the license plate fields are the same, or when the license plate fields are ambiguous but the electronic tag fields are the same and the vehicle type and wheel axle fields correspond to each other, the corresponding passage event records are determined to belong to the same vehicle.

[0010] Preferably, when constructing the path continuity chain, the dual-chain evidence construction module connects the traffic event records in series according to the connection order of the entrance station, the gantries along the route, and the exit station in the road network topology, and verifies whether the time interval between two adjacent traffic event records falls within the preset travel time interval of the corresponding road segment, so as to determine whether the candidate traffic path is continuous.

[0011] Preferably, the conflict classification module determines the corresponding conflict category when the following situations occur: When the license plate fields are the same and the electronic tag fields are missing or inconsistent, it is determined to be a weak read conflict of the electronic tag. When the electronic tag fields are the same and there are multiple candidate recognition results in the license plate field, it is determined to be a license plate recognition ambiguity conflict; When the combination relationship between the vehicle model field and the axle field does not meet the pre-stored vehicle model and axle correspondence rules, it is determined that there is a conflict between the vehicle model and the axle. When a candidate passage path lacks a passage event record corresponding to a predetermined gantry, it is determined to be a gantry event missing conflict.

[0012] Preferably, the supplementary acquisition strategy issuing module generates supplementary acquisition control instructions according to the preset mapping relationship between conflict type and supplementary acquisition action. Electronic tag weak read conflict corresponds to secondary electronic tag reading and writing and exit close-up capture, license plate recognition ambiguity conflict corresponds to exit close-up capture, vehicle type and wheel axle conflict corresponds to wheel axle retest and vehicle type retest, and gantry event missing conflict corresponds to adjacent gantry event retrieval.

[0013] Preferably, the supplementary data collection strategy issuing module determines the vehicle's pre-arrival time range based on the occurrence time of the most recent effective gantry passage event and the preset travel time interval from the most recent effective gantry to the corresponding road segment of the target exit toll lane, and issues a supplementary data collection control command to the exit vehicle roadside controller within a preset advance time before the start of the pre-arrival time range, so that the exit vehicle roadside controller can complete the corresponding supplementary data collection parameter switching.

[0014] Preferably, the billing control module is used to perform full-path billing when the updated identity consistency chain and path continuity chain simultaneously meet the direct billing conditions; to perform confirmed road segment toll locking and generate a road segment to be confirmed record when the updated identity consistency chain meets the direct billing conditions and there are unconfirmed road segments in the path continuity chain; and to generate an abnormal billing review record when the updated identity consistency chain and path continuity chain still do not meet the direct billing conditions. The system also includes an evidence storage server connected to the cloud control platform. The evidence storage server is used to store the captured images of the corresponding conflict segments, gantry passage event segments and conflict category identifiers when an abnormal billing review record is generated, and to store a passage event record summary when no abnormal billing review record is generated.

[0015] In summary, the present invention has the following main beneficial effects: This application establishes a unified link for traffic event collection and processing between the entrance roadside controller, the gantry roadside controller along the route, the exit vehicle roadside controller, and the cloud control platform. Through an event standardization module, entrance traffic events, gantry traffic events, exit traffic events, and supplementary traffic events are uniformly converted into traffic event records. Combined with a candidate route generation module, possible vehicle routes are generated and filtered. This achieves the effect of unifying identification results, originally scattered across different stations, devices, and times, into a single toll determination system. The system no longer relies solely on single entry or exit identification results for billing, but can form a continuous description of the passage process for the same vehicle. Therefore, even with weak electronic tag readings, license plate recognition ambiguities, or fluctuations in local gantry information, the overall stability of route determination and vehicle identification can be maintained, thereby improving the completeness and accuracy of toll determination.

[0016] The system constructs an identity consistency chain and a path continuity chain using a dual-chain evidence construction module. A conflict classification module categorizes conflicts such as weak reads of electronic tags, ambiguous license plate recognition, vehicle type and axle conflicts, and missing gantry events. A supplementary data collection strategy issuance module then issues corresponding supplementary data collection control commands before the vehicle reaches or passes through the target exit toll lane, achieving a coordinated processing of anomaly detection, equipment control, and chain state correction. This ensures that when the system detects unconfirmed parts in the identity or path chain, it doesn't immediately resort to manual review or simply estimate charges based on missing information. Instead, it triggers secondary electronic tag readings and writing, close-up capture, axle retesting, vehicle type retesting, or retrieval of adjacent gantry events based on the conflict type. This improves the effectiveness of supplementary data collection results under abnormal conditions, shortens the processing time at exit toll lanes, and reduces the probability of incorrect or missed charges.

[0017] The chain status update module updates the identity consistency chain and path continuity chain based on the supplementary data collection results. The billing control module then performs full-path billing, confirmed segment toll locking, or abnormal billing review based on the updated dual-chain status. Simultaneously, the evidence storage server selectively saves conflicting segments during abnormal billing review, achieving coordinated implementation of toll processing, anomaly tracking, and evidence management. This allows the system to complete full-path billing promptly when the dual chains are closed, and to lock confirmed segments for tolls even when some paths are not yet confirmed, initiating abnormal billing review when necessary, thus balancing real-time billing and accuracy. Furthermore, by focusing on storing only abnormal conflicting segments and only saving summary information for normal passage events, the system reduces evidence storage pressure and subsequent review workload. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall architecture of the highway toll collection system of the present invention; Figure 2 This is a schematic diagram of the access event collection and uploading process of the present invention. Detailed Implementation

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

[0020] Example 1 refer to Figure 1-2 A highway toll collection system based on electronic information, comprising: Entrance roadside controller, multiple gantry roadside controllers along the route, exit vehicle roadside controller, and cloud control platform; The entrance roadside controller is used to collect entrance passage events when vehicles enter the highway. The roadside controller for the gantry along the route is used to collect gantry passage events when a vehicle passes through the corresponding gantry along the route. The exit vehicle roadside controller is used to collect exit passage events when vehicles leave the highway, and executes corresponding supplementary collection operations to generate supplementary passage events when it receives a supplementary collection control command. The cloud control platform is communicatively connected to the entrance roadside controller, the gantry roadside controller along the route, and the exit vehicle roadside controller. The cloud control platform includes an event standardization module, a candidate path generation module, a dual-chain evidence construction module, a conflict classification module, a supplementary collection strategy distribution module, a chain status update module, and a billing control module. The event standardization module is used to convert entrance access events, gantry access events, exit access events, and supplementary collection access events into access event records with a unified field structure; The candidate path generation module is used to generate candidate travel paths based on the road network topology and the temporal relationship of each travel event record. The dual-chain evidence construction module is used to construct an identity consistency chain and a path continuity chain for each candidate passage path. The identity consistency chain is used to characterize the identity association relationship of the same vehicle between different passage event records, and the path continuity chain is used to characterize the path association relationship of the vehicle between the entrance station, the gantries along the line, and the exit station. The conflict classification module is used to determine the corresponding conflict category based on the verification results of the identity consistency chain and the path continuity chain; The supplementary sampling strategy issuing module is used to issue corresponding supplementary sampling control commands to the exit vehicle roadside controller and the gantry roadside controller along the route according to the conflict category. The chain state update module is used to update the identity consistency chain and the path continuity chain based on the supplementary passage event; The billing control module is used to determine the billing status of the corresponding vehicle based on the updated identity consistency chain and path continuity chain and to perform the corresponding billing processing.

[0021] This invention is used to continuously collect, uniformly model, perform dual-chain verification, targeted supplementary collection, and charge control of the passage process of vehicles at highway entrances, gantries along the route, and exit toll lanes. It aims to solve the problems of existing toll collection systems, such as unstable passage path confirmation, miscalculation and omission, prolonged lane dwell time, and high cost of retaining abnormal evidence, which are prone to occur when there are weak readings of electronic tags, ambiguity in license plate recognition, conflicts between vehicle type recognition and wheel axle recognition, missing gantry events, and local communication anomalies.

[0022] The highway toll collection system in this embodiment includes an entrance roadside controller, multiple gantry roadside controllers along the route, an exit vehicle roadside controller, a cloud control platform, and an evidence storage server. The entrance roadside controllers, gantry roadside controllers, and exit vehicle roadside controllers are connected to the cloud control platform via a dedicated toll collection network, and the evidence storage server is also connected to the cloud control platform. Each roadside controller has a local clock synchronization function, and its local clock is periodically calibrated according to the unified time source of the toll collection system to enable time-series comparison and chaining of traffic event records generated at different collection locations.

[0023] The entrance roadside controller is connected to the entrance electronic tag reader / writer, the entrance license plate capture device, the entrance vehicle type detection device, and the entrance wheel axle detection device. The entrance electronic tag reader / writer reads the electronic tag identification information of vehicles passing through the entrance lane. The entrance license plate capture device acquires images of vehicles passing through the entrance lane, and the entrance roadside controller uses a preset recognition program to obtain the license plate recognition information. The entrance vehicle type detection device outputs vehicle type detection information. The entrance wheel axle detection device outputs wheel axle detection information. The entrance roadside controller also records entrance station identification and entrance time information, organizing this information into an entrance passage event.

[0024] Each gantry roadside controller is connected to both a gantry electronic tag reader / writer and a gantry image capture device. The gantry electronic tag reader / writer reads the electronic tag identification information of vehicles passing through the corresponding gantry. The gantry image capture device acquires the gantry image of the corresponding vehicle, and the roadside controller calls a preset recognition program to obtain the license plate recognition information. The roadside controller also records gantry markings, gantry time information, and lane markings, and organizes this information into gantry passage events.

[0025] To support chain replacement in case of missing gantry events, each gantry roadside controller is equipped with a gantry event caching unit. The gantry event caching unit uses a rolling cache method to store gantry passage events within the most recent preset time period. The preset time period is determined jointly based on the maximum reasonable travel time from the corresponding gantry to the downstream exit toll lane, communication delay redundancy, and equipment clock error redundancy. This ensures that when a missing gantry event conflict occurs, candidate gantry passage event records can be retrieved from the local cache of the upstream or downstream adjacent gantry.

[0026] The exit roadside controller is connected to the exit electronic tag reader / writer, the exit panoramic capture device, the exit close-up capture device, the exit wheel axle re-measurement device, and the exit vehicle model re-measurement device. The exit electronic tag reader / writer reads the electronic tag identification information of vehicles in the exit toll lane. The exit panoramic capture device acquires an overall image of the vehicle passing through. The exit close-up capture device acquires high-resolution images of the vehicle's license plate area or a portion of the vehicle's front. The exit wheel axle re-measurement device re-measures the vehicle's wheel axle information when needed. The exit vehicle model re-measurement device re-measures the vehicle's model information when needed. The exit roadside controller records the exit station sign, exit time information, and toll lane sign, and organizes the collected results into exit passage events. Upon receiving a supplementary data collection control command from the cloud control platform, the exit roadside controller executes the corresponding supplementary data collection operation and organizes the data generated by this operation into a supplementary passage event.

[0027] The cloud control platform includes an event standardization module, a candidate path generation module, a dual-chain evidence construction module, a conflict classification module, a supplementary data collection strategy issuance module, a chain status update module, and a billing control module. The event standardization module converts entry passage events, gantry passage events, exit passage events, and supplementary passage events into unified passage event records. The candidate path generation module generates candidate passage paths based on road network topology and event sequence relationships. The dual-chain evidence construction module constructs identity consistency chains and path continuity chains separately. The conflict classification module determines conflict categories based on dual-chain verification results. The supplementary data collection strategy issuance module issues supplementary data collection control commands to the exit vehicle roadside controller and the gantry roadside controllers along the route based on the conflict category. The chain status update module updates the dual-chain status based on supplementary passage events or retrieved candidate gantry passage event records. The billing control module determines the billing status based on the updated dual-chain status and executes corresponding billing processing.

[0028] The evidence storage server is used to store captured images of conflict segments, gantry passage event segments, and conflict category identifiers during abnormal billing review, and only stores passage event record summaries during non-abnormal billing review.

[0029] In this embodiment, the entrance passage event includes at least the entrance station identifier, entrance time information, license plate recognition information, electronic tag recognition information, vehicle type detection information, wheel axle detection information, and entrance image information. The gantry passage event includes at least the gantry identifier, gantry time information, license plate recognition information, electronic tag recognition information, lane identifier, and gantry image information. The exit passage event includes at least the exit station identifier, exit time information, license plate recognition information, electronic tag recognition information, vehicle type detection information, wheel axle detection information, exit image information, and toll lane identifier. The supplementary data collection passage event includes at least the supplementary data collection trigger time, supplementary data collection source device identifier, supplementary data collection result information, and the association identifier with the corresponding vehicle.

[0030] The event standardization module converts all the above-mentioned traffic events into traffic event records. The traffic event records include fields for event source, station, lane, event time, license plate, electronic tag, vehicle type, axle, image feature index, and conflict marker.

[0031] The Event Source field identifies whether the passage event record originates from the entrance, gantry, exit, or is a supplementary data collection. The Station field records the identifier of the entrance station, gantry, or exit station. The Lane field records the entrance lane, gantry lane, or exit toll lane. The Event Time field records the data collection time of each device under a unified time source. The License Plate field records the license plate recognition result and its corresponding recognition confidence information. The Electronic Tag field records the electronic tag recognition result and its read / write status information. The Vehicle Type field records the vehicle type detection result. The Axle field records the axle detection result. The Conflict Marker field records whether the current passage event record has been marked as a candidate conflict event.

[0032] The image feature index field is a fixed-length structured encoding result output by a pre-set feature extraction program based on entrance image information, gantry image information, or exit image information. It is used to assist in identity verification and is not directly used as an independent basis for billing. The fixed-length structured encoding result may include the vehicle's external contour aspect ratio encoding, the vehicle front area texture encoding, the license plate area relative position encoding, and the vehicle body local structure encoding. Under normal billing conditions, the image feature index field is used to assist in the construction of the dual-chain, and the original image does not need to be stored long-term; under abnormal billing review conditions, the evidence storage server stores the original image of the corresponding conflicting segment.

[0033] The candidate route generation module starts with the entrance passage event record and ends with the exit passage event record. It combines the topological connection relationship between the entrance station, the gantries along the route, and the exit station in the target road network to generate one or more candidate passage routes for the corresponding vehicle.

[0034] Specifically, the candidate route generation module first determines the vehicle's starting toll station based on the entrance station identifier, and then retrieves the set of gantries that may be traversed from the entrance station to the corresponding exit station based on the road network topology table. Subsequently, the candidate route generation module concatenates the gantry passage event records that have candidate associations with the vehicle according to the chronological order of the event time field, generating one or more candidate passage routes. For paths that are topologically connected but have missing local gantry events, the candidate route generation module retains them for subsequent conflict classification and chain completion processing.

[0035] To ensure that candidate path generation has clear and feasible boundaries, this example uses the minimum and maximum travel times between adjacent checkpoints as basic constraints. For any two adjacent checkpoints... and Its minimum travel time and maximum travel time are determined according to the following formulas: in, Indicates the check point With checkpoint The mileage length of the path along the road network topology between them; This indicates the reliable maximum speed of traffic on that road segment; This indicates the reliable minimum speed limit for that road segment; This indicates the amount of additional time compensation. This information is derived from electronic maps of highways, gantry deployment data, or basic ledgers of toll road sections. and The data is derived from speed limit information for the corresponding road segment, speed distribution of historical normal traffic samples, and deceleration characteristics of the station area. The data is derived from statistical results of queuing, ramp merging, short-term congestion, and equipment time synchronization errors in historical operational data. Therefore, the preset travel time interval can be obtained based on real road network parameters and historical operational data, rather than being arbitrarily set.

[0036] The identity consistency chain is used to characterize the identity association relationship of the same vehicle across different passage event records. When constructing the identity consistency chain, the dual-chain evidence construction module uses the license plate field and electronic tag field as the primary association fields, and the vehicle model field, axle field, and image feature index field as auxiliary verification fields.

[0037] To clarify how this process is implemented, this embodiment defines an identity consistency score. for: in, Indicates the degree of license plate matching. Indicates the electronic tag matching degree. Indicates vehicle model consistency. Indicates wheel and axle alignment. Indicates the consistency of image features; , , , and Let each of the five indicators represent its weight, and satisfy the following conditions: Among them, license plate matching degree The electronic tag matching degree is determined based on whether the license plate field matches, whether there is any character ambiguity, and the confidence difference of each candidate result. The consistency is determined based on whether the electronic tag fields are consistent, and whether there are weak reads, intermittent reads, or verification failures. Vehicle model consistency. Determined based on whether the vehicle model field matches. Wheel and axle consistency. Determined based on whether the wheel axle field is consistent. Image feature consistency. Determined based on the similarity of the image feature index fields.

[0038] The weight , , , and The initial values ​​are obtained through offline calibration using historical normal and abnormal traffic samples that have been manually verified. Under the same road segment, similar equipment, and the same recognition model version, the above weights remain fixed within a calibration period. If there are significant changes in equipment type, recognition model, or toll collection environment, recalibration is performed.

[0039] In this embodiment, a first identity threshold is set. Second identity threshold And satisfy .when When the corresponding passage event records are determined to belong to the same vehicle, the identity consistency chain is closed; when When the identity is determined to be in an unconfirmed state, a supplementary data collection needs to be triggered; when At that time, it was determined to be an identity conflict. and The distribution results derived from historical normal and abnormal samples can be determined through offline calibration.

[0040] When the license plate fields are the same, or the license plate fields are ambiguous but the electronic tag fields are the same and the vehicle model and wheel axle fields correspond to each other, and the corresponding identity consistency score is satisfied. At that time, the dual-chain evidence construction module determines that the corresponding passage event records belong to the same vehicle. Therefore, the correspondence in the claims has a clear basis for determination in the embodiments.

[0041] The path continuity chain is used to characterize the path association relationship of the same vehicle between the entrance station, along-line gantries, and the exit station. When constructing the path continuity chain, the dual-chain evidence construction module concatenates the passage event records on the candidate passage path according to the connection order of the entrance station, along-line gantries, and exit station in the road network topology, and verifies at least the following two items for any two adjacent passage event records: first, whether the corresponding stations are connected in the road network topology; and second, whether the time interval between the two passage event records falls within the preset travel time interval of the corresponding road segment.

[0042] To ensure that the path continuity chain has clear decision boundaries, this example defines the adjacent verification results. for: in, Indicates the first candidate path The event and the first The time interval between events; and These represent the minimum and maximum travel times between adjacent checkpoints, respectively. Further, a path continuity score is defined. for: in, This indicates the total number of passage event records on the candidate passage path. When, it means that all adjacent events on the candidate path satisfy the requirements of topological continuity and temporal continuity, and the path continuity chain is closed; when When, it indicates that there is at least one unconfirmed segment in the path continuity chain; when If a candidate path has obvious topological breaks, it is determined that the path cannot be used as a valid billing path.

[0043] Among them, the unconfirmed road segment refers to a road segment in the path continuity chain that has been closed or has met the billing requirements, but cannot be confirmed temporarily due to missing gantry events or failure to meet the continuity verification of adjacent events.

[0044] The conflict classification module determines the conflict category based on the verification results of the identity consistency chain and the path continuity chain. In this embodiment, the following four types of conflicts are included.

[0045] The first category is weak read conflicts of electronic tags. When the license plate field is the same or the identity consistency chain initially determines that it is the same vehicle, but the electronic tag field is empty, the electronic tag verification fails, the return value of the same vehicle is unstable within the continuous read and write cycle, or the electronic tag fields of the entry and exit are still inconsistent after the initial screening by the identity consistency chain, it is determined to be a weak read conflict of electronic tags.

[0046] The second category is license plate recognition ambiguity conflict. When electronic tag fields are identical or the path continuity chain shows corresponding passage events belonging to the same path, but the license plate field has multiple candidate recognition results, it is determined to be a license plate recognition ambiguity conflict. "Multiple candidate recognition results for the license plate field" refers to license plate results where, after license plate recognition, two or more confidence values ​​fall within a preset difference threshold range for the same image frame or consecutive image frames of the same vehicle. The preset difference threshold is obtained statistically based on the distribution of confidence values ​​of historical license plate recognition results.

[0047] The third category is vehicle type and axle conflict. A conflict is identified when the combination of the vehicle type field and the axle field does not meet the pre-stored vehicle type and axle correspondence rules. These pre-stored vehicle type and axle correspondence rules are derived from the vehicle type classification rules used in the target road network, the vehicle type detection equipment calibration results, and the axle detection equipment calibration results.

[0048] The fourth category is gantry event missing conflict. When a candidate passage path is topologically connected, but the corresponding passage event record for the scheduled gantry is missing, and the upstream and downstream gantry events can still form a sandwich relationship with the missing segment in terms of time and topology, it is identified as a gantry event missing conflict.

[0049] The aforementioned conflict categories can appear individually or in combination. When they appear in combination, the supplementary sampling strategy distribution module prioritizes processing the conflict categories that have a greater impact on the closure of the identity consistency chain, and then processes the conflict categories that have a greater impact on the closure of the path continuity chain.

[0050] The supplementary data acquisition strategy distribution module generates supplementary data acquisition control instructions according to the preset mapping relationship between conflict categories and supplementary data acquisition actions. Among them, weak read conflicts of electronic tags correspond to secondary electronic tag reading and writing and close-up capture at the exit; ambiguous conflicts of license plate recognition correspond to close-up capture at the exit; conflicts between vehicle type and wheel axle correspond to wheel axle retest and vehicle type retest; and conflicts with missing gantry events correspond to retrieval of events from adjacent gantry locations.

[0051] The supplementary data collection in this application is not a post-event certificate collection after the vehicle leaves the toll lane, but rather a pre-emptive switching of the exit lane data collection parameters and retrieval of gantry cache records before the vehicle reaches the target exit toll lane or during the process of passing through the target exit toll lane.

[0052] Under the supplementary acquisition strategy, the module is based on the occurrence time of the most recent valid gantry passage event. The minimum and maximum travel times from the nearest effective gantry to the corresponding road segment of the target exit toll lane are used to determine the vehicle's estimated arrival time range. : in, Indicates the most recently active gantry. Indicates the target exit toll lane. and These represent the minimum and maximum travel times from the nearest available gantry to the target exit toll lane, respectively. Indicates the expected arrival buffer amount. (Selected) The data is derived from historical statistics on equipment communication delays, lane queue length fluctuations, and station deceleration fluctuations.

[0053] The supplementary data collection strategy distribution module issues a supplementary data collection control command to the exit vehicle roadside controller within a preset lead time before the start of the expected arrival time range. This preset lead time is determined based on the equipment response time required for the exit vehicle roadside controller to complete the switching of electronic tag reading / writing parameters, close-up capture parameters, wheel axle re-measurement trigger parameters, and vehicle model re-measurement trigger parameters, as well as the average lane queuing change time in the most recent statistical period. Therefore, the preset lead time has a clear source and is not arbitrarily set.

[0054] For weak read conflicts of electronic tags, the exit electronic tag reading and writing equipment switches to supplementary acquisition reading and writing mode, extends the reading and writing window duration, and increases the polling frequency. At the same time, the exit close-range capture equipment increases the priority of license plate area acquisition. For license plate recognition ambiguity conflicts, the exit close-range capture equipment switches to high-resolution mode and increases the number of close-range acquisition frames. For vehicle type and axle conflicts, the exit axle retesting equipment and the exit vehicle type retesting equipment are triggered when the vehicle enters the corresponding detection area. For gantry event missing conflicts, the supplementary acquisition strategy distribution module simultaneously sends event retrieval requests to the upstream adjacent gantry roadside controller and the downstream adjacent gantry roadside controller of the missing gantry to retrieve candidate gantry passage event records in their respective gantry event cache units.

[0055] In the event of missing gantry events, multiple candidate gantry passage event records may exist in the event buffer units of the upstream and downstream adjacent gantry. To avoid unclear chain boundaries, this embodiment specifies the following selection rule.

[0056] When multiple candidate gantry passage event records exist in adjacent gantry event cache units, the chain state update module prioritizes selecting the candidate gantry passage event record that simultaneously satisfies topological order constraints and preset travel time interval constraints, and has the highest identity consistency score, as the supplementary passage event. If two or more candidate gantry passage event records have the same identity consistency score, the deviation between the event time of each candidate gantry passage event record and the theoretical passage time of the missing gantry is further compared, and the candidate gantry passage event record with the smaller deviation value is prioritized. The theoretical passage time of the missing gantry is determined based on the effective event time upstream of the missing gantry, the length of the upstream and downstream road segments, and the average passage time in historical normal samples.

[0057] Through the above selection rules, the chain replacement process under the conflict of missing gantry events has a clear objective basis for judgment, avoiding unclear selection boundaries between different candidate gantry passage event records.

[0058] Upon receiving a supplementary access event or retrieving a candidate gantry access event record, the chain state update module recalculates the identity consistency score. and path continuity score And update the identity consistency chain and path continuity chain.

[0059] For weak read conflicts of electronic tags, if the secondary electronic tag read / write results after re-collection are consistent with the fields of the input electronic tag, the electronic tag matching degree will be improved. If the license plate field of the most recently captured vehicle is restored from an ambiguous state to a unique recognition result, the license plate matching accuracy will be improved accordingly. For vehicle model and axle conflicts, if the retest results meet the pre-stored vehicle model and axle correspondence rules, the corresponding vehicle model field and axle field will be updated to reflect the retest results. For gantry event missing conflicts, if a candidate gantry passage event record that meets the topological order constraints and preset travel time interval constraints is retrieved according to the selection rules, the candidate gantry passage event record will be written into the path continuity chain as a supplementary passage event.

[0060] The billing control module determines the billing status of the corresponding vehicle based on the updated identity consistency chain and path continuity chain. In this embodiment, the following three billing statuses are included.

[0061] The first scenario is full-path billing. When the updated identity consistency score meets... And the updated path continuity score satisfies At that time, the billing control module performs full-path billing.

[0062] The second type is the confirmed toll-locked status of the road segment. When the updated identity consistency score meets... And the updated path continuity score satisfies At this time, the billing control module performs fee locking for confirmed road segments and generates records for road segments awaiting confirmation. Fee locking for confirmed road segments refers to temporarily storing and recording the corresponding costs for road segments already confirmed by the path continuity chain and generating a non-repeating billing identifier. The final full-path settlement result is not output until all road segments awaiting confirmation are completed. The records for road segments awaiting confirmation include the confirmed road segment identifier, the missing gantry segment identifier, the locked costs, and the pending billing status identifier.

[0063] The third type is the abnormal billing review status. When the updated identity consistency score meets... Or the updated path continuity score satisfies If, after supplementary data collection, critical conflicts remain unresolved, the billing control module generates an abnormal billing review record. These critical conflicts include unresolved electronic tag weak read conflicts, license plate recognition ambiguity conflicts, vehicle type and axle group conflicts, and gantry event missing conflicts.

[0064] The three billing states correspond to clear objective trigger boundaries, preventing the billing control module from being perceived as a result-limited module driven solely by abstract rules.

[0065] When the billing control module generates an abnormal billing review record, the cloud control platform writes the captured image of the corresponding conflict segment, the gantry passage event segment, and the conflict category identifier to the evidence storage server. The captured image preferably includes relevant portions of the entrance image information, the gantry image information that triggered the conflict, the exit image information, and the supplementary close-up image information. The gantry passage event segment includes at least the corresponding gantry's station field, event time field, license plate field, electronic tag field, image feature index field, and conflict marker field.

[0066] When no abnormal billing review record is generated, the evidence storage server only stores a summary of the passage event record. The summary of the passage event record includes the event source field, station field, event time field, license plate field, electronic tag field, billing status, and billing result.

[0067] The road network topology is derived from highway electronic maps, gantry deployment records, and toll station configuration data; Road segment length It originates from basic road network data or calibrated inter-gantry mileage data; Selectable maximum passage speed and optional minimum passage speed The data is derived from the corresponding road segment speed limit information, historical normal traffic sample speed distribution, and station area deceleration characteristics. Additional time compensation and expected arrival buffer amount Statistics on queuing, short-term congestion, equipment communication delays, and clock errors are derived from historical operational data. Weights in identity consistency scoring , , , and and threshold and The results are derived from offline calibration of historical normal and abnormal samples that have been manually verified. The confidence difference threshold for license plate candidate results is derived from the statistical distribution of confidence values ​​of historical license plate recognition results; The preset advance time is derived from the equipment response time required for the exit vehicle roadside controller to complete the parameter switching and the average lane queuing change time in the most recent statistical period. The vehicle type and axle correspondence rules are derived from the toll vehicle type classification rules and the on-site equipment calibration results.

[0068] The working principle of this application is as follows: When a vehicle enters the highway, the entrance roadside controller collects entrance passage events; when the vehicle passes through each gantry along the route, the roadside controllers of each gantry collect the corresponding gantry passage events; when the vehicle enters the exit toll lane, the exit roadside controller collects the exit passage events and sends the above passage events to the cloud control platform. The cloud control platform first converts the entrance passage events, gantry passage events, exit events, and any subsequent supplementary passage events into passage event records with a unified field structure through the event standardization module. Then, the candidate path generation module combines the road network topology and the temporal relationship of each passage event record to generate candidate passage paths for the corresponding vehicle. Subsequently, the dual-chain evidence construction module constructs an identity consistency chain and a path continuity chain to determine whether different passage event records belong to the same vehicle and whether the corresponding passage paths are continuous and complete. If both the identity consistency chain and the path continuity chain meet the billing conditions, the billing control module directly executes full-path billing. If there are unconfirmed parts in either the identity consistency chain or the path continuity chain, the conflict classification module determines the corresponding conflict category, and the supplementary data collection strategy issuance module issues supplementary data collection control instructions to the exit vehicle roadside controller and / or the gantry roadside controller along the route based on the conflict category. This triggers secondary electronic tag reading and writing, close-up capture, wheel axle retesting, vehicle type retesting, or adjacent gantry event retrieval. The chain status update module then updates the identity consistency chain and the path continuity chain based on the supplementary data collection results. The billing control module then executes full-path billing, confirmed road segment toll locking, or abnormal billing review based on the updated dual-chain status. For abnormal billing review status, the evidence storage server saves the captured image of the corresponding conflict segment, the gantry passage event segment, and the conflict category identifier. For normal billing status, only the passage event record summary is saved. Therefore, this application improves the stability of vehicle route confirmation and the accuracy of toll collection results by linking the collection of passage events, candidate route generation, dual-chain verification, pre-arrival supplementary collection, chain status update and toll status switching in a closed loop. This can improve the stability of vehicle route confirmation and the accuracy of toll collection results in complex situations such as weak electronic tag reading, license plate recognition ambiguity, vehicle type and wheel axle conflict, and missing gantry events.

[0069] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A highway toll collection system based on electronic information, characterized in that, include: Entrance roadside controller, multiple gantry roadside controllers along the route, exit vehicle roadside controller and cloud control platform; The entrance roadside controller is used to collect entrance passage events when vehicles enter the highway. The roadside controller for the gantry along the route is used to collect gantry passage events when a vehicle passes through the corresponding gantry along the route. The exit vehicle roadside controller is used to collect exit passage events when vehicles leave the highway, and executes corresponding supplementary collection operations to generate supplementary passage events when it receives a supplementary collection control command. The cloud control platform is communicatively connected to the entrance roadside controller, the gantry roadside controller along the route, and the exit vehicle roadside controller. The cloud control platform includes an event standardization module, a candidate path generation module, a dual-chain evidence construction module, a conflict classification module, a supplementary collection strategy distribution module, a chain status update module, and a billing control module. The event standardization module is used to convert entrance access events, gantry access events, exit access events, and supplementary data collection events into access event records with a unified field structure; The candidate path generation module is used to generate candidate travel paths based on the road network topology and the temporal relationship of each travel event record. The dual-chain evidence construction module is used to construct an identity consistency chain and a path continuity chain for each candidate passage path. The identity consistency chain is used to characterize the identity association relationship of the same vehicle between different passage event records, and the path continuity chain is used to characterize the path association relationship of the vehicle between the entrance station, the gantries along the line, and the exit station. The conflict classification module is used to determine the corresponding conflict category based on the verification results of the identity consistency chain and the path continuity chain. The supplementary sampling strategy issuing module is used to issue corresponding supplementary sampling control commands to the exit vehicle roadside controller and the gantry roadside controller along the route according to the conflict category. The chain state update module is used to update the identity consistency chain and the path continuity chain based on the supplementary passage event; The billing control module is used to determine the billing status of the corresponding vehicle based on the updated identity consistency chain and path continuity chain and to perform the corresponding billing processing.

2. The highway toll collection system based on electronic information according to claim 1, characterized in that, The entrance roadside controller is connected to the entrance electronic tag reader / writer, the entrance license plate capture device, the entrance vehicle type detection device, and the entrance wheel axle detection device to collect entrance passage events including entrance station identification, entrance time information, license plate recognition information, electronic tag recognition information, vehicle type detection information, wheel axle detection information, and entrance image information. The roadside controllers for the gantries along the route are connected to the gantry electronic tag reading and writing equipment and the gantry capture equipment, respectively, to collect gantry passage events including gantry identification, gantry time information, license plate recognition information, electronic tag recognition information, lane markings and gantry image information; The exit roadside controller is connected to the exit electronic tag reader / writer, the exit panoramic capture device, the exit close-up capture device, the exit wheel axle retesting device, and the exit vehicle model retesting device to collect exit passage events including exit station signage, exit time information, license plate recognition information, electronic tag recognition information, vehicle model detection information, wheel axle detection information, exit image information, and toll lane signage.

3. A highway toll collection system based on electronic information according to claim 2, characterized in that, The unified field structure of the passage event record includes the event source field, station field, lane field, event time field, license plate field, electronic tag field, vehicle type field, wheel axle field, image feature index field, and conflict marker field.

4. A highway toll collection system based on electronic information according to claim 3, characterized in that, The candidate route generation module generates and filters multiple feasible routes for the same vehicle based on the road network topology connection between the entrance station, the gantries along the route and the exit station, as well as the chronological order of the events recorded in each passage event and the preset travel time constraints.

5. A highway toll collection system based on electronic information according to claim 4, characterized in that, When constructing the identity consistency chain, the dual-chain evidence construction module uses the license plate field and electronic tag field in the passage event record as the main association fields, and the vehicle type field, wheel axle field and image feature index field as auxiliary verification fields. When the license plate fields are the same, or when the license plate fields are ambiguous but the electronic tag fields are the same and the vehicle type and wheel axle fields correspond to each other, the corresponding passage event records are determined to belong to the same vehicle.

6. A highway toll collection system based on electronic information according to claim 5, characterized in that, When constructing the path continuity chain, the dual-chain evidence construction module connects the traffic event records in series according to the connection order of the entrance station, the gantries along the route, and the exit station in the road network topology, and verifies whether the time interval between two adjacent traffic event records falls within the preset travel time interval of the corresponding road segment, so as to determine whether the candidate traffic path is continuous.

7. A highway toll collection system based on electronic information according to claim 6, characterized in that, The conflict classification module determines the corresponding conflict category when the following situations occur: When the license plate fields are the same and the electronic tag fields are missing or inconsistent, it is determined to be a weak read conflict of the electronic tag. When the electronic tag fields are the same and there are multiple candidate recognition results in the license plate field, it is determined to be a license plate recognition ambiguity conflict; When the combination relationship between the vehicle model field and the axle field does not meet the pre-stored vehicle model and axle correspondence rules, it is determined that there is a conflict between the vehicle model and the axle. When a candidate passage path lacks a passage event record corresponding to a predetermined gantry, it is determined to be a gantry event missing conflict.

8. A highway toll collection system based on electronic information according to claim 7, characterized in that, The supplementary acquisition strategy distribution module generates supplementary acquisition control instructions according to the preset mapping relationship between conflict type and supplementary acquisition action. Electronic tag weak read conflict corresponds to secondary electronic tag reading and writing and exit close-up capture. License plate recognition ambiguity conflict corresponds to exit close-up capture. Vehicle type and wheel axle conflict corresponds to wheel axle retest and vehicle type retest. Gantry event missing conflict corresponds to adjacent gantry event retrieval.

9. A highway toll collection system based on electronic information according to claim 8, characterized in that, The supplementary data collection strategy issuing module determines the vehicle's pre-arrival time range based on the occurrence time of the most recent effective gantry passage event and the preset travel time interval from the most recent effective gantry to the corresponding road segment of the target exit toll lane. Within a preset advance time before the start of the pre-arrival time range, the module issues a supplementary data collection control command to the exit vehicle roadside controller so that the exit vehicle roadside controller can complete the corresponding supplementary data collection parameter switching.

10. A highway toll collection system based on electronic information according to claim 9, characterized in that, The billing control module is used to perform full-path billing when the updated identity consistency chain and path continuity chain simultaneously meet the direct billing conditions; to perform confirmed road segment toll locking and generate a road segment to be confirmed record when the updated identity consistency chain meets the direct billing conditions and there are unconfirmed road segments in the path continuity chain; and to generate an abnormal billing review record when the updated identity consistency chain and path continuity chain still do not meet the direct billing conditions. The system also includes an evidence storage server connected to the cloud control platform. The evidence storage server is used to store the captured images of the corresponding conflict segments, gantry passage event segments and conflict category identifiers when an abnormal billing review record is generated, and to store the passage event record summary when no abnormal billing review record is generated.

Images