Highway vehicle fee evasion inspection method and system based on multi-source data fusion

Abstract

The application belongs to the technical field of intelligent transportation, and provides a highway vehicle fee evasion inspection method and system based on multi-source data fusion, which comprises the following steps: obtaining the time stamp and vehicle characteristic information of a vehicle when the vehicle enters and exits a highway; obtaining the actual dynamic driving time of the vehicle based on the time stamp; when the actual dynamic driving time exceeds the dynamic driving time threshold, there is a potential fee evasion phenomenon; otherwise, the path gantry information of the vehicle during driving is obtained to determine whether there is a potential fee evasion phenomenon; when the path gantry information is missing, there is a potential fee evasion phenomenon; otherwise, the vehicle characteristic information is used to determine whether the vehicle has a potential fee evasion phenomenon; when the vehicle characteristic information of the vehicle when entering and exiting the highway does not match, it is determined that there is a potential fee evasion phenomenon, and the highway vehicle fee evasion inspection based on multi-source data fusion is completed; and then the potential fee evasion behavior of the highway vehicle is accurately and efficiently identified, and the highway toll rights and interests are protected.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent transportation, and particularly relates to a highway vehicle fee evasion inspection method and system based on multi-source data fusion. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute prior art.

[0003] The highway is a multi-lane highway specially used for the split direction and split lane driving of automobiles, with all controlled access; due to the relatively high cost and management cost, the highway is generally charged by referring to the driving distance, bridge and tunnel conditions. At present, with the continuous popularization of non-stop charging, the phenomenon of highway toll evasion sometimes occurs.

[0004] At present, the main ways of highway fee evasion include changing the charging vehicle type, inconsistent entry and exit information, card switching, inconsistent entry and exit license plates, inconsistent charging path and license plate recognition path, inconsistent entry and exit vehicle type, abnormal large vehicle small license plate, abnormal large vehicle small axle number, abnormal large vehicle small vehicle type, malicious U / J driving, no snapshot throughout the journey, overtime, bottom-up charging, etc.

[0005] For highway fee evasion vehicles, the current inspection method is that the inspection personnel manually analyze each item to find suspected fee evasion vehicles, and then manually verify whether it is a highway vehicle fee evasion; however, the manual inspection method usually requires a large amount of manpower, and the efficiency needs to be improved. At the same time, the existing highway vehicle fee evasion inspection adopts single-dimensional detection, the traditional inspection method relies on a single data source, and cannot adapt to the dynamic changes of weather, vehicle type and road conditions within a fixed time threshold, lacks multi-dimensional data spatio-temporal correlation analysis, and thus the recognition degree of fee evasion behavior is not high. Therefore, how to inspect the highway vehicle fee evasion behavior through effective technical means to maintain the normal and orderly progress of highway toll management is a problem that should be paid attention to. SUMMARY

[0006] To solve the above problems, the present application provides a highway vehicle fee evasion inspection method and system based on multi-source data fusion, which acquires the time of vehicle entering and exiting the highway, the vehicle type characteristics and the license plate information, combines the vehicle portal information acquired on the highway, and comprehensively judges whether the vehicle has a path abnormality and other fee evasion behaviors according to the acquired information, accurately and efficiently identifies the potential fee evasion behavior of the highway vehicle, and safeguards the rights and interests of highway tolls.

[0007] According to some embodiments, the first aspect of the present application provides a highway vehicle fee evasion inspection method based on multi-source data fusion, which adopts the following technical scheme:

[0008] A highway vehicle fee evasion inspection method based on multi-source data fusion, comprising:

[0009] Obtaining the time stamp and vehicle characteristic information of the vehicle when entering and leaving the highway;

[0010] Based on the obtained time stamp, the actual dynamic driving time of the vehicle on the highway is obtained, and when the actual dynamic driving time exceeds the dynamic driving time threshold, the vehicle has a potential fee evasion phenomenon, otherwise, by obtaining the path gantry information of the vehicle during the driving process on the highway, it is judged whether the vehicle has a potential fee evasion phenomenon;

[0011] When the obtained path gantry information has information missing, the vehicle has a potential fee evasion phenomenon, otherwise, according to the obtained vehicle characteristic information, it is judged whether the vehicle has a potential fee evasion phenomenon;

[0012] When the obtained vehicle characteristic information of the vehicle when entering and leaving the highway does not match, it is judged that the vehicle has a potential fee evasion phenomenon, and the highway vehicle fee evasion inspection based on multi-source data fusion is completed.

[0013] As a further technical limitation, the vehicle has a potential fee evasion phenomenon, including driving time anomaly, vehicle information mismatch, and path gantry information missing; the driving time anomaly includes J-type fee evasion, long-distance driving and short-distance buying, and exit without card; the vehicle information mismatch includes half-way off-hanging and large vehicle with small tag; the path gantry information missing includes OBU information shielding, CPC card reverse card.

[0014] As a further technical limitation, the acquisition process of the dynamic driving time threshold is:

[0015] Obtaining the historical traffic data and current traffic data of the target section;

[0016] Based on the obtained historical traffic data, the key features affecting the driving time are extracted;

[0017] According to the extracted key features, a target section traffic sub-model of each key feature is constructed;

[0018] According to the obtained current traffic data, the matching target section traffic sub-model is selected;

[0019] The dynamic driving time threshold is calculated in combination with the selected target section traffic sub-model.

[0020] As a further technical limitation, in the process of judging whether the vehicle has a potential fee evasion phenomenon by acquiring the path gantry information of the vehicle during the driving process on the expressway, the number of missing path gantry information is detected in real time by the path gantry, and when the number of continuously missing path gantries is less than a preset value, the track space-time restoration mechanism is automatically triggered, and a time window constraint is constructed by combining the passing time stamp of the adjacent path gantry and the design speed of the expressway. A hidden Markov model is used to define the hidden state set as the actual gantry node and its virtual topological connection state, and a space-time transition probability matrix is introduced, and the state transition weight is dynamically calculated and generated by the distance between the gantries and the average driving speed of the vehicle. Based on the topological connectivity of the road network, a candidate virtual gantry set is generated, and a space-time check is performed in combination with the passing time interval of the adjacent gantries. A Bayesian inference network is constructed, and multi-dimensional probability calculation is performed by combining vehicle historical path preference, real-time traffic state and time series data. The first N high-confidence track hypotheses are output, each path is accompanied by a space-time matching degree evaluation value, and a path restoration credibility analysis report containing time dimension verification is generated. When the number of missing path gantries is not less than the preset value and the space-time deviation rate of the derived path from the shortest charging path exceeds the set value, the fee evasion inspection is automatically activated.

[0021] As a further technical limitation, in the process of judging whether the vehicle has a potential fee evasion phenomenon according to the acquired vehicle characteristic information, vehicle characteristic information is collected when the vehicle enters and exits the expressway, and the number of vehicle axles is detected in real time. The vehicle axle number information acquired by the path gantry along the way is synchronously called, and a multi-stage verification method is used to match the vehicle characteristic information to determine whether the vehicle has a potential fee evasion phenomenon.

[0022] As a further technical limitation, the actual dynamic driving time of the vehicle on the expressway is the time difference between the time stamp when the vehicle exits the expressway and the time stamp when the vehicle enters the expressway.

[0023] According to some embodiments, the second aspect of the present application provides a highway vehicle fee evasion inspection system based on multi-source data fusion, which adopts the following technical scheme:

[0024] A highway vehicle fee evasion inspection system based on multi-source data fusion, comprising:

[0025] An acquisition module configured to acquire the time stamp and vehicle characteristic information when the vehicle enters and exits the expressway;

[0026] A judgment module is configured to obtain an actual dynamic driving time of the vehicle on the expressway based on the obtained timestamp, and when the actual dynamic driving time exceeds a dynamic driving time threshold, the vehicle has a potential fee evasion phenomenon, otherwise, the judgment module is configured to obtain path gantry information of the vehicle during driving on the expressway to determine whether the vehicle has a potential fee evasion phenomenon, and when the obtained path gantry information has information missing, the vehicle has a potential fee evasion phenomenon, otherwise, the judgment module is configured to determine whether the vehicle has a potential fee evasion phenomenon according to the obtained vehicle characteristic information, and when the obtained vehicle characteristic information of the vehicle driving into the expressway and driving out of the expressway is not matched, it is determined that the vehicle has a potential fee evasion phenomenon, and the highway vehicle fee evasion inspection based on multi-source data fusion is completed.

[0027] According to some embodiments, a third aspect of the present application provides a computer readable storage medium, adopting the technical scheme as follows:

[0028] A computer readable storage medium, having a program stored thereon, which, when executed by a processor, implements the steps in the highway vehicle fee evasion inspection method based on multi-source data fusion according to the first aspect of the present application.

[0029] According to some embodiments, a fourth aspect of the present application provides an electronic device, adopting the technical scheme as follows:

[0030] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor, and the processor implements the steps in the highway vehicle fee evasion inspection method based on multi-source data fusion according to the first aspect of the present application when executing the program.

[0031] According to some embodiments, a fifth aspect of the present application provides a computer program product, adopting the technical scheme as follows:

[0032] A computer program product includes software code, and the program in the software code implements the steps in the highway vehicle fee evasion inspection method based on multi-source data fusion according to the first aspect of the present application.

[0033] Compared with the prior art, the present application has the following beneficial effects:

[0034] The present application obtains the time of the vehicle entering and leaving the expressway, the vehicle model characteristics, and the license plate information, combines the obtained vehicle gantry information on the expressway, and comprehensively judges whether the vehicle has a path abnormality and other fee evasion behaviors according to the obtained information, accurately and efficiently identifies the potential fee evasion behavior according to the multi-element information of the vehicle, and protects the rights and interests of the expressway toll. BRIEF DESCRIPTION OF DRAWINGS

[0035] The description and drawings of the accompanying drawings, which form a part of this example, are presented to provide a further understanding of this example, illustrate the preferred example and explain the principles of the example, but are not intended to limit the example.

[0036] Figure 1 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0037] Figure 2 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0038] Figure 3 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0039] Figure 4 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0040] Figure 5 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0041] Figure 6 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0042] Figure 7 Flow chart of the method for checking the highway vehicle fee evasion based on multi-source data fusion in the first example of the present application;

[0043] Figure 8 Structure block diagram of the system for checking the highway vehicle fee evasion based on multi-source data fusion in the second example of the present application. DETAILED DESCRIPTION

[0044] The present application will be further described with reference to the drawings and examples.

[0045] It should be noted that the following detailed description is merely exemplary in nature and is intended to provide further description of the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0046] It is also to be understood that the terminology used herein is for the purpose of describing the particular examples only and is not intended to be limiting of the example. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising", when used in this specification, specify the presence of stated features, steps, operations, elements, components and / or groups thereof, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components and / or groups thereof.

[0047] In the present application, the terms such as "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "side", "bottom", etc. indicate the orientation or positional relationship shown in the drawings, which are only the relationship words determined for the convenience of describing the structural relationship of the components or elements of the present application, and cannot be understood as the limitation of the present application.

[0048] In the present application, the terms such as "fixedly connected", "connected", "connected" should be understood broadly, which can be fixedly connected, integrally connected or detachably connected; it can be directly connected or indirectly connected through an intermediate medium. For relevant scientific research or technical personnel in the field, the specific meaning of the above terms in the present application can be determined according to the specific circumstances, and cannot be understood as the limitation of the present application.

[0049] In the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0050] Embodiment one

[0051] The present application embodiment one introduces a kind of highway vehicle escape fee inspection method based on multi-source data fusion.

[0052] As shown in Figure 1 And Figure 2 A kind of highway vehicle escape fee inspection method based on multi-source data fusion, comprising:

[0053] Obtain the time stamp and vehicle characteristic information of vehicle when driving into highway and driving out of highway;

[0054] Based on the obtained time stamp, the actual dynamic driving time of vehicle on highway is obtained, when the actual dynamic driving time exceeds dynamic driving time threshold, then vehicle exists potential escape fee phenomenon, otherwise, whether vehicle exists potential escape fee phenomenon is judged by obtaining path gantry information in the process of vehicle driving on highway;

[0055] When the obtained path gantry information exists information missing, then vehicle exists potential escape fee phenomenon, otherwise, whether vehicle exists potential escape fee phenomenon is judged according to the obtained vehicle characteristic information;

[0056] When the obtained vehicle characteristic information of vehicle when driving into highway and driving out of highway is not matched, then it is judged that vehicle exists potential escape fee phenomenon, and the highway vehicle escape fee inspection based on multi-source data fusion is completed.

[0057] As shown in Figure 3The types of toll evasion on highways shown include potential toll evasion phenomena such as abnormal travel time, vehicle information mismatch, and missing route gantry information; abnormal travel time includes J-type toll evasion, traveling long distances to buy short distances, and no card at the exit; vehicle information mismatch includes trailer swapping and large vehicles with small labels; missing route gantry information includes OBU information obscured and CPC card swapping.

[0058] like Figure 4 The flowchart shown is for calculating vehicle travel time. It is applicable to calculating the time taken for a vehicle to travel from entering the highway to exiting the highway, and specifically includes the following steps:

[0059] S310: When a vehicle passes through a highway entrance toll station, if it is a manual toll lane, the staff will record the vehicle's entry timestamp through the system while the vehicle receives the toll card; if it is an ETC lane, the electronic tag on the vehicle communicates with the lane equipment, and the equipment automatically records the time the vehicle passes through the entrance.

[0060] S320: When a vehicle arrives at the highway exit toll station, there will also be a corresponding recording operation. In the manual toll lane, the staff will record the time the vehicle arrives at the exit in the system based on the vehicle's toll card information and calculate the travel time; for the ETC lane, the system will automatically identify the vehicle's electronic tag, record the timestamp of the vehicle passing through the exit, and compare it with the entry time for calculation.

[0061] S330: The system calculates the vehicle's travel time on the highway by subtracting the entry time from the exit time. It compares the actual travel time with the estimated time. If the difference is within a set threshold, the data is considered normal, and billing or processing is performed as usual. If the difference exceeds the threshold, an exception handling procedure is initiated.

[0062] like Figure 5 The illustrated calculation process for the dynamic threshold of the longest vehicle travel time is applicable to calculating dynamic time thresholds for different road segments and scenarios. It specifically includes the following steps:

[0063] S410: Obtain historical traffic data for the target road segment, including vehicle travel time, traffic flow, weather conditions, and time period information; clean the data, remove outliers and missing values, and perform standardization processing.

[0064] S420: Based on historical data, key features affecting travel time are extracted, distinguishing between morning peak, evening peak, and off-peak periods. The average vehicle speed for each time period is calculated, severe weather levels are defined, and their attenuation coefficients on vehicle speed are quantified. The travel time distribution patterns of different vehicle types are analyzed by classifying them as passenger cars / freight cars and ETC / CPC users. The importance of features is evaluated through the XGBoost algorithm, and a general prediction model based on XGBoost covers 80% of common scenarios.

[0065] S430: Adjust model parameters based on current traffic conditions and combine historical data. If the current rainfall is detected to reach the level of a rainstorm, automatically switch to the severe weather sub-model and increase the weather weight. When a sudden accident occurs on the road segment, real-time queue length data is introduced, the "accident impact coefficient" is temporarily added, the maximum travel time threshold of the road segment is appropriately increased, and the normal, Poisson, or Weiber distribution of travel time under different scenarios is fitted. The 95th percentile time value of the probability distribution is taken as the reference value for the longest travel time of the road segment.

[0066] like Figure 6 The flowchart shown is for vehicle trajectory reconstruction and is applicable when vehicle gantry information is incomplete. It includes the following steps:

[0067] S510: Dynamic gantry missing detection, based on the target vehicle's ETC gantry passage records, detects the number of missing gantries in real time. When the number of consecutive missing gantries is found to be < At that time, the system automatically triggers the trajectory spatiotemporal reconstruction mechanism, and constructs a time window constraint by combining the passage timestamps of adjacent gantry and the road design speed;

[0068] Missing gantry The calculation method is as follows: calculate the gantry density (number of gantry / km) based on the road segment length and the total number of gantry. Divide the density into three intervals: low density (<0.2 gantry / km), medium density (0.2-0.5 gantry / km), and high density (>0.5 gantry / km), and assign them weight coefficients of 0.7, 0.5, and 0.3 respectively.

[0069] The failure frequency of the gantry equipment on this section of the road over the past month was statistically analyzed. If the failure rate was less than 2%, a correction factor of 0.9 was applied; if it was between 2% and 5%, a correction factor of 1.0 was applied; and if it was greater than 5%, a correction factor of 1.1 was applied.

[0070] Considering the different requirements for gantry information in simple and complex road sections, in ordinary road sections, the gantry layout is relatively sparse. Even if a certain proportion of gantries are missing, such as 20%-30%, the system can still roughly deduce the trajectory based on the remaining gantry information and general vehicle movement patterns. In complex road sections, the dependence on gantry information is extremely high. Even a 5%-10% or less missing gantry can make trajectory determination difficult. Taking all factors into consideration, 10% is chosen as the baseline proportion for missing gantry information. This is calculated using the formula: "Proportion Threshold = Road Section Gantry Density Weight × Historical Equipment Failure Rate Correction Coefficient × 10%". The percentage threshold is multiplied by the total number of gantries in the road section, and the result is rounded up.

[0071] S520: Multimodal path reasoning modeling: Hidden Markov Model (HMM) is adopted, and the hidden state set is defined as the actual gantry nodes (G1, G2, ..., Gn) and their virtual topological connection states. A spatiotemporal transition probability matrix is ​​introduced, in which the state transition weights are dynamically calculated and generated by the gantry spacing and the average vehicle speed.

[0072] S530: Intelligent Route Reconstruction Engine: Innovatively embeds a dynamic interpolation mechanism for virtual nodes into the traditional Viterbi algorithm framework. It generates a set of candidate virtual gantries based on road network topology connectivity, and performs spatiotemporal verification by combining the passage time intervals between adjacent gantries; it constructs a Bayesian inference network, comprehensively considering vehicle historical route preferences, real-time traffic conditions, and time-series data for multi-dimensional probability calculation; it outputs the top N (N≥3) high-confidence trajectory hypotheses, each path accompanied by a spatiotemporal matching degree evaluation value (0-1 interval), generating a route reconstruction credibility analysis report including time-dimensional verification.

[0073] S540: When the system detects a missing gantry count ≥ Furthermore, when the spatiotemporal deviation rate between the derived path and the shortest toll-based path exceeds 15%, the toll evasion detection module is automatically activated. Through multi-feature fusion decision-making, including the path spatiotemporal anomaly index (reconstructed based on gantry passage time series), license plate feature cross-gantry similarity (SSIM≥0.85), and service area dwell time anomaly detection (compared with historical data of similar vehicle types), accurate identification of toll evasion behavior caused by OBU signal obstruction is achieved, forming a complete chain of evidence for investigation.

[0074] like Figure 7 The flowchart shown is applicable to matching export vehicle information and includes the following steps:

[0075] S610: When a vehicle enters the highway, it collects vehicle feature information through a multispectral high-definition camera (resolution ≥ 8 million pixels, frame rate 60fps) deployed at the entrance, uses an improved YOLOv8 model to detect the number of vehicle axles in real time, and performs cross-verification by combining wheel ground pressure distribution data.

[0076] S620: Encrypts data such as the number of axles, vehicle 3D contour, license plate information, and vehicle type classification and writes it into the OBU device and the provincial data center to form an unalterable feature vector group.

[0077] When a vehicle exits the highway, it obtains the entry binding data in real time through V2X communication and simultaneously retrieves the axle number change information recorded by the gantries along the route.

[0078] S630: Employs a multi-stage verification algorithm to automatically associate data such as service area surveillance video clips, ETC transaction records, and mobile signaling trajectories to construct a complete chain of evidence that includes spatiotemporal features, physical features, and behavioral features;

[0079] S640: If the weighing and axle count at the exit do not match the information at the entrance, the system will automatically initiate a second weighing and retrieve data from the three adjacent gantry frames for local path analysis.

[0080] This embodiment obtains the time of vehicle entry and exit from the highway, vehicle type characteristics, and license plate information, and combines this with vehicle gantry information obtained on the highway. Based on the obtained information, it makes a comprehensive judgment on whether the vehicle has any toll evasion behavior such as abnormal route. It accurately and efficiently identifies potential toll evasion behavior based on the vehicle's diverse information, thus protecting the toll collection rights of the highway.

[0081] Example 2

[0082] Embodiment 2 of the present invention introduces a highway vehicle toll evasion investigation system based on multi-source data fusion.

[0083] like Figure 8 The system shown is a highway toll evasion detection system based on multi-source data fusion, comprising:

[0084] The acquisition module is configured to acquire timestamps and vehicle characteristic information of vehicles entering and exiting highways;

[0085] The judgment module is configured to obtain the actual dynamic driving time of a vehicle on the highway based on the acquired timestamp. If the actual dynamic driving time exceeds the dynamic driving time threshold, the vehicle is suspected of potential toll evasion. Otherwise, the module determines whether the vehicle is suspected of potential toll evasion by acquiring the path gantry information of the vehicle during its highway journey. If the acquired path gantry information is missing, the vehicle is suspected of potential toll evasion. Otherwise, the module determines whether the vehicle is suspected of potential toll evasion based on the acquired vehicle feature information. If the acquired vehicle feature information for entering and exiting the highway does not match, the module determines that the vehicle is suspected of potential toll evasion, thus completing the highway vehicle toll evasion inspection based on multi-source data fusion.

[0086] The detailed steps are the same as those provided in Example 1 for the highway vehicle toll evasion investigation method based on multi-source data fusion, and will not be repeated here.

[0087] Example 3

[0088] Embodiment 3 of the present invention provides a computer-readable storage medium.

[0089] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the highway vehicle toll evasion investigation method based on multi-source data fusion as described in Embodiment 1 of the present invention.

[0090] The detailed steps are the same as the method for checking highway vehicle fee evasion based on multi-source data fusion provided in Embodiment One, and thus are not described again here.

[0091] Embodiment Four

[0092] Embodiment Four of the present application provides an electronic device.

[0093] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor, and the processor implements the steps in the method for checking highway vehicle fee evasion based on multi-source data fusion as described in Embodiment One of the present application when executing the program.

[0094] The detailed steps are the same as the method for checking highway vehicle fee evasion based on multi-source data fusion provided in Embodiment One, and thus are not described again here.

[0095] Embodiment Five

[0096] Embodiment Five of the present application provides a computer program product.

[0097] A computer program product includes software code, and the program in the software code implements the steps in the method for checking highway vehicle fee evasion based on multi-source data fusion as described in Embodiment One of the present application.

[0098] The detailed steps are the same as the method for checking highway vehicle fee evasion based on multi-source data fusion provided in Embodiment One, and thus are not described again here.

[0099] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0100] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks. Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks.

[0101] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks. Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks.

[0102] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks. Figure 1 one or more functions specified in the flowchart or multiple flows and / or blocks.

[0103] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments described and shown, and it is therefore intended that the appended claims cover all such variations and modifications that come within the scope of the application.

[0104] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

[0105] The above description is only preferred embodiments of the present application, and is not intended to limit the present application. The present application can be variously changed and modified by those skilled in the art without departing from the spirit and scope of the present application. Therefore, any modification, equivalent replacement, and improvement, etc. made within the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A method for investigating toll evasion on highways based on multi-source data fusion, characterized in that, include: Obtain the timestamps and vehicle characteristic information of vehicles entering and exiting highways; Based on the obtained timestamp, the actual dynamic driving time of the vehicle on the highway is obtained. When the actual dynamic driving time exceeds the dynamic driving time threshold, the vehicle has a potential toll evasion phenomenon. Otherwise, the potential toll evasion phenomenon of the vehicle is determined by obtaining the path gantry information of the vehicle during its journey on the highway. The process of obtaining the dynamic driving time threshold is as follows: acquiring historical traffic data and current traffic data of the target road segment; extracting key features affecting driving time based on the acquired historical traffic data; Based on the extracted key features, a target road segment traffic sub-model is constructed for each key feature; a matching target road segment traffic sub-model is selected based on the acquired current traffic data; and a dynamic driving time threshold is calculated by combining the selected target road segment traffic sub-models. In the process of determining whether a vehicle has a potential toll evasion by acquiring the route gantry information of the vehicle during its journey on the highway, the number of missing route gantry information is detected in real time by the route gantry. When the number of consecutively missing route gantry is less than a preset value, the trajectory spatiotemporal restoration mechanism is automatically triggered, and a time window constraint is constructed by combining the passage timestamps of adjacent route gantry with the design speed of the highway. A hidden Markov model is adopted, and the hidden state set is defined as the actual gantry node and its virtual topological connection state. A spatiotemporal transition probability matrix is ​​introduced, and the state transition weight is dynamically calculated and generated by the gantry spacing and the average vehicle speed. A candidate virtual gantry set is generated based on the road network topology connectivity, and spatiotemporal verification is performed by combining the passage time interval between adjacent gantries. A Bayesian inference network is constructed to perform multi-dimensional probability calculations by integrating vehicle historical route preferences, real-time traffic conditions, and time series data; the top N high-confidence trajectory hypotheses are output, each path is accompanied by a spatiotemporal matching degree evaluation value, and a path reconstruction credibility analysis report with time dimension verification is generated; when the number of missing path gantry is detected to be not less than a preset value and the spatiotemporal deviation rate between the derived path and the shortest toll path exceeds a set value, toll evasion investigation is automatically activated; If the obtained route gantry information is missing, the vehicle may be suspected of evading tolls; otherwise, the vehicle may be suspected of evading tolls based on the obtained vehicle characteristic information. When the vehicle characteristic information obtained does not match when entering and exiting the highway, it is determined that the vehicle has a potential toll evasion problem, and the highway toll evasion investigation based on multi-source data fusion is completed.

2. The method for investigating highway toll evasion based on multi-source data fusion as described in claim 1, characterized in that, The potential toll evasion issues of the vehicles include abnormal travel time, mismatched vehicle information, and missing route gantry information; abnormal travel time includes J-type toll evasion, traveling long distances to buy short-distance tickets, and no card at the exit; mismatched vehicle information includes trailer swapping and large vehicles with small labels; missing route gantry information includes obscured onboard electronic tag information and reversed highway composite toll cards.

3. The method for investigating highway toll evasion based on multi-source data fusion as described in claim 1, characterized in that, In the process of determining whether a vehicle has the potential toll evasion based on the acquired vehicle feature information, vehicle feature information is collected when the vehicle enters and exits the highway, and the number of vehicle axles is detected in real time. The number of vehicle axles obtained from the gantries along the route is retrieved simultaneously. A multi-stage verification method is used to match the vehicle feature information in order to determine whether the vehicle has the potential toll evasion.

4. The method for investigating highway toll evasion based on multi-source data fusion as described in claim 1, characterized in that, The actual dynamic driving time of the vehicle on the highway is the time difference between the timestamp when the vehicle leaves the highway and the timestamp when it enters the highway.

5. A highway toll evasion detection system based on multi-source data fusion, characterized in that, include: The acquisition module is configured to acquire timestamps and vehicle characteristic information of vehicles entering and exiting highways; The judgment module is configured to obtain the actual dynamic driving time of a vehicle on the highway based on the acquired timestamp. If the actual dynamic driving time exceeds the dynamic driving time threshold, the vehicle is suspected of potential toll evasion. Otherwise, the module determines whether the vehicle is suspected of potential toll evasion by acquiring the path gantry information of the vehicle during its highway journey. If the acquired path gantry information is missing, the vehicle is suspected of potential toll evasion. Otherwise, the module determines whether the vehicle is suspected of potential toll evasion based on the acquired vehicle feature information. If the acquired vehicle feature information for entering and exiting the highway does not match, the module determines that the vehicle is suspected of potential toll evasion, thus completing the highway vehicle toll evasion inspection based on multi-source data fusion. The process of obtaining the dynamic driving time threshold is as follows: acquiring historical traffic data and current traffic data of the target road segment; extracting key features affecting driving time based on the acquired historical traffic data; Based on the extracted key features, a target road segment traffic sub-model is constructed for each key feature; a matching target road segment traffic sub-model is selected based on the acquired current traffic data; and a dynamic driving time threshold is calculated by combining the selected target road segment traffic sub-models. In the process of determining whether a vehicle has a potential toll evasion by acquiring the route gantry information of the vehicle during its journey on the highway, the number of missing route gantry information is detected in real time by the route gantry. When the number of consecutively missing route gantry is less than a preset value, the trajectory spatiotemporal restoration mechanism is automatically triggered, and a time window constraint is constructed by combining the passage timestamps of adjacent route gantry with the design speed of the highway. A hidden Markov model is adopted, and the hidden state set is defined as the actual gantry node and its virtual topological connection state. A spatiotemporal transition probability matrix is ​​introduced, and the state transition weight is dynamically calculated and generated by the gantry spacing and the average vehicle speed. A candidate virtual gantry set is generated based on the road network topology connectivity, and spatiotemporal verification is performed by combining the passage time interval between adjacent gantries. A Bayesian inference network is constructed to perform multi-dimensional probability calculations by integrating vehicle historical route preferences, real-time traffic conditions, and time series data. The top N high-confidence trajectory hypotheses are output, with each path accompanied by a spatiotemporal matching degree evaluation value, generating a path reconstruction credibility analysis report that includes time dimension verification. When the number of missing path gantry is detected to be no less than a preset value and the spatiotemporal deviation rate between the derived path and the shortest toll path exceeds a set value, toll evasion investigation is automatically activated.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the highway vehicle toll evasion investigation method based on multi-source data fusion as described in any one of claims 1-4.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the highway vehicle toll evasion investigation method based on multi-source data fusion as described in any one of claims 1-4.

8. A computer program product, comprising software code, characterized in that, The program in the software code performs the steps of the highway vehicle toll evasion investigation method based on multi-source data fusion as described in any one of claims 1-4.

Images