Abnormal checkpoint detection method, computer device and storage medium

Abstract

The present application relates to an abnormal checkpoint detection method, computer equipment and storage medium. The method comprises: combining single vehicle passing data under multiple checkpoint pairs based on checkpoint collection data within a preset time; the single vehicle passing data comprises checkpoint collection data collected when a vehicle passes through a checkpoint pair once; calculating the vehicle driving data corresponding to the single vehicle passing data; determining the position abnormality and / or identifying the abnormal single vehicle passing data respectively based on the vehicle driving data corresponding to the multiple single vehicle passing data; determining the abnormal position checkpoint and / or identifying the abnormal checkpoint based on the position abnormality and identifying the abnormal single vehicle passing data. The use of this method can achieve the effect of improving the accuracy and efficiency of checkpoint equipment recognition.

Description

Technical Field

[0001] The present application relates to the technical field of equipment detection, and in particular to an abnormal checkpoint detection method, computer equipment and storage medium. Background Art

[0002] With the increasing popularity of the toll gate system, perception data has become an important part of building smart cities. However, due to problems at different levels of the toll gate equipment itself, the recognition of objects on the road may be biased. For example, there are actually multiple different objects under the recognition set of the same object, or the same object exists in the recognition sets of multiple objects. The wrong object may cause incorrect calculation results in subsequent analysis.

[0003] Traditional technology uses similarity-based license plate correction to overcome the problem of recognition results not corresponding to the actual object. However, there are too many similar license plates in the traffic network, especially in heavy traffic conditions. The similarity-based method places too high demands on the accuracy of the algorithm, resulting in large amounts of calculation and low accuracy.

[0004] It can be seen that the current bayonet system still has the problem of low accuracy and efficiency of bayonet device recognition. Summary of the invention

[0005] Based on this, it is necessary to provide an abnormal checkpoint detection method, computer equipment and storage medium that can improve the accuracy and efficiency of checkpoint equipment recognition in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for detecting an abnormal checkpoint, the method comprising:

[0007] Based on the checkpoint collection data within a preset time, the single vehicle passing data under multiple checkpoint pairs are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once;

[0008] Calculating vehicle travel data corresponding to the single vehicle passing data;

[0009] Based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or identifying the single vehicle passing data with abnormality respectively;

[0010] Based on the single vehicle passing data of the position abnormality and the identification abnormality, the position abnormality checkpoint is determined and / or the abnormal checkpoint is identified.

[0011] In one embodiment, the vehicle driving data includes a passing speed and a passing time, and the calculating of the vehicle driving data corresponding to the single passing data includes:

[0012] Calculate the passing time corresponding to each single vehicle passing data; the passing time is the difference in time when the vehicle is captured when passing through the checkpoint;

[0013] The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair, the vehicle passing time, and the farthest capture distance of each single checkpoint in the single vehicle passing data.

[0014] In one embodiment, after calculating the vehicle passing time corresponding to each single vehicle passing data, the method further includes:

[0015] Based on the passing times corresponding to the multiple single passing data, the passing times corresponding to the multiple single passing data are verified to obtain the verified passing time corresponding to each single passing data;

[0016] The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair, the verified vehicle passing time, and the farthest capture distance of each single checkpoint in the single vehicle passing data.

[0017] In one embodiment, based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or the single vehicle passing data with abnormal position includes:

[0018] If, among the plurality of single vehicle passing data, the number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in the plurality of single vehicle passing data meet a first preset condition, the speeding vehicle passing data is determined as single vehicle passing data with an identified abnormality; the speeding vehicle passing data is single vehicle passing data in which the passing speed exceeds a preset speed threshold;

[0019] If, among the multiple single vehicle passing data, the number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in the multiple single vehicle passing data meet the second preset condition, the speeding vehicle passing data is determined as single vehicle passing data with abnormal position.

[0020] In one embodiment, determining the position abnormal checkpoint and / or identifying the abnormal checkpoint based on the position abnormality and the single vehicle passing data with the abnormal identification includes:

[0021] Calculate the proportion of single vehicle passing data with abnormal identification to single vehicle passing data with abnormal location at each single checkpoint;

[0022] If the proportion is less than a preset ratio threshold, the single checkpoint is determined as a checkpoint with abnormal position; otherwise, the single checkpoint is determined as a checkpoint with abnormal identification.

[0023] In one embodiment, the abnormal checkpoint detection method further includes:

[0024] The confidence of the position abnormality checkpoint is calculated based on the proportion of the speeding vehicle passing data in the multiple single vehicle passing data and the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data.

[0025] In one embodiment, based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining and identifying abnormal single vehicle passing data includes:

[0026] Based on the vehicle passing times corresponding to the plurality of single vehicle passing data and an abnormality detection algorithm, the single vehicle passing data for identifying abnormalities is determined.

[0027] In one embodiment, the determining and identifying abnormal single vehicle passing data based on the vehicle passing time and the abnormality detection algorithm corresponding to the plurality of single vehicle passing data further includes:

[0028] Eliminate single vehicle passing data whose passing time is greater than the preset time threshold;

[0029] Based on the vehicle passing times corresponding to the eliminated multiple single vehicle passing data and the abnormality detection algorithm, the abnormal single vehicle passing data is determined and identified.

[0030] In a second aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0031] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method described above when executed by a processor.

[0032] The above-mentioned abnormal checkpoint detection method, computer equipment and storage medium obtain single vehicle passing data under multiple checkpoint pairs by combining the checkpoint collection data within a preset time; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes the checkpoint pair once; calculate the vehicle driving data corresponding to the single vehicle passing data; based on the vehicle driving data corresponding to the multiple single vehicle passing data, determine the single vehicle passing data with abnormal position and / or abnormal identification respectively; determine the abnormal checkpoint and / or identify the abnormal checkpoint based on the single vehicle passing data with abnormal position and abnormal identification, so as to realize the detection and root cause judgment of the position abnormality and identification abnormality of the checkpoint based on the vehicle passing trajectory, so as to further sort out the possible checkpoint defects, adjust the checkpoint system parameters or notify the equipment operation and maintenance personnel to perform maintenance, which can achieve the effect of improving the accuracy and efficiency of checkpoint equipment recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A diagram showing an application environment of an abnormal checkpoint detection method in an embodiment;

[0034] Figure 2 Schematic diagram of a flow chart of an abnormal checkpoint detection method in one embodiment;

[0035] Figure 3 A schematic diagram of a flow chart of an abnormal checkpoint detection method in another embodiment;

[0036] Figure 4 is a schematic diagram of a user interaction interface in one embodiment;

[0037] Figure 5 is a schematic diagram of a user interaction interface in another embodiment;

[0038] Figure 6 is a schematic diagram of a user interaction interface in another embodiment;

[0039] Figure 7 is a structural block diagram of an abnormal checkpoint detection device in one embodiment;

[0040] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] The abnormal checkpoint detection method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains the checkpoint data within the preset time from the server 104, and based on the checkpoint collection data within the preset time, the single vehicle passing data under multiple checkpoints are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes the checkpoint pair once; calculate the vehicle driving data corresponding to the single vehicle passing data; based on the vehicle driving data corresponding to the multiple single vehicle passing data, determine the position abnormality and / or identify the abnormal single vehicle passing data; based on the position abnormality and the identification of the abnormal single vehicle passing data, determine the position abnormal checkpoint and / or identify the abnormal checkpoint. Among them, the terminal 102 can be but not limited to various personal computers, laptops, smart phones and tablets. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0043] In one embodiment, Figure 2 As shown, a method for detecting abnormal checkpoints is provided, and the method is applied to Figure 1 The terminal 102 in the example is used as an example to illustrate, and the following steps are included:

[0044] Step S100, based on the checkpoint collection data within a preset time, single vehicle passing data under multiple checkpoint pairs are combined.

[0045] The checkpoint collected data within the preset time may be the checkpoint collected data obtained by all or part of the checkpoints in the road network within the preset time. The checkpoint collected data, i.e., the collected data obtained by each checkpoint at the corresponding position, may include the perception and recording of the license plate, and may also include one or more of the information such as vehicle brand, vehicle type, vehicle color, and capture time, and may also include other vehicle-related information, which is not limited in this embodiment.

[0046] A checkpoint pair may include single checkpoints at both ends of a specific road section. The checkpoint devices of two single checkpoints may be arranged adjacently or oppositely to capture one-way, two-way or multi-way traffic flows. Single vehicle passing data includes the checkpoint collection data collected when a vehicle passes through a checkpoint pair once, which is used to record relevant information of the vehicle.

[0047] Based on the collected data within a preset time period, the corresponding single checkpoints in the checkpoint collection data can be combined according to different road sections to obtain single vehicle passing data. According to the driving trajectory of the vehicle, the checkpoints through which the vehicle passes can be combined. For example, a specific road section may include one or more intersections, or may pass through one or more checkpoint devices. When the vehicle passes through the first checkpoint, the second checkpoint, and the third checkpoint in sequence and is captured, the first checkpoint and the second checkpoint, the second checkpoint and the third checkpoint, and the first checkpoint and the third checkpoint can be respectively combined into checkpoint pairs to obtain corresponding single vehicle passing data.

[0048] Step S200, calculating the vehicle driving data corresponding to the single vehicle passing data.

[0049] The vehicle driving data may be driving status data of a vehicle passing through in a single vehicle passing data, wherein the driving status data may be, for example, the passing speed, passing time, passing route, etc. of the vehicle.

[0050] Exemplarily, the calculation of the passing time can be obtained based on the time difference between passing two single checkpoints in a single passing data, the passing speed can be obtained based on the distance between the two single checkpoints and the passing time, and the passing route can be determined based on the order in which the two single checkpoints record the same vehicle to determine the passing direction and route.

[0051] Step S300, based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or identifying the abnormal single vehicle passing data respectively.

[0052] Among them, single vehicle passing data with abnormal position may be single vehicle passing data with errors due to abnormal position information. For example, when the position information of the bayonet device is abnormal, since the position information of the bayonet device may be manually set during installation, data deviation may occur in subsequent analysis due to inaccurate setting; when the vehicle's position information is abnormal, for example, when the vehicle's position information is the position of the bayonet device based on the vehicle's position in its field of view, and the position in the actual space is mapped and inferred, there may be an inaccurate mapping relationship, an inaccurate mapping algorithm, or an error in the calculation of the vehicle's position information caused by a change in the angle of the bayonet device.

[0053] Determining a single vehicle passing data with an abnormal position based on the vehicle driving data corresponding to multiple single vehicle passing data can be done by combining different single vehicle passing data of the same vehicle within a preset time to identify whether the vehicle speed change is abnormal, or by combining whether the average value of the passing speeds of multiple vehicles when passing the checkpoint is greater than a threshold to determine whether the corresponding single vehicle passing data is abnormal. It can also be determined whether the corresponding single vehicle passing data is abnormal based on the number or proportion of multiple vehicles whose passing speeds are greater than a threshold when passing the checkpoint. It can also be based on a combined judgment of two or more of the above methods, which is not limited in this embodiment.

[0054] The single vehicle passing data with abnormal recognition may be the single vehicle passing data with errors due to abnormal license plate recognition, for example, due to inaccurate recognition algorithm, interference from environmental factors such as weather, light, smoke, etc., the license plate itself is dirty or blocked, etc., resulting in abnormal license plate recognition. It is understandable that in the case of abnormal license plate recognition, it usually leads to the single vehicle passing data obtained by recognition, and the time of passing the checkpoint is obviously too long or too short. For example, the first vehicle and the second vehicle that pass through two single checkpoints in a short period of time may be recognized as the same vehicle, resulting in the situation where the time of passing the vehicle in the single vehicle passing data is too short.

[0055] Determine the single vehicle passing data with abnormal identification based on the vehicle driving data corresponding to the multiple single vehicle passing data. It can be based on the passing speeds or passing times of multiple vehicles when passing the checkpoints, and regard the single vehicle passing data with individual abnormal passing speeds or passing times as the single vehicle passing data with abnormal identification. It can also be based on the collected data of two single checkpoints, the vehicle type, vehicle color, vehicle brand, etc. of the same license plate, to determine whether there is any misidentification, and regard it as the single vehicle passing data with abnormal identification.

[0056] Step S400, based on the single vehicle passing data of the position abnormality and the identification abnormality, determine the position abnormality checkpoint and / or identify the abnormal checkpoint.

[0057] Among them, based on the single vehicle passing data with abnormal position and abnormal identification, determining the abnormal position checkpoint and / or identifying the abnormal checkpoint can be performed by analyzing the possibility of position abnormality or identification abnormality for each single checkpoint based on the single vehicle passing data with abnormal position and the single vehicle passing data with abnormal identification. For example, when the proportion of single vehicle passing data with abnormal position in the single vehicle passing data corresponding to a single checkpoint exceeds a preset threshold, it means that the positioning information of the single checkpoint may be abnormal; corresponding adjustments need to be made, and when the proportion of single vehicle passing data with abnormal identification in the single vehicle passing data corresponding to a single checkpoint exceeds a preset threshold, it can be characterized that the checkpoint device of the single checkpoint itself has problems such as inaccurate recognition algorithm or long-term interference from environmental factors, and corresponding software or hardware adjustments need to be made.

[0058] The present embodiment provides an abnormal checkpoint detection method, which obtains single vehicle passing data under multiple checkpoint pairs by combining the checkpoint collection data within a preset time; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes the checkpoint pair once; the vehicle driving data corresponding to the single vehicle passing data is calculated; based on the vehicle driving data corresponding to the multiple single vehicle passing data, the single vehicle passing data with position abnormality and / or identification abnormality are determined respectively; based on the single vehicle passing data with position abnormality and identification abnormality, the checkpoint with abnormal position and / or identification of abnormal checkpoint is determined, so that the position abnormality and identification abnormality of the checkpoint are detected and the root cause judgment is made based on the vehicle passing trajectory, so that possible checkpoint defects can be further sorted out, the checkpoint system parameters can be adjusted, or the equipment operation and maintenance personnel can be notified to perform maintenance, which can achieve the effect of improving the accuracy and efficiency of checkpoint equipment recognition.

[0059] In one embodiment, the combining of the data collected at the checkpoints within a preset time to obtain the single vehicle passing data under multiple checkpoints includes:

[0060] Determining multiple license plates based on the data collected by the checkpoint within the preset time;

[0061] Based on the data collected by the checkpoints, the trajectory of each license plate is determined, and multiple groups of checkpoint pairs are obtained according to the relative positions of multiple single checkpoints in the road network;

[0062] Based on the checkpoint collection data of a single checkpoint in each checkpoint pair, multiple single vehicle passing data are combined.

[0063] Wherein, based on the checkpoint collected data within the preset time, multiple license plates are determined, and the license plates identified in the checkpoint collected data of multiple single checkpoints are summarized to obtain the multiple license plates.

[0064] By grouping the data collected by the checkpoints according to the license plate, and sorting the data according to the time when the license plate passes through different single checkpoints, the trajectory of each license plate can be determined. According to the relative positions of multiple single checkpoints in the road network, multiple single checkpoints can be combined in pairs according to the track of the license plate to obtain multiple groups of checkpoint pairs. In a specific embodiment, multiple groups of checkpoint pairs can be obtained by combining adjacent single checkpoints in the track of the license plate in pairs.

[0065] Based on the checkpoint collection data of a single checkpoint in each checkpoint pair, multiple single vehicle passing data are combined. The checkpoint collection data obtained by collecting data from two single checkpoints on the same license plate in a single checkpoint pair can be combined to obtain single vehicle passing data of a single checkpoint pair for a single license plate. Multiple single vehicle passing data can be obtained by combining multiple checkpoint pairs separately.

[0066] The present embodiment provides an abnormal checkpoint detection method, which collects data based on the checkpoint, determines the trajectory of each license plate, and obtains multiple groups of checkpoint pairs according to the relative positions of multiple single checkpoints in the road network, and then combines to obtain multiple single vehicle passing data. This can achieve effective data extraction of single vehicle passing data, thereby providing a basis for subsequent checkpoint abnormality identification based on vehicle driving data, which can not only improve data organization and availability, but also further improve the accuracy and efficiency of checkpoint equipment identification.

[0067] In one embodiment, the vehicle driving data includes a passing speed and a passing time, and the calculating of the vehicle driving data corresponding to the single passing data includes:

[0068] Calculate the passing time corresponding to each single passing data; the passing time is the difference in time when the vehicle is captured when passing through the checkpoint;

[0069] The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair, the vehicle passing time, and the farthest capture distance of each single checkpoint in the single vehicle passing data.

[0070] The calculation of the passing time corresponding to each single vehicle passing data may be based on the time when two single checkpoints respectively capture and record the vehicle in each single vehicle passing data, and the capture time difference may be calculated based on the two capture times.

[0071] The shortest distance of the road network can be obtained by calculating the straight-line distance between the two checkpoint devices based on the latitude and longitude data reported by the two checkpoint devices in the checkpoint alignment or other coordinate data used to represent the position. If the checkpoint device has a ranging function and there is no obstruction between the two checkpoint devices, the shortest distance of the road network between the checkpoint devices can also be obtained by ranging.

[0072] The farthest capture distance of each single checkpoint may be the farthest distance that the checkpoint device can identify and capture. The farthest capture distances of different single checkpoints may vary due to the checkpoint device model, terrain, etc. In the calculation of the passing speed, it can be calculated based on the actual farthest capture distance, or it can be manually set based on the average value of the checkpoint device.

[0073] In a specific embodiment, based on the shortest distance of the road network, the vehicle passing time, and the farthest capture distance of each single checkpoint, the vehicle passing speed can be calculated by the following formula:

[0074] v=(d-α 1 -α 2 ) /

[0075] Among them, v is the passing speed, d is the shortest distance of the road network, α 1 and α 2 They are the farthest capture distances of the two bayonet devices. is the time of passing the vehicle, i.e. the time difference of capturing. It can be understood that when the farthest capturing distance is set manually, the above formula can also be simplified to v=(d-2×α) / , where α is the maximum capture distance setting value of the camera device.

[0076] The abnormal checkpoint detection method provided in this embodiment calculates the passing speed by combining the shortest distance of the road network and the actual capture time difference, which can provide more accurate vehicle speed information, help to effectively detect potential abnormal behaviors, such as speeding, data misreading, etc., and can improve the accuracy and efficiency of checkpoint equipment recognition, and further improve the overall performance of the system.

[0077] In one embodiment, after calculating the vehicle passing time corresponding to each single vehicle passing data, the method further includes:

[0078] Based on the passing times corresponding to the multiple single passing data, the passing times corresponding to the multiple single passing data are verified to obtain the verified passing time corresponding to each single passing data;

[0079] The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair, the verified vehicle passing time, and the farthest capture distance of each single checkpoint in the single vehicle passing data.

[0080] It is understandable that judging the abnormality of a single vehicle passing data based on the passing speed can be used to judge whether there is an abnormality in the position information of the checkpoint device itself. In order to determine the abnormality of the position information, the passing time of each single vehicle passing data can be verified by eliminating unfavorable factors, so as to avoid the influence of the calculation result of the passing speed due to the wrong passing time.

[0081] Among them, the passing times corresponding to multiple single passing data are verified to obtain the verified passing time corresponding to each single passing data. This can be by obtaining single passing records of multiple license plates passing under the same checkpoint, and verifying the passing times that obviously deviate from the normal value based on the passing times of multiple license plates. It can also be by determining the median value of the passing times of multiple license plates as the verified passing time of multiple license plates.

[0082] The present embodiment provides an abnormal checkpoint detection method, which verifies the vehicle passing time and calculates the vehicle passing speed based on the verified vehicle passing time. In this way, when there are both position abnormalities and recognition abnormalities, the influence of recognition abnormalities on the vehicle passing time and speed can be eliminated, thereby more accurately obtaining single vehicle passing data with position abnormalities, thereby achieving the effect of improving the recognition accuracy and efficiency of the checkpoint equipment.

[0083] In one embodiment, based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or the single vehicle passing data with abnormal position includes:

[0084] If, among the plurality of single vehicle passing data, the number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in the plurality of single vehicle passing data meet a first preset condition, the speeding vehicle passing data is determined as the single vehicle passing data for identifying abnormality;

[0085] If, among the multiple single vehicle passing data, the number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in the multiple single vehicle passing data meet the second preset condition, the speeding vehicle passing data is determined as single vehicle passing data with abnormal position.

[0086] Among them, the speeding vehicle passing data is a single vehicle passing data in which the passing speed exceeds the preset speed threshold. It can be understood that when the single vehicle passing data is identified as speeding vehicle passing data, the number of abnormalities can be accumulated.

[0087] The number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in multiple single vehicle passing data meet the first preset condition. The first preset condition can be that the number is lower than the first threshold and / or the proportion of its number in the total amount of single vehicle passing data is less than the first proportion threshold. It can be characterized that these speeding vehicle passing data may be speed deviations caused by misidentification, and therefore they can be classified as abnormally identified vehicle passing data.

[0088] The number of speeding vehicle passing data and / or the proportion of speeding vehicle passing data in multiple single vehicle passing data meet the second preset condition. The second preset condition can be that the number is higher than the second threshold and / or the proportion of its number in the total single vehicle passing data is greater than the second proportion threshold. This can indicate that there is a deviation in the passing speed during the multiple data collection processes of the checkpoint device. It can be used as a basis for judging the abnormality of the checkpoint position and be identified as single vehicle passing data with abnormal position.

[0089] The present embodiment provides an abnormal checkpoint detection method, which can effectively identify which data are abnormal due to recognition errors and which data are abnormal due to location or equipment problems based on the number and / or proportion of speeding vehicle data in a single vehicle passing data, so as to maintain the cause of the abnormality and improve the accuracy and efficiency of checkpoint equipment recognition.

[0090] In one embodiment, determining the position abnormal checkpoint and / or identifying the abnormal checkpoint based on the position abnormality and the single vehicle passing data with the abnormal identification includes:

[0091] Calculate the proportion of single vehicle passing data with abnormal identification to single vehicle passing data with abnormal location at each single checkpoint;

[0092] If the proportion is less than a preset ratio threshold, the single checkpoint is determined as a checkpoint with abnormal position; otherwise, the single checkpoint is determined as a checkpoint with abnormal identification.

[0093] Among them, it is understandable that a single vehicle passing data with abnormal position may also be identified as a single vehicle passing data with abnormal identification due to the simultaneous existence of abnormal vehicle passing time. When a single vehicle passing data is identified as having abnormal identification, the judgment basis is often determined based on the abnormality of the vehicle passing time. Therefore, the position abnormality determined by the vehicle passing speed may also be caused by the abnormality of the vehicle passing time. Therefore, when a single vehicle passing data has both position abnormality and identification abnormality, it may be the vehicle passing data abnormality caused by the identification abnormality, which in turn causes it to be judged as position abnormality. Therefore, this type of situation needs to be excluded.

[0094] Furthermore, the proportion of single vehicle passing data with identification abnormalities in the single vehicle passing data with position abnormalities in each single checkpoint can be calculated. If the proportion is less than a preset ratio threshold, the position abnormality caused by identification abnormalities can be ruled out, and the single checkpoint can be determined as a checkpoint with position abnormalities. Otherwise, it can be determined as a checkpoint with identification abnormalities.

[0095] The present embodiment provides a method for detecting abnormal checkpoints, which determines the checkpoints with abnormal positions and identifies abnormal checkpoints by identifying the proportion of abnormal single vehicle passing data in single vehicle passing data with abnormal positions. This method can effectively determine which checkpoints are abnormal due to position problems and which are abnormal due to identification problems, and perform targeted equipment maintenance, thereby achieving the effect of improving the accuracy and efficiency of checkpoint equipment identification.

[0096] In one embodiment, the abnormal checkpoint detection method further includes:

[0097] The confidence of the position abnormality checkpoint is calculated based on the proportion of the speeding vehicle passing data in the multiple single vehicle passing data and the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data.

[0098] It can be understood that the higher the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data, the lower the possibility of the checkpoint position abnormality and the higher the possibility of identifying the abnormality, and the lower the proportion of the speeding vehicle passing data in the multiple single vehicle passing data, the lower the possibility of the checkpoint itself being abnormal, and vice versa. Therefore, based on the proportion of the speeding vehicle passing data in the multiple single vehicle passing data and the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data, the confidence of the checkpoint in the abnormal position can be calculated, and the proportions of the two can be converted accordingly through a preset ratio to obtain the corresponding confidence result.

[0099] The present embodiment provides an abnormal checkpoint detection method, which calculates the confidence level of the abnormal checkpoint based on the proportion of the speeding vehicle passing data in the multiple single vehicle passing data, and the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data. This can provide users with a reference to the abnormal situation and degree of abnormality, thereby providing more accurate data support and decision-making basis.

[0100] In one embodiment, based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining and identifying abnormal single vehicle passing data includes:

[0101] Based on the vehicle passing times corresponding to the plurality of single vehicle passing data and an abnormality detection algorithm, the single vehicle passing data for identifying abnormalities is determined.

[0102] Among them, based on the passing time corresponding to multiple single vehicle passing data and the abnormality detection algorithm, the abnormal single vehicle passing data can be determined and identified. The single vehicle passing data of multiple license plates under the same checkpoint can be subjected to abnormality detection of the passing time, so as to determine the abnormal single vehicle passing data.

[0103] Exemplarily, the anomaly detection algorithm may be a detection method based on statistics, such as a box plot, standard deviation, etc., or a method based on machine learning, such as an isolation forest, an autoencoder, etc.

[0104] An abnormal checkpoint detection method provided in this embodiment can effectively identify data deviations caused by recognition errors by combining the vehicle passing time and the abnormal detection algorithm, thereby providing more reliable data support for the equipment operation and maintenance department, and can achieve the effect of improving the accuracy and efficiency of checkpoint equipment recognition.

[0105] In one embodiment, the determining and identifying abnormal single vehicle passing data based on the vehicle passing time and the abnormality detection algorithm corresponding to the plurality of single vehicle passing data further includes:

[0106] Eliminate single vehicle passing data whose passing time is greater than the preset time threshold;

[0107] Based on the vehicle passing times corresponding to the eliminated multiple single vehicle passing data and the abnormality detection algorithm, the abnormal single vehicle passing data is determined and identified.

[0108] It is understandable that when a vehicle passes through a checkpoint pair, it may stop between two single checkpoints, such as temporary parking, long-term parking, traffic accidents, etc. The identification of abnormal single-vehicle passing data is often an abnormal identification of single-vehicle passing data of normal driving. Therefore, it is necessary to exclude vehicles that may have stopped.

[0109] Furthermore, when determining the single process data for identifying abnormalities, the single vehicle passing data with a passing time greater than a preset time threshold may be eliminated. In a specific embodiment, it may be determined whether the passing time satisfies the following formula:

[0110] > ×T 路网

[0111] in, For the passing time, is the preset coefficient, T 路网 To calculate the average passing time of the road network, if the passing time satisfies the formula, it can be considered that the vehicle has stopped or reached the end of its trajectory between passing two checkpoints, and the passing time needs to be eliminated. Then, based on the passing time corresponding to the eliminated multiple single passing data and the anomaly detection algorithm, the abnormal single passing data is determined and identified.

[0112] The present embodiment provides an abnormal checkpoint detection method, which can more effectively identify data deviations caused by recognition errors by eliminating the passing times in which vehicles stop, and then performing abnormality detection on the passing times in multiple single passing data. This can provide more reliable data support for the equipment operation and maintenance department, and can achieve the effect of improving the accuracy and efficiency of checkpoint equipment recognition.

[0113] In one embodiment, the determining of the position abnormal checkpoint and identifying the abnormal checkpoint based on the position abnormality and the single vehicle passing data of the abnormal identification includes:

[0114] The abnormal position checkpoint, the abnormal identification checkpoint, and the checkpoint collection data corresponding to the abnormal position checkpoint and the abnormal identification checkpoint are displayed through the user interaction interface.

[0115] The user interaction interface may be in the form of a map view, showing the locations of all or part of the checkpoints, or may be in the form of emphatic expressions such as color, icon size, etc., to indicate location anomalies and identify abnormal checkpoints.

[0116] When there is a confidence calculation for the position abnormal checkpoint and the identification abnormal checkpoint, the corresponding part of the checkpoint can also be screened out for display according to the minimum confidence selected by the user. In a specific embodiment, the user can select the checkpoint abnormality confidence on the page, and the suspected abnormal checkpoint will be displayed on the page at the same time. Click the abnormal checkpoint to display multiple snapshots of the abnormal identification of the vehicle passing data under the checkpoint as a chain of evidence, and at the same time display the estimated cause of the checkpoint abnormality and the corresponding snapshot time of each abnormal identification of the vehicle passing data.

[0117] The present embodiment provides an abnormal checkpoint detection method, which displays the abnormal position checkpoint and the identified abnormal checkpoint, as well as the checkpoint collection data corresponding to the abnormal position checkpoint and the identified abnormal checkpoint, through a user interaction interface. This can provide comprehensive information support for management or equipment operation and maintenance personnel, assist in making more accurate and rapid maintenance and decisions, and achieve the effect of improving the accuracy and efficiency of checkpoint equipment identification.

[0118] In one embodiment, the displaying of the abnormal position checkpoint and the abnormal identification checkpoint, and the checkpoint collection data corresponding to the abnormal position checkpoint and the abnormal identification checkpoint also includes:

[0119] Based on the abnormal license plate detection algorithm and the abnormal license plate correction algorithm, the abnormal identification license plate in the checkpoint collection data corresponding to the abnormal checkpoint is obtained and the correction result of the abnormal identification license plate is obtained;

[0120] Determining the abnormal cause of the abnormal identification checkpoint based on the plurality of abnormal identification license plates and the correction results of the abnormal identification license plates;

[0121] The abnormal identification checkpoints caused by the abnormality of the license plate itself will be removed.

[0122] It is understandable that the single-pass data for identifying abnormalities can be obtained by side judgment based on the time of passing the vehicle, so the correctness of the license plate recognition in the single-pass data has not been verified. Based on the abnormal license plate detection algorithm, the abnormal identification license plate in the bayonet acquisition data corresponding to the abnormal bayonet is obtained, which can be through the abnormal license plate detection algorithm, the license plate number identified in the single-pass data and the actual captured image are error detected, so that the wrongly identified license plate, that is, the abnormal identification license plate, can be determined; based on the abnormal license plate correction algorithm, the correction result of the abnormal identification license plate is obtained, which can be corrected for the abnormal identification license plate, and the license plate actually corresponding to the captured image of the abnormal identification license plate is determined as the correction result.

[0123] By comparing the abnormal recognition license plate with the correction result, the abnormal reason for the abnormal recognition of the checkpoint can be determined. For example, when the license plate is blocked or damaged, the abnormal reason is the abnormality of the license plate itself, not the recognition abnormality of the checkpoint. When the contrast or brightness of the captured image of the license plate is low, resulting in abnormal recognition, the abnormal reason may be insufficient lighting. Accordingly, maintenance and adjustment can be made by setting a fill light.

[0124] Removing recognition abnormal checkpoints whose abnormality is caused by abnormality of the license plate itself can eliminate recognition abnormalities that are not caused by abnormality of the checkpoint itself, and more resources can be invested in systems or equipment that really need to be optimized.

[0125] The present embodiment provides a method for detecting abnormal checkpoints, which can effectively identify and correct abnormal license plate data by removing abnormal checkpoints caused by abnormalities in the license plate itself, and can also intuitively display these abnormal checkpoints and their related collected data through a user interaction interface, and by removing abnormal checkpoints caused by abnormalities in the license plate itself, optimize the system's performance and user experience.

[0126] In order to more clearly illustrate the technical solution of the present application, the present application also provides a detailed embodiment.

[0127] In one embodiment, Figure 3 As shown, a method for detecting abnormal checkpoints is provided, comprising:

[0128] Step S01, obtaining recent vehicle capture data information and related road network information, the vehicle capture data information includes the checkpoint latitude and longitude, capture time, captured license plate and other information.

[0129] Step S02: analyzing abnormal vehicle speed based on the captured vehicle data and the road network data.

[0130] Step S03, performing abnormal analysis on the bayonet.

[0131] Step S04, analyzing the abnormal captured data.

[0132] Step S05, providing a user interface interactive capability to display checkpoint anomalies and license plate obstruction and dirt.

[0133] Wherein, step S02 includes:

[0134] Step S02.1, group the recent vehicle capture data information according to the license plate and sort them by time, calculate the vehicle speed (i.e., passing speed) based on the shortest road network distance between two consecutive capture checkpoints and the capture time difference (i.e., passing time), and if there is no road network data, calculate the straight-line distance based on the longitude and latitude data. The vehicle speed calculation formula is as follows:

[0135] v = (d - 2 × α) /

[0136] Among them, d is the shortest distance of the road network, α is the farthest capture distance of the camera device, is the capture time difference, that is, the time when the vehicle passes. If the speed v is greater than the threshold, it can be considered that the capture data of the two front and rear checkpoints are abnormal.

[0137] Step S02.2: According to the abnormal snapshot data calculated in step S02.1, statistical analysis is performed on the abnormal checkpoint pairs and single abnormal checkpoints. 1 and A 2 When the single passing speed is greater than the threshold, the checkpoint ab(A 1 , A 2 ) accumulates the number of abnormalities, where ab(A 1 , A 2 ) and ab(A 2 , A 1 ) is consistent, and the single bayonet abnormality ab(A 1 ) and ab(A 2 ) to accumulate the number of exceptions.

[0138] Analyze the cumulative number of abnormalities of the checkpoint pair and the single checkpoint. If the cumulative number is less than the first threshold, the corresponding captured data is high-confidence identification of abnormal vehicle passing data. If the cumulative number is greater than the second threshold, the corresponding captured data is high-confidence checkpoint abnormal data.

[0139] Step S02.3, group the recent vehicle capture data information according to the license plate, and sort them by time, perform statistical analysis based on the capture time difference between two consecutive capture checkpoints, and analyze the capture time difference sequence between all checkpoints. For example, all vehicles that pass through two checkpoints A consecutively 1 and A 2 A total of n times, then the time difference sequence includes T 1 ,…,T n .

[0140] According to the shortest time of the road network, the time difference sequence is filtered. If the capture time difference meets > ×T 路网 , it can be considered that the vehicle stopped or its trajectory was interrupted between the two capture checkpoints, and the time difference needs to be eliminated.

[0141] For the deleted snapshot time difference sequence, abnormal time detection analysis is performed to mine abnormal time differences that are too small. If the time difference is an abnormal value, it can be considered that the snapshot data corresponding to the abnormal time difference of the two front and rear checkpoints are abnormal.

[0142] Step S02.4, verifying the actual speed of the checkpoints by the distribution of the time it takes for multiple vehicles to pass through the two checkpoints. Specifically, the abnormal passing time can be adjusted based on the passing time of multiple vehicles when passing through the checkpoints, so as to eliminate the influence of the abnormal passing time on the calculation of the passing speed.

[0143] If the single vehicle passing data meets the abnormal conditions of step S02.1 (i.e. the passing speed is greater than the threshold) and the single abnormal conditions of S02.3 (i.e. there is an abnormality in the capture time difference), or the single vehicle passing data is the high-confidence abnormal vehicle passing data in S02.1, then the single vehicle passing data is considered to be the single vehicle passing data with identification abnormality, and then the full amount of vehicle passing data of the vehicle is analyzed according to the vehicle type, vehicle brand, and body color, the vehicle passing data with misidentified license plates is mined, and the confidence is output.

[0144] Count the proportion of single vehicle passing data that meets the high confidence checkpoint abnormal data defined in step S02.1, but does not meet the abnormal situation in step S02.3. 1 and A 2 The cumulative number of abnormalities ab(A 1 , A 2 )=e 1 , that is, there exists e 1 There is speeding in the single vehicle passing data, and e 1 There is e 2If the single vehicle passing data does not meet the abnormal situation of the capture time difference in step S02.3, the real abnormality ratio of the checkpoint pair is real_ab(A 1 , A 2 )=(e 1 -e 2 ) / e 1 If the proportion is less than the threshold, the single checkpoint under the checkpoint pair can be considered as a high-confidence position abnormal checkpoint, and the confidence of the position abnormal checkpoint is output.

[0145] Step S03 includes:

[0146] Step S03.1: If the checkpoint is a high-confidence position abnormal checkpoint, it can be considered that there is an abnormality in the longitude and latitude settings of the checkpoint.

[0147] Step S03.2, based on the analysis of the passing vehicle data and the license plate misidentification vehicle data, classify the abnormal reasons for identifying abnormal checkpoints. For example, if the amount of data captured by the checkpoint is small, the checkpoint is built in a bad position or angle; if the license plate resolution rate of the checkpoint is low, the clarity of the checkpoint is poor; if the license plate recognition accuracy of the checkpoint is low throughout the day, the hardware algorithm of the checkpoint needs to be upgraded; if the license plate recognition accuracy of the checkpoint is low at night, the equipment fill light of the checkpoint is insufficient, etc.

[0148] Step S04 includes:

[0149] Step S04.1, the single vehicle passing data that was misidentified mined in step S02.4 is combined with an abnormal license plate analysis method to detect abnormally identified license plates. The abnormal license plate analysis method includes but is not limited to analyzing whether the license plate number meets the issuance rules of the vehicle management office, whether the license plate number and license plate type match, whether the vehicle passing data has temporal and spatial anomalies, whether the front-end and back-end parsed data are consistent, the number of recent license plate appearances is abnormal, the person-vehicle relationship is abnormal, the person-vehicle relationship is inconsistent with the registration of the vehicle management office, the license plate is not registered, the vehicle attributes of the vehicle passing data are inconsistent with the registration data, and other misidentified license plate detection algorithms.

[0150] The license plate type correction algorithm, special license plate number and license plate type correction algorithm, person-vehicle relationship data and license plate correction algorithm, vehicle file data and license plate correction algorithm, nearby time and space correction algorithm, track must-pass point and license plate correction algorithm are used to correct the license plate of the vehicle that has been misidentified. Obtain the abnormally recognized license plate and the corresponding corrected license plate.

[0151] Step S04.2, analyze the abnormal recognition license plate data and analyze the license plate recognition errors. For example, if several digits of the license plate number of vehicle M are misidentified for a long time, and the misidentification situation is not unique, then the vehicle may have a license plate obstruction or dirt; if vehicle N is not a trailer, and the proportion of the front (rear) capture in the passing vehicle capture is much higher than the proportion of the rear (front) capture, then it is considered that the rear (front) of the vehicle is suspected to have a license plate obstruction or dirt.

[0152] Step S05 includes:

[0153] Step S05.1, such as Figure 4 and Figure 5 As shown, a user interaction interface is provided to display multiple abnormal checkpoints and trajectory analysis to the user, and a "search license plate number" function is provided for trajectory analysis. The user can select the confidence level of the checkpoint abnormality on the page. For example, when the confidence threshold is 80%, suspected abnormal checkpoints with a confidence level greater than 80% are displayed on the page at the same time. When the abnormal checkpoint is clicked, multiple snapshots of the abnormally identified vehicle passing data under the checkpoint are displayed as a chain of evidence. The inferred cause of the checkpoint abnormality and the corresponding snapshot time of each abnormally identified vehicle passing data can also be displayed. For example, "Insufficient equipment fill light" and "2022-10-17 16:30" are displayed. For different reasons for the checkpoint abnormality, according to the user's input instructions, the user is supported to manually modify the reason for the checkpoint abnormality and delete the checkpoint in the abnormal checkpoint list.

[0154] Step S05.2, based on the user's selection of the license plate dirtiness and occlusion confidence, such as Figure 6 As shown, vehicles suspected of having dirty or obstructed license plates are displayed on the user interaction interface, and based on the user's click instructions for a specific vehicle, a chain of evidence of captured images of the vehicle suspected of having dirty or obstructed license plates is displayed, and the corresponding captured images are displayed based on the click instructions. According to the user's input instructions, the user is supported to manually modify the cause of the abnormal license plate recognition and remove the vehicle from the suspected dirty or obstructed vehicles.

[0155] The present embodiment provides an abnormal checkpoint detection method, which detects and determines the root cause of the checkpoint position abnormality and recognition abnormality based on the vehicle passing trajectory, so as to further sort out possible checkpoint defects, adjust the checkpoint system parameters or notify the equipment operation and maintenance personnel to perform maintenance, so as to improve the accuracy and efficiency of checkpoint equipment recognition; by collecting data based on the checkpoint, determining the trajectory of each license plate, and obtaining multiple groups of checkpoint pairs based on the relative positions of multiple single checkpoints in the road network, and then combining multiple single vehicle passing data, it is possible to achieve effective data extraction of single vehicle passing data, thereby providing a basis for subsequent checkpoint abnormality recognition based on vehicle driving data; by combining the shortest distance of the road network and the actual snapshot Calculating the passing speed based on the time difference can provide more accurate vehicle speed information, which helps to effectively detect potential abnormal behaviors, such as speeding, data misreading, etc.; by verifying the passing time and calculating the passing speed based on the verified passing time, it is possible to eliminate the influence of the recognition abnormality on the passing time and passing speed when there are both position abnormalities and recognition abnormalities, so that the single passing data with position abnormalities can be obtained more accurately; by using the number and / or proportion of speeding passing data in a single passing data, it is possible to effectively identify which data is due to recognition errors, and which data is due to location or equipment problems, so that the cause of the abnormality can be identified. Maintenance; by determining the position abnormal checkpoints and identification abnormal checkpoints by the proportion of the single vehicle passing data with abnormal identification in the single vehicle passing data with abnormal location, it is possible to effectively determine which checkpoints are abnormal due to location problems and which are abnormal due to identification problems, and perform targeted equipment maintenance; by calculating the confidence level of the position abnormal checkpoints based on the proportion of the speeding vehicle passing data in multiple single vehicle passing data, and the proportion of the identification abnormal vehicle passing data in the speeding vehicle passing data, a reference for abnormal conditions and abnormality levels can be provided to users; by eliminating the vehicle passing time with vehicle stoppage, and then performing abnormality detection on the vehicle passing time in multiple single vehicle passing data, it is possible to better In order to effectively identify data deviations caused by recognition errors, thereby providing more reliable data support for the equipment operation and maintenance department; through the user interaction interface, the abnormal position checkpoint and the abnormal identification checkpoint, as well as the checkpoint collection data corresponding to the abnormal position checkpoint and the abnormal identification checkpoint are displayed, which can provide comprehensive information support for management or equipment operation and maintenance personnel; by removing the abnormal identification checkpoint whose abnormal reason is the abnormality of the license plate itself, the abnormal license plate data can be effectively identified and corrected, and these abnormal checkpoints and their related collection data can be intuitively displayed through the user interaction interface, and by removing the abnormal identification checkpoint due to the abnormality of the license plate itself, the effect of optimizing the system performance and user experience is achieved.

[0156] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0157] Based on the same inventive concept, the embodiment of the present application also provides an abnormal checkpoint detection device for implementing the abnormal checkpoint detection method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more abnormal checkpoint detection device embodiments provided below can refer to the limitations of the abnormal checkpoint detection method above, and will not be repeated here.

[0158] In one embodiment, Figure 7 As shown, an abnormal checkpoint detection device is provided, comprising: a data combination module 100, a data calculation module 200, an abnormality identification module 300 and a checkpoint determination module 400, wherein:

[0159] The data combination module 100 is used to combine the checkpoint collection data within a preset time to obtain single vehicle passing data under multiple checkpoint pairs; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once;

[0160] A data calculation module 200, used to calculate the vehicle travel data corresponding to the single vehicle passing data;

[0161] An abnormality identification module 300, used to determine the position abnormality and / or identify the abnormal single vehicle passing data based on the vehicle driving data corresponding to the plurality of single vehicle passing data;

[0162] The checkpoint determination module 400 is used to determine the checkpoint with abnormal position and / or identify the abnormal checkpoint based on the single vehicle passing data of the abnormal position and the abnormal identification.

[0163] Each module in the abnormal checkpoint detection device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0164] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, an abnormal checkpoint detection method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse, etc.

[0165] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0166] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the abnormal checkpoint detection method of any of the above embodiments is implemented:

[0167] Based on the checkpoint collection data within a preset time, the single vehicle passing data under multiple checkpoint pairs are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once;

[0168] Calculating vehicle travel data corresponding to the single vehicle passing data;

[0169] Based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or identifying the single vehicle passing data with abnormality respectively;

[0170] Based on the single vehicle passing data of the position abnormality and the identification abnormality, the position abnormality checkpoint is determined and / or the abnormal checkpoint is identified.

[0171] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the abnormal checkpoint detection method of any of the above embodiments is implemented:

[0172] Based on the checkpoint collection data within a preset time, the single vehicle passing data under multiple checkpoint pairs are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once;

[0173] Calculating vehicle travel data corresponding to the single vehicle passing data;

[0174] Based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or identifying the single vehicle passing data with abnormality respectively;

[0175] Based on the single vehicle passing data of the position abnormality and the identification abnormality, the position abnormality checkpoint is determined and / or the abnormal checkpoint is identified.

[0176] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the abnormal checkpoint detection method of any of the above embodiments:

[0177] Based on the checkpoint collection data within a preset time, the single vehicle passing data under multiple checkpoint pairs are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once;

[0178] Calculating vehicle travel data corresponding to the single vehicle passing data;

[0179] Based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining the position abnormality and / or identifying the single vehicle passing data with abnormality respectively;

[0180] Based on the single vehicle passing data of the position abnormality and the identification abnormality, the position abnormality checkpoint is determined and / or the abnormal checkpoint is identified.

[0181] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0182] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0183] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0184] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for detecting abnormal checkpoints, characterized in that: The abnormal checkpoint detection method comprises: Based on the checkpoint collection data within a preset time, the single vehicle passing data under multiple checkpoint pairs are combined; the single vehicle passing data includes the checkpoint collection data collected when the vehicle passes through the checkpoint pair once; Calculating vehicle travel data corresponding to the single vehicle passing data; Based on the vehicle driving data corresponding to the multiple single vehicle passing data, the single vehicle passing data with abnormal position and / or abnormal identification are determined respectively; based on the vehicle driving data corresponding to the multiple single vehicle passing data, the single vehicle passing data with abnormal position and / or abnormal identification is determined, including: if the number of speeding vehicle passing data and / or the proportion of the number of speeding vehicle passing data in the multiple single vehicle passing data meets the first preset condition, then the speeding vehicle passing data is determined as the single vehicle passing data with abnormal identification; the speeding vehicle passing data is a single vehicle passing data whose passing speed exceeds a preset speed threshold. data; if, among the multiple single vehicle passing data, the number of speeding vehicle passing data and / or the proportion of the number of speeding vehicle passing data in the multiple single vehicle passing data meet the second preset condition, the speeding vehicle passing data is determined as the single vehicle passing data with abnormal position; the first preset condition is that the number of speeding vehicle passing data is lower than the first threshold value and / or the proportion of the number of speeding vehicle passing data in the total amount of single vehicle passing data is lower than the first proportion threshold value; the second preset condition is that the number of speeding vehicle passing data is higher than the second threshold value and / or the proportion of the number of speeding vehicle passing data in the total amount of single vehicle passing data is greater than the second proportion threshold value; Based on the single vehicle passing data of the position abnormality and the identification abnormality, the position abnormality checkpoint is determined and / or the abnormal checkpoint is identified.

2. The abnormal checkpoint detection method according to claim 1, characterized in that: The vehicle driving data includes a passing speed and a passing time, and the calculation of the vehicle driving data corresponding to the single passing data includes: Calculate the passing time corresponding to each single passing data; the passing time is the difference in time when the vehicle is captured when passing through the checkpoint; The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair, the vehicle passing time, and the farthest capture distance of each single checkpoint in the single vehicle passing data.

3. The abnormal checkpoint detection method according to claim 2, characterized in that: After calculating the vehicle passing time corresponding to each single vehicle passing data, the method further includes: Based on the passing times corresponding to the multiple single passing data, the passing times corresponding to the multiple single passing data are verified to obtain the verified passing time corresponding to each single passing data; The vehicle passing speed is determined based on the shortest road network distance of the checkpoint pair in the single vehicle passing data, the verified vehicle passing time, and the farthest capture distance of each single checkpoint.

4. The abnormal checkpoint detection method according to claim 1, characterized in that: The determining of the position abnormal checkpoint and / or identifying the abnormal checkpoint based on the position abnormality and the single vehicle passing data of the identification abnormality comprises: Calculate the proportion of single vehicle passing data with abnormal identification to single vehicle passing data with abnormal location at each single checkpoint; If the proportion is less than a preset ratio threshold, the single checkpoint is determined as a checkpoint with abnormal position; otherwise, the single checkpoint is determined as a checkpoint with abnormal identification.

5. The abnormal checkpoint detection method according to claim 4, characterized in that: The abnormal checkpoint detection method also includes: The confidence of the position abnormality checkpoint is calculated based on the proportion of the speeding vehicle passing data in the multiple single vehicle passing data and the proportion of the identified abnormal vehicle passing data in the speeding vehicle passing data.

6. The abnormal checkpoint detection method according to claim 2, characterized in that: Based on the vehicle driving data corresponding to the plurality of single vehicle passing data, determining and identifying abnormal single vehicle passing data includes: Based on the vehicle passing times corresponding to the plurality of single vehicle passing data and an abnormality detection algorithm, the single vehicle passing data for identifying abnormalities is determined.

7. The abnormal checkpoint detection method according to claim 6, characterized in that: The determining and identifying abnormal single vehicle passing data based on the vehicle passing time and the abnormality detection algorithm corresponding to the plurality of single vehicle passing data further includes: Eliminate single vehicle passing data whose passing time is greater than the preset time threshold; Based on the vehicle passing times corresponding to the eliminated multiple single vehicle passing data and the abnormality detection algorithm, the abnormal single vehicle passing data is determined and identified.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images