A vehicle monitoring method and system based on virtual electronic fence

Abstract

The present invention relates to the technical field of vehicle supervision, and specifically discloses a vehicle supervision method and system based on a virtual electronic fence, wherein the method comprises the following steps: S1: determining a target vehicle based on an admitted vehicle model, and determining a target distance according to the location information of the target vehicle; S2: determining a pending vehicle according to the target distance, collecting the target distance of the pending vehicle within a collection period, generating data points and fitting to obtain a fitting curve; determining coordinate points, marking coordinate points that meet preset conditions as abnormal points, and determining the abnormal proportion in the coordinate point set; S3: calculating an evaluation coefficient according to the fitting curve and the abnormal proportion and determining the abnormal vehicle, and sending an early warning message for prompting according to the target distance of the abnormal vehicle. The present invention determines whether to perform an early warning prompt according to the comprehensive change of the target distance, thereby improving the accuracy of vehicle supervision.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle supervision, and in particular to a vehicle supervision method and system based on a virtual electronic fence. Background Art

[0002] Physical fences are a traditional security measure, usually built with materials such as metal and concrete. They are used to define boundaries and prevent unauthorized people or vehicles from entering specific areas. Their main function is to provide security protection through physical barriers. With the rapid development of technology today, physical fences can no longer meet the requirements of modern security management due to their high construction and maintenance costs and their inherent flexibility limitations, and are gradually being replaced by virtual electronic fences.

[0003] In the prior art, warnings are usually issued by monitoring the distance between the vehicle and the virtual electronic fence. When the distance between the vehicle and the virtual electronic fence is less than a preset distance threshold, the system will issue a warning message. However, in actual applications, the system may continue to alarm. For example, when certain lanes prohibit certain types of vehicles from passing, and the distance between lanes is close, although certain types of vehicles are close to the virtual electronic fence (such as the lane dividing line), they have not actually entered the limited area of ​​the virtual electronic fence. At this time, the system may still issue a continuous alarm signal, thereby affecting the accuracy of vehicle supervision. Summary of the invention

[0004] The purpose of the present invention is to provide a vehicle supervision method and system based on a virtual electronic fence to solve the following technical problems:

[0005] When early warning is given based on the distance between the vehicle and the virtual electronic fence, continuous alarms may occur, affecting the accuracy of vehicle supervision.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A vehicle supervision method based on a virtual electronic fence comprises the following steps:

[0008] S1: Determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, wherein the target distance is the distance between the target vehicle and the virtual electronic fence;

[0009] S2: When the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t);

[0010] Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn);

[0011] S3: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send a warning message according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

[0012] As a further solution of the present invention: in the step S3, the process of calculating the evaluation coefficient K according to the fitting curve and the abnormal ratio specifically includes:

[0013] A straight line passing through the reference point (0, f(Tsta)) and parallel to the horizontal plane is used as a reference line, and the proportion c of the time length corresponding to the fitting curve above the reference line to the acquisition period is counted;

[0014] The evaluation coefficient K is calculated by a formula, which specifically includes:

[0015]

[0016] Wherein, the target curve F(t)=f(t)-f(Tsta), ε represents a preset correction coefficient and ε>0.

[0017] As a further solution of the present invention: when the abnormal proportion b=0, the subsequent steps are not executed, and it is determined that the to-be-determined vehicle is not an abnormal vehicle.

[0018] As a further solution of the present invention: when the time length corresponding to the fitting curve above the reference line accounts for a proportion c≥0.9 of the acquisition period, the subsequent steps are not executed, and it is determined that the pending vehicle is not an abnormal vehicle.

[0019] As a further solution of the present invention, the process of sending warning information for prompting according to the target distance of the abnormal vehicle specifically includes:

[0020] Set a target distance range [0, Dmax], set m target distance levels within the target distance range at preset target distance intervals, where m is a preset value, and the shorter the target distance, the higher the target distance level;

[0021] Determine the target distance f(Tsta+Tys) corresponding to the time point at which the acquisition cycle ends, determine the target distance level A where the target distance f(Tsta+Tys) is located, and send an A-level warning message for prompting.

[0022] As a further solution of the present invention: in the step S2, when the target vehicle enters the target area within the collection period, the subsequent steps are not executed and the highest level warning information is sent for prompting.

[0023] As a further solution of the present invention: in the step S1, the process of determining the target distance according to the position information specifically includes: verifying the position information, taking the position information that meets the preset verification rules as the target information, and determining the target distance based on the target information.

[0024] A vehicle monitoring system based on a virtual electronic fence, comprising:

[0025] Acquisition module: determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, where the target distance is the distance between the target vehicle and the virtual electronic fence;

[0026] Processing module: when the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t);

[0027] Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn);

[0028] Evaluation module: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send warning information according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

[0029] Beneficial effects of the present invention: In the present invention, the target vehicle is first determined based on the allowed vehicle type, and the target distance is determined based on the position information of the target vehicle; it is worth noting that the reason for determining the target vehicle is: vehicles of specific models are allowed to enter certain areas (such as the left lane is generally for small vehicles to pass, and large vehicles need to drive on the right lane, then the left lane is the target area), there is no need to monitor the vehicle types that can enter the target area, and the target distance is the basis for subsequent determination of whether to issue an early warning. Before entering the target area, the shorter the target distance, the more it should issue an early warning. As recorded in the background technology, simply considering the distance may result in continuous alarms, because the distance between the lanes is short, which may have reached the level of early warning. Therefore, after the distance reaches the level of early warning, In order to improve the accuracy of vehicle supervision, the target distance is periodically obtained during the acquisition period, and fitting is performed to determine the change trend of the target distance, and the evaluation coefficient is comprehensively calculated according to the change trend of the target distance during the acquisition period; it can be understood that in the process of determining the abnormal point, when the target distance Da≤f(Tsta) and the tangent slope Ka≤0, it means that compared with the position at the time point Tsta, the vehicle is closer to the virtual electronic fence, and the trend is to continue to approach the virtual electronic fence, so it should be marked as an abnormal point, and the more abnormal points there are, the greater the abnormal proportion, and the more warning information should be issued; finally, when the evaluation coefficient is less than or equal to the preset evaluation coefficient threshold, the pending vehicle is marked as an abnormal vehicle, and a warning message is sent according to the target distance of the abnormal vehicle for prompting. The present invention determines whether to issue a warning prompt through the comprehensive change of the target distance, thereby improving the accuracy of vehicle supervision. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below in conjunction with the accompanying drawings.

[0031] Figure 1 This is a structural schematic diagram of a vehicle monitoring method based on a virtual electronic fence of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] See also Figure 1 As shown, the present invention is a vehicle supervision method based on a virtual electronic fence, comprising the following steps:

[0034] S1: Determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, wherein the target distance is the distance between the target vehicle and the virtual electronic fence;

[0035] S2: When the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t);

[0036] Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn);

[0037] S3: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send a warning message according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

[0038] It should be noted that the target vehicle is first determined based on the allowed vehicle type, and the target distance is determined based on the location information of the target vehicle; it is worth noting that the reason for determining the target vehicle is: vehicles of specific models are allowed to enter certain areas (such as the left lane is generally for small vehicles to pass, and large vehicles need to drive on the right lane, then the left lane is the target area), and there is no need to monitor the vehicle types that can enter the target area. The target distance is the basis for subsequent determination of whether to issue an early warning. Before entering the target area, the shorter the target distance, the more it should issue an early warning. As recorded in the background technology, simply considering the distance may result in continuous alarms, because the distance between the lanes is short, and it may have reached the level of early warning. Therefore, after the distance reaches the level of early warning, in order to improve the accuracy of vehicle supervision, the target distance is periodically obtained during the collection cycle. Target distance, and fitting, determine the change trend of target distance, comprehensively calculate the evaluation coefficient according to the change trend of target distance within the acquisition period, and subsequent judgment will be made only after the target distance reaches a certain level, so as to ensure the efficiency of analysis and judgment and effective use of resources; it can be understood that in the process of determining abnormal points, when the target distance Da≤f(Tsta) and the tangent slope Ka≤0, it means that compared with the position at the time point Tsta, the vehicle is closer to the virtual electronic fence, and the trend is to continue to approach the virtual electronic fence, so it should be marked as an abnormal point, the more abnormal points there are, the greater the abnormal proportion, and the more warning information should be issued; finally, when the evaluation coefficient is less than or equal to the preset evaluation coefficient threshold, the pending vehicle is marked as an abnormal vehicle, and a warning message is sent according to the target distance of the abnormal vehicle for prompting;

[0039] It is understandable that when the warning information is sent, the vehicle's trajectory data will be generated synchronously, and the time granularity of the trajectory data is determined based on the actual situation, such as 5 minutes, 10 minutes, etc., which is convenient for the staff to view and perform analysis and other operations; it is worth noting that in the present invention, the jurisdiction of the target vehicle can be determined according to the target information, and the changes in the jurisdiction of the target vehicle can be recorded to ensure that the vehicle's jurisdiction change record is updated in real time to facilitate vehicle supervision. Specifically, if the vehicle enters another jurisdiction for the first time, the location where the vehicle first enters the jurisdiction is recorded in the cache and output to the entry and exit record table; if the vehicle has a historical record, it is determined whether the current vehicle position is the same as the jurisdiction position recorded last time. If If they are the same, no operation will be performed; if they are different, it means that the vehicle has left the last jurisdiction and entered another jurisdiction; in this process, the current location information of the vehicle is obtained according to the longitude and latitude of the vehicle, and the data is obtained based on the national offline vector map to ensure the timeliness of the acquisition and analysis of the vehicle's location information. The national offline vector map ensures that it is applicable to different server network environments and meets the customizable location information requirements; and in the process of recording the jurisdiction of the vehicle, it is determined according to the vehicle's trajectory data. For example, for five minutes of trajectory data, it is necessary to determine whether the trajectory exceeding 2.5 minutes is in the jurisdiction (the proportion can be set by experience), so as to avoid the sporadic nature of trajectory data in the area.

[0040] In another preferred embodiment of the present invention, in step S3, the process of calculating the evaluation coefficient K according to the fitting curve and the abnormal ratio specifically includes:

[0041] A straight line passing through the reference point (0, f(Tsta)) and parallel to the horizontal plane is used as a reference line, and the proportion c of the time length corresponding to the fitting curve above the reference line to the acquisition period is counted;

[0042] The evaluation coefficient K is calculated by a formula, which specifically includes:

[0043]

[0044] Wherein, the target curve F(t)=f(t)-f(Tsta), ε represents a preset correction coefficient and ε>0.

[0045] It is worth noting that when the target distance Da≤f(Tsta) and the tangent slope Ka≤0, it means that compared with the position at the time point Tsta, the vehicle is closer to the virtual electronic fence, and the trend is to continue to approach the virtual electronic fence. Therefore, it should be marked as an abnormal point. The more abnormal points there are, the greater the abnormal proportion, and the more warning information should be issued. Therefore, the larger the abnormal proportion b, the smaller the evaluation coefficient K should be. The relationship between the remaining parameters and the evaluation coefficient K can refer to the above ideas and will not be repeated here. The correction coefficient ε can be set specifically by experience.

[0046] In another preferred embodiment of the present invention, when the abnormality ratio b=0, the subsequent steps are not executed, and it is determined that the to-be-determined vehicle is not an abnormal vehicle.

[0047] It can be understood that the abnormal proportion b=0 means that the behavior of the vehicle during the entire collection period is completely normal and there is no tendency to approach the virtual electronic fence. Therefore, there is no need to perform additional steps and it is directly determined that the target vehicle is not an abnormal vehicle.

[0048] In another preferred embodiment of the present invention, when the time length corresponding to the fitting curve above the reference line accounts for a proportion c≥0.9 of the acquisition period, subsequent steps are not executed, and it is determined that the pending vehicle is not an abnormal vehicle.

[0049] It should be noted that when the ratio c ≥ 0.9, it means that the target vehicle is not closer to the virtual electronic fence than the initial state (i.e., the position of the target vehicle at time point Tsta) for most of the time. Therefore, there is no abnormality in the target vehicle. In this way, vehicles that do not need to be monitored can be quickly identified, thereby improving the efficiency and accuracy of the system and avoiding invalid monitoring and false alarms.

[0050] In another preferred embodiment of the present invention, the process of sending warning information for prompting according to the target distance of the abnormal vehicle specifically includes:

[0051] Set a target distance range [0, Dmax], set m target distance levels within the target distance range at preset target distance intervals, where m is a preset value, and the shorter the target distance, the higher the target distance level;

[0052] Determine the target distance f(Tsta+Tys) corresponding to the time point at which the acquisition cycle ends, determine the target distance level A where the target distance f(Tsta+Tys) is located, and send an A-level warning message for prompting.

[0053] It should be noted that graded warnings are provided based on the actual distance between the vehicle and the fence to ensure that the warning information can accurately reflect the risk level of the vehicle. For example, when the vehicle is far away from the fence, a lower level prompt is sent, and when the vehicle is closer, a higher level warning is sent. Through the intelligent multi-level warning mechanism, the system can adjust the warning intensity in time according to the proximity of the vehicle to ensure the accuracy and timeliness of the warning.

[0054] In another preferred embodiment of the present invention, in the step S2, when the target vehicle enters the target area within the collection period, the subsequent steps are not executed and the highest level warning information is sent for prompting.

[0055] It is understandable that when a target vehicle (i.e. a vehicle that does not meet the access conditions) directly enters the target area (such as a large vehicle entering a lane that only allows small vehicles to pass), it means that the vehicle has broken through the virtual electronic fence and violated the access regulations of the area. In this case, the vehicle's behavior is no longer a "potential risk" but an "actual violation". The system needs to respond immediately instead of continuing to perform additional analysis processes, thereby skipping unnecessary steps and shortening response time.

[0056] In another preferred embodiment of the present invention, in the step S1, the process of determining the target distance according to the position information specifically includes: verifying the position information, taking the position information that meets the preset verification rules as the target information, and determining the target distance based on the target information.

[0057] It is worth noting that the verification rules include but are not limited to longitude and latitude range verification, timestamp verification, etc. For example, the longitude and latitude in the vehicle location information should be within the longitude and latitude range. If it exceeds the range, it means that the location information is inaccurate. At this time, these data need to be eliminated to ensure the accuracy of subsequent judgments; the data is cleaned by analyzing indicators such as positioning time, speed, direction, longitude and latitude, and altitude to ensure its accuracy and reliability.

[0058] A vehicle monitoring system based on a virtual electronic fence, comprising:

[0059] Acquisition module: determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, where the target distance is the distance between the target vehicle and the virtual electronic fence;

[0060] Processing module: when the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t);

[0061] Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn);

[0062] Evaluation module: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send warning information according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

[0063] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A vehicle supervision method based on a virtual electronic fence, characterized in that: The following steps are involved: S1: Determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, wherein the target distance is the distance between the target vehicle and the virtual electronic fence; S2: When the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t); Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn); S3: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send a warning message according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

2. A vehicle monitoring method based on virtual electronic fence according to claim 1, characterized in that: In the step S3, the process of calculating the evaluation coefficient K according to the fitting curve and the abnormal ratio specifically includes: A straight line passing through the reference point (0, f(Tsta)) and parallel to the horizontal plane is used as a reference line, and the proportion c of the time length corresponding to the fitting curve above the reference line to the acquisition period is counted; The evaluation coefficient K is calculated by a formula, which specifically includes: Wherein, the target curve F(t)=f(t)-f(Tsta), ε represents a preset correction coefficient and ε>0.

3. A vehicle monitoring method based on virtual electronic fence according to claim 2, characterized in that: When the abnormality ratio b=0, the subsequent steps are not executed, and it is determined that the to-be-determined vehicle is not an abnormal vehicle.

4. A vehicle monitoring method based on virtual electronic fence according to claim 2, characterized in that: When the time length corresponding to the fitting curve above the reference line accounts for a proportion c≥0.9 of the acquisition period, the subsequent steps are not performed, and it is determined that the pending vehicle is not an abnormal vehicle.

5. The vehicle monitoring method based on virtual electronic fence according to claim 1 is characterized in that: The process of sending warning information according to the target distance of the abnormal vehicle specifically includes: Set a target distance range [0, Dmax], set m target distance levels within the target distance range at preset target distance intervals, where m is a preset value, and the shorter the target distance, the higher the target distance level; Determine the target distance f(Tsta+Tys) corresponding to the time point at which the acquisition cycle ends, determine the target distance level A where the target distance f(Tsta+Tys) is located, and send an A-level warning message for prompting.

6. A vehicle monitoring method based on virtual electronic fence according to claim 1, characterized in that: In the step S2, when the target vehicle enters the target area within the collection period, the subsequent steps are not executed and the highest level warning information is sent for prompting.

7. A vehicle monitoring method based on virtual electronic fence according to claim 1, characterized in that: In the step S1, the process of determining the target distance according to the position information specifically includes: verifying the position information, taking the position information that meets the preset verification rules as the target information, and determining the target distance based on the target information.

8. A vehicle monitoring system based on virtual electronic fence, characterized in that: include: Acquisition module: determine the allowed vehicle type in the preset target area, use the boundary of the target area as a virtual electronic fence, mark the vehicle that is not the allowed vehicle type as the target vehicle, collect the position information of the target vehicle, and determine the target distance D according to the position information, where the target distance is the distance between the target vehicle and the virtual electronic fence; Processing module: when the target distance is less than or equal to the preset distance threshold Dmax, the target vehicle is marked as a pending vehicle, and a collection period [Tsta, Tsta+Tys] is set, where Tsta represents the time point when the target vehicle is marked as a pending vehicle, and Tys represents the preset time length. The target distance of the pending vehicle is periodically collected within the collection period to generate data points Sa(a, Da), where a≤n, n represents the total number of times the target distance is collected, and Da represents the target distance of the pending vehicle collected for the ath time. The data points are fitted to obtain a fitting curve f(t); Determine the coordinate point Za(Da, Ka), where Ka represents the tangent slope of the data point Sa in the fitting curve. When the target distance Da≤f(Tsta) and the tangent slope Ka≤0, mark the coordinate point Za(Da, Ka) as an abnormal point, and determine the abnormal proportion b. The abnormal proportion b represents the proportion of the number of abnormal points in the coordinate point set to the total number of elements. The coordinate point set PDjh=(Z1, Z2, ..., Zn); Evaluation module: Calculate the evaluation coefficient K according to the fitting curve and the abnormal proportion. When the evaluation coefficient K≤K', mark the pending vehicle as an abnormal vehicle, send warning information according to the target distance of the abnormal vehicle, and K' represents a preset evaluation coefficient threshold.

Images