Vehicle tracking method and system, storage medium and server

Abstract

The present application provides a vehicle tracking method, a vehicle system, a storage medium and a server. The vehicle tracking method comprises: recording vehicle feature information of a target vehicle when the target vehicle is driven in; performing multi-frame license plate recognition on the basis of the vehicle feature information; if a target license plate number that meets a recognition threshold earliest is present, using the target license plate number as a final license plate number of the target vehicle, and adding the final license plate number as a corresponding driving-in vehicle license plate number of a parking area corresponding to the target vehicle; and when a driving-out vehicle which cannot be recognized is present in the parking area, using, to match the current driving-out vehicle, the driving-in vehicle license plate number corresponding to the parking area, so as to determine vehicle information of the driving-out vehicle.

Description

Vehicle tracking method, system, storage medium and server

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on July 4, 2023, with application number 202310812852.X. The entire contents of this application are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of image detection, for example, to a vehicle tracking method, system, storage medium and server. Background Art

[0003] The vehicle entry and exit recognition algorithm can be used to automatically report the vehicle's entry or exit status, entry or exit time, and license plate information when a vehicle enters or exits a monitored parking area. In some scenarios such as self-service charging stations, 4S shop car washes, and gas station refueling stations, this algorithm can facilitate vehicle management in individual parking spaces. While reducing manpower and financial resources, it can intelligently provide vehicle entry and exit warnings, monitor parking time, and manage vehicle information, facilitating subsequent vehicle tracking by the platform or relevant personnel.

[0004] However, using the vehicle target detected by the vehicle entry and exit recognition algorithm as the basis for vehicle entry and exit judgment is unstable. For example, when the vehicle is blocked or the vehicle features are not obvious, the vehicle target identifier (Identifier, ID) is not unique, tracking is unstable, and the vehicle target cannot be locked, resulting in low vehicle tracking detection accuracy.

[0005] Summary of the Invention

[0006] The present application provides a vehicle tracking method, system, storage medium and server, which can improve the accuracy of vehicle tracking detection.

[0007] This application provides a vehicle tracking method, comprising:

[0008] The vehicle characteristic information of the target vehicle is recorded when the target vehicle enters; multi-frame license plate recognition is performed based on the vehicle characteristic information; if there is a target license plate number that meets the recognition threshold the earliest, the target license plate number is used as the final license plate number of the target vehicle, and the final license plate number is added as the entry license plate number of the parking area corresponding to the target vehicle; when the exiting vehicle of the parking area cannot be identified, the exiting vehicle is matched with the entry license plate number of the parking area to determine the vehicle information of the exiting vehicle.

[0009] The present application also provides a vehicle tracking system, comprising:

[0010] The vehicle detection module is configured to record the vehicle characteristic information of the target vehicle when the target vehicle enters; the entry number detection module is configured to perform multi-frame license plate recognition based on the vehicle characteristic information; the vehicle alarm module is configured to use the target license plate number as the final license plate number of the target vehicle when there is a target license plate number that meets the recognition threshold the earliest, and add the final license plate number as the entry license plate number of the parking area corresponding to the target vehicle; the vehicle matching module is configured to use the entry license plate number of the parking area to match the exiting vehicle when the exiting vehicle cannot be identified, so as to determine the vehicle information of the exiting vehicle.

[0011] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the vehicle tracking method described above is implemented.

[0012] The present application also provides a server, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the vehicle tracking method described above when calling the computer program in the memory. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] FIG1 is a flow chart of a vehicle tracking method provided in an embodiment of the present application;

[0014] FIG2 is a schematic structural diagram of a vehicle tracking system provided in an embodiment of the present application;

[0015] FIG3 is a schematic diagram of the structure of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. The described embodiments are part of the embodiments of the present application.

[0017] Please refer to FIG1 , which is a flow chart of a vehicle tracking method provided by an embodiment of the present application, the method comprising:

[0018] S101: Recording vehicle feature information of a target vehicle when the target vehicle enters.

[0019] S102: Perform multi-frame license plate recognition based on the vehicle feature information.

[0020] S103: If there is a target license plate number that meets the recognition threshold the earliest, the target license plate number is used as the final license plate number of the target vehicle, and the final license plate number is added as the entry license plate number of the parking area corresponding to the target vehicle.

[0021] S104: When the vehicle leaving the parking area cannot be identified, the vehicle leaving the parking area is matched with the license plate number of the vehicle entering the parking area to determine the vehicle information of the vehicle leaving the parking area.

[0022] Step S101 determines when a target vehicle enters and records its vehicle characteristic information. The target vehicle can be any vehicle; in this embodiment, it refers to the vehicle being tracked and monitored. The vehicle characteristic information is not limited here. Typically, the license plate is used as vehicle characteristic information due to its unique identification. Vehicle characteristic information may also include other information, including vehicle appearance and vehicle type.

[0023] There are no specific restrictions on how to detect vehicle entry. In one embodiment, if a target vehicle is detected in the video stream corresponding to the entrance, it can be considered as the target vehicle entering. In another embodiment, the vehicle's front direction can also be detected to determine whether the target vehicle is entering or exiting. Vehicle detection can be performed on the front of the vehicle when entering or exiting the vehicle in the forward direction. When entering or exiting the vehicle in the reverse direction, the rear of the vehicle can be detected.

[0024] The vehicle's entry and exit status can be determined based on the relationship between the front or rear of the vehicle and the monitored parking area. For example, if more than three-quarters of the front or rear of the vehicle intersects the parking area, the vehicle is considered to be within the parking area. If less than half of the front or rear of the vehicle intersects the parking area, the vehicle is considered to be outside the parking area. The vehicle's status at adjacent moments is determined (vehicle within the parking area, vehicle not within the parking area, or the proportion of intersection between the vehicle and the parking area), and these states are compared to determine the vehicle's entry and exit status based on the comparison results.

[0025] The parking area in this step can be the parking area within the monitoring range. That is, the monitoring range of each camera is the corresponding parking area. It can also be understood that the parking spaces within the monitoring range are considered the parking area. This parking area usually corresponds to a monitored parking space. At any time, only one entry or exit occurs in the same parking area.

[0026] The vehicle's state is determined based on the intersection of the vehicle detection frame and the parking area. The vehicle's entry or exit is determined based on the changes between the previous and current states. Specifically, if the vehicle's previous state was "out of the parking area" and its current state is "in the parking area," the vehicle is considered to have entered the parking area. After determining that the vehicle has entered the parking area, the current vehicle position is used as the previous state position to facilitate subsequent vehicle exit determinations. Conversely, if the vehicle's previous state was "in the parking area" and its current state is "out of the parking area," the target vehicle is considered to have exited the parking area. Similarly, after determining that the target vehicle has exited the parking area, the current vehicle position is used as the previous state position to facilitate subsequent vehicle entry determinations. The vehicle's previous state position is not updated in every frame; it is only updated after a vehicle is determined to have entered or exited the parking area.

[0027] When recording the vehicle characteristic information of the target vehicle, the target vehicle can be continuously recorded, and the entry information of the target vehicle can also be uploaded.

[0028] Afterwards, the incoming vehicle's vehicle characteristic information is detected and identified, requiring multi-frame license plate recognition based on this vehicle characteristic information. To ensure the accuracy of the target vehicle's license plate number, multi-frame license plate recognition can be performed on the incoming target vehicle, and the optimal result from the multi-frame license plate recognition is selected as the target vehicle's final license plate number. Taking the license plate as the vehicle characteristic information, for example, when the number of identical recognitions for any incoming license plate number reaches a threshold the earliest, that license plate number is selected as the target license plate number, i.e., the alarm license plate number for the incoming vehicle. The target license plate number is also recorded as the entry license plate number for that parking area, graphPlateNo[graphAlarmIn]. In this embodiment, only one vehicle can enter or exit a parking area at a time, and each entry and exit constitutes a closed-loop state for the same vehicle. For exiting vehicles, a strong association is performed between the identified exiting vehicle and the parking area license plate, and the entry license plate number recorded with the same parking area ID is assigned as the alarm license plate number for the exiting vehicle. Afterwards, the license plate number of the vehicle entering the parking area can be cleared to facilitate the recording and matching of the next target license plate number. When monitoring parking spaces, it is easy for a vehicle to be captured temporarily passing through the monitoring scene. To prevent a temporarily passing vehicle from being mistaken for an exiting vehicle and thus mistakenly associated with the entry license plate number, based on the above embodiment, a character matching strategy can be added to prevent the abnormal situation where different vehicles entering and exiting are associated. That is, the license plate number of the exiting vehicle is only recognized once. If the length of consecutive similar characters between the license plate number of the exiting vehicle and the license plate number entering the parking area is less than a preset character length threshold, for example, if the length of consecutive similar characters is less than 3, it is determined that the exiting vehicle and the target vehicle are different vehicles, and no parking area association operation is performed on the exiting vehicle. If the length of consecutive similar characters between the license plate number of the exiting vehicle and the license plate number entering the parking area is greater than or equal to the preset character length threshold, it can be determined that the exiting vehicle and the target vehicle are the same vehicle, and the entry license plate number recorded with the same parking area ID is then assigned as the alarm license plate number of the exiting vehicle.

[0029] In an embodiment of the present application, when tracking a vehicle, for scenarios where the vehicle behavior is uncontrollable or the scene survey is uncontrollable, such as excessive speed or tilt of the vehicle affecting normal vehicle identification, or a camera height being too low resulting in a larger parking area for detection, the parking area status characteristics and the license plate information of the vehicle entering the parking area can be used to match the vehicle leaving a parking area, and an association mechanism between the vehicle and the parking area can be established. This improves the robustness of vehicle detection, reduces the limitations of scene and visual conditions, and makes the vehicle entry and exit identification less affected by environmental or human interference, thereby ensuring the integrity of the alarm evidence chain of the parking vehicle entry and exit process and the reliability of the license plate recognition results.

[0030] On the basis of the previous embodiment, after recording the vehicle characteristic information of the target vehicle, the entry and exit information of the target vehicle can be continuously recorded and uploaded. The vehicle ID of the target vehicle can be stored, and the corresponding entry and exit information can be stored according to the vehicle ID.

[0031] For entry information, after determining the entering vehicle, the entry information of the vehicle ID can be continuously recorded. If the vehicle is subsequently determined to have been in the parking area, the entry status information and detection frame information of the vehicle ID will be continuously recorded and uploaded to facilitate the subsequent selection of the vehicle license plate number. At the same time, it also prevents the risk of underreporting caused by uploading the entry information only once but not detecting the target license plate number.

[0032] For exit information, after determining that a vehicle has exited, the vehicle ID can continuously record the exit information, which is convenient for character matching operations when later associating with the incoming license plate number, to avoid the failure to detect the exit license plate in one go affecting the character matching. After that, if the vehicle is determined to be completely outside the parking area, the exit status information and detection frame information of the vehicle ID can be continuously recorded and uploaded, which can also prevent missed reports. Although there is no need to select a license plate number based on the exit information, but to directly associate the exit vehicle with the incoming license plate number, the above-mentioned character matching operation must be performed to determine whether the exit vehicle is the same vehicle as the target vehicle. If the exit information is only uploaded once, there may be a risk that the license plate cannot be detected in this frame, resulting in the inability to use character matching. Therefore, the exit information will also be continuously reported.

[0033] To avoid multiple alarms for a vehicle, you can use not only the vehicle ID as an alarm reference for the vehicle's entry and exit process, but also the parking area ID of the parking area where the vehicle is located to solve the multiple alarm problem. The following steps can be included:

[0034] When a vehicle enters the parking area, the exit reporting detection switch of the parking area is turned on; when the exit reporting detection switch is not turned on, the alarm is stopped for vehicles leaving the parking area; similarly, after the vehicle exits the parking area, the entry reporting detection switch of the parking area can be turned on; when the entry reporting detection switch is not turned on, the alarm is stopped for vehicles entering the parking area.

[0035] Independent detection switches are set for the entry and exit processes. The target vehicle will be alarmed only when the detection switch is turned on, realizing alarm filtering.

[0036] On this basis, the entry reporting detection switch and the exit reporting detection switch can be controlled to report the final entry and exit of the vehicle, including the following steps:

[0037] Determine whether the entry reporting detection switch of the parking area is turned on; if so, alarm the entering vehicle and upload the vehicle feature information and vehicle shooting information, and turn off the entry reporting detection switch; determine whether the exit reporting detection switch of the parking area is turned on; if so, alarm the exiting vehicle and upload the vehicle feature information and vehicle shooting information; turn off the exit reporting detection switch.

[0038] Alarms are issued for target vehicles that have passed the above alarm filtering process, and judgment is performed according to the above process. If the entry reporting detection switch is on, an alarm is executed, including the entry vehicle target alarm entering the "IN" state, and uploading the event image, vehicle license plate snapshot image and vehicle license plate information. Similarly, the above operations can also be performed for the exit process.

[0039] Based on the above embodiments, there may be a situation where the vehicle leaving the parking area cannot be identified. The reason for this situation may be that the vehicle exit behavior cannot be effectively detected due to engineering survey reasons, such as at night or in extreme weather conditions. It may be that the vehicle has exceeded the monitoring screen when it leaves, making it impossible to detect the front or rear of the vehicle. It may be that the vehicle ID is variable due to night scenes and partial occlusion of the vehicle. It may also be that the license plate is exposed and the vehicle is leaving too fast, resulting in the vehicle not being detected. In the case that the vehicle cannot be effectively detected and tracked, the exit event can be supplemented based on the vacant space exit strategy. After the exit event is supplemented, the exit vehicle corresponding to the exit event can be matched with the license plate number of the vehicle entering the parking area to supplement the license plate number of the unidentified exit vehicle. The vacant space exit strategy can be as follows to report the unreported exit vehicles:

[0040] The first step is to detect the vehicle target in the parking area if an entry alarm target is detected, wherein the vehicle target includes the whole vehicle detection frame, the front or rear vehicle detection frame and the license plate detection frame; the second step is to determine that the parking area is in a vacant state if the whole vehicle detection frame, the front or rear vehicle detection frame and the license plate detection frame are not in the parking area, and the vacant space count is increased by 1; the third step is to return the vacant space count to zero when at least one of the whole vehicle detection frame, the front or parking space detection frame and the license plate detection frame is in the parking area; the fourth step is to report the unreported exit event if the vacant space count meets the set vacant space threshold, and virtualize the vehicle detection frame; the license plate number of the exiting vehicle corresponding to the exit event is the license plate number of the last entry vehicle in the parking area.

[0041] Specifically, for detected entering vehicles, vacancy detection is performed based on the parking area they are located in. To prevent adjacent parking areas from being too close, causing vehicles in other parking areas to partially intersect with the currently monitored area and affect vacancy determination in that area, a threshold intersection condition can be set when detecting vehicle targets within the parking area. When performing vacancy supplementation, if a vehicle target cannot be identified, the unidentified exiting vehicle is directly associated with the entering vehicle's license plate number, without the need for character matching.

[0042] After a normal exit alarm (i.e., detection of a vehicle exit), or after a supplementary report of an unreported exit event (i.e., no vehicle exit is detected), the license plate number recorded for the parking area can be cleared. This is also considered as the vehicle exiting the parking area, meaning the parking area's entry detection switch can be enabled.

[0043] Based on the above embodiment, detecting a vehicle target in a parking area includes:

[0044] If the area ratio of the intersection parking space area of ​​the whole vehicle detection frame and the maximum inscribed rectangle of the parking space area to the whole vehicle detection frame is greater than the first threshold, it is confirmed that the whole vehicle is in the parking space area (or the whole vehicle detection frame is in the parking space area); if the area ratio of the intersection parking space area of ​​the front or rear vehicle detection frame and the maximum inscribed rectangle of the parking space area to the front or rear vehicle detection frame is greater than the second threshold, it is confirmed that the front or rear vehicle is in the parking space area (or the front or rear vehicle detection frame is in the parking space area); if there is an intersection parking space area between the license plate detection frame and the parking space area, it is confirmed that the license plate is in the parking space area (or the license plate detection frame is in the parking space area).

[0045] The evaluation criteria for the entire vehicle and the front and rear ends of a vehicle are based on comparing the intersection of the target detection frame and the maximum inscribed rectangle of the parking space with the target detection frame. This is because the parking space is an irregular quadrilateral. When calculating the intersection with the target detection frame, the general program logic automatically changes to calculating the intersection of the quadrilateral's circumscribed rectangle with the target detection frame. Furthermore, due to the camera's viewing angle and the size of passing vehicles, some passing vehicles may intersect with the circumscribed rectangle of the parking space in some cases. To avoid the influence of passing vehicles, the maximum inscribed rectangle of the parking space is used, combined with an intersection threshold. This allows for identifying whether there are vehicle targets within the parking space while preventing passing vehicles from affecting the parking space status. The above criteria are not used for license plates because they are relatively small. By setting the license plate intersection threshold higher, passing vehicles cannot meet the evaluation criteria corresponding to this intersection threshold. Therefore, the default circumscribed rectangle of the parking space can be used for the calculation of the license plate detection frame.

[0046] The first threshold and the second threshold are not limited here and can be set by those skilled in the art. They can be the same or different values. For example, if the first threshold and the second threshold are both 0.5, a feasible threshold intersection condition can be as follows:

[0047] The intersection condition of the vehicle threshold is: the vehicle detection box S m The maximum inscribed rectangle A of the parking area Internal The intersection parking area S mc With the vehicle detection frame S m When the area ratio is greater than 0.5, the vehicle is considered to be in the parking area (vehicle detection frame S m In the parking area), the vehicle detection frame S m The maximum inscribed rectangle A of the parking area Internal The intersection parking area S mc With the vehicle detection frame S m When the area ratio is not greater than 0.5, the vehicle is considered not to be within the parking area (vehicle detection frame S m Not in the parking area). The mathematical conditions are as follows: (S m ∩A Internal ) / S m >0.5

[0048] The intersection condition of the front and rear thresholds is: the front or rear detection box S v The maximum inscribed rectangle A of the parking area Internal The intersection parking area S vc With the front or rear detection frame S v When the area ratio is greater than 0.5, the front or rear of the vehicle is considered to be within the parking area (the front or rear detection box S v In the parking area), the front or rear detection frame S v The maximum inscribed rectangle A of the parking area Internal The intersection parking area S vc With the front or rear detection frame S v When the area ratio is not greater than 0.5, it is considered that the front or rear of the vehicle is not in the parking area (the front or rear detection box S v Not in the parking area). The mathematical conditions are as follows: (S v ∩A Internal ) / S v >0.5

[0049] The license plate threshold intersection condition is: license plate detection frame S p Intersecting parking area S with parking area A pc If the area is greater than 0.5, the license plate is considered to be within the parking area (license plate detection frame S p In the parking area), the license plate detection frame S pIntersecting parking area S with parking area A pc If the area is not greater than 0.5, it is considered that the license plate is not within the parking area (license plate detection frame S p Not in the parking area). The mathematical conditions are as follows: (S p ∩A) / A>0.5

[0050] In addition, after a normal exit alarm or a supplementary report of an unreported exit event, the entry license plate number recorded in the parking area can be cleared in time to facilitate the recording of the license plate number of the next vehicle entering the parking area.

[0051] A vehicle tracking system provided in an embodiment of the present application is introduced below. The vehicle tracking system described below and the vehicle tracking method described above can be referenced to each other.

[0052] Referring to FIG. 2 , FIG. 2 is a schematic diagram of the structure of a vehicle tracking system provided in an embodiment of the present application. The present application further provides a vehicle tracking system, comprising:

[0053] The vehicle detection module 10 is configured to record the vehicle characteristic information of the target vehicle when the target vehicle enters; the entry number detection module 20 is configured to perform multi-frame license plate recognition based on the vehicle characteristic information; the vehicle alarm module 30 is configured to use the target license plate number as the final license plate number of the target vehicle when there is a target license plate number that meets the recognition threshold the earliest, and add the final license plate number as the entry license plate number of the parking area corresponding to the target vehicle; the vehicle matching module 40 is configured to use the entry license plate number of the parking area to match the vehicle leaving the parking area when the vehicle leaving the parking area cannot be identified, so as to determine the vehicle information of the vehicle leaving the area.

[0054] Based on the above embodiment, the vehicle matching module 40 includes:

[0055] The matching unit is configured to perform a license plate recognition on the outgoing vehicle in the parking area by using character matching when there is an outgoing vehicle in the parking area; if the length of consecutive similar characters between the license plate number of the outgoing vehicle and the target license plate number of the target vehicle entering the parking area is less than a preset character length threshold, it is confirmed that the outgoing vehicle and the target vehicle are different vehicles; if the length of consecutive similar characters between the license plate number of the outgoing vehicle and the target license plate number is greater than or equal to the preset character length threshold, it is confirmed that the outgoing vehicle and the target vehicle are the same vehicle.

[0056] Based on the above embodiment, the vehicle tracking system further includes:

[0057] The exit detection switch setting module is configured to turn on the exit reporting detection switch of the parking area when a vehicle enters the parking area; and to stop issuing an alarm for vehicles exiting the parking area when the exit reporting detection switch is not turned on. The entry detection switch setting module is configured to turn on the entry reporting detection switch of the parking area after a vehicle exits the parking area; and to stop issuing an alarm for vehicles entering the parking area when the entry reporting detection switch is not turned on.

[0058] Based on the above embodiment, the vehicle tracking system further includes:

[0059] The first alarm module is configured to determine whether the entry reporting detection switch of the parking area is turned on, and when the judgment result is yes, it warns the entering vehicle and uploads the vehicle feature information and vehicle shooting information, and turns off the entry reporting detection switch; the second alarm module is configured to determine whether the exit reporting detection switch of the parking area is turned on, and when the judgment result is yes, it warns the exiting vehicle and uploads the vehicle feature information and vehicle shooting information; and turns off the exit reporting detection switch.

[0060] Based on the above embodiment, the vehicle tracking system further includes:

[0061] The vacant space detection module is configured to detect vehicle targets in the parking area after detecting an entry alarm target, wherein the vehicle targets include a whole vehicle detection frame, a front or rear vehicle detection frame, and a license plate detection frame; if the whole vehicle detection frame, the front or rear vehicle detection frame, and the license plate detection frame are not within the parking area, the parking area is determined to be in a vacant state, and a vacant space count is increased by 1; when at least one of the whole vehicle detection frame, the front or parking space detection frame, and the license plate detection frame is within the parking area, the vacant space count is reset to zero; if the vacant space count meets a set vacant space threshold, an unreported exit event is reported, and a vehicle detection frame is virtualized; the license plate number of the exiting vehicle corresponding to the exit event is the license plate number of the last entry into the parking area.

[0062] Based on the above embodiment, the vehicle tracking system further includes:

[0063] The information clearing module is configured to clear the entry license plate number recorded in the parking area after a normal exit alarm or a supplementary report of an unreported exit event.

[0064] Based on the above embodiment, the vacancy detection module includes:

[0065] The whole vehicle threshold detection module is configured to confirm that the whole vehicle detection frame is within the parking space if the area ratio of the intersection parking space area of ​​the whole vehicle detection frame and the maximum inscribed rectangle of the parking space area to the whole vehicle detection frame is greater than a first threshold; confirm that the front or rear detection frame is within the parking space if the area ratio of the intersection parking space area of ​​the front or rear detection frame and the maximum inscribed rectangle of the parking space area to the front or rear detection frame is greater than a second threshold; and confirm that the license plate detection frame is within the parking space if there is an intersection parking space area between the license plate detection frame and the parking space area.

[0066] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, can implement the steps provided in the above embodiment. The storage medium may include: a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a non-transitory storage medium, a magnetic disk, or an optical disk, and other media that can store program code.

[0067] The present application also provides a server, as shown in FIG3 , which may include a memory 1 and a processor 2. The memory 1 stores a computer program, and when the processor 2 calls the computer program in the memory, the steps provided in the above embodiment may be implemented. The server may also include various network interfaces, a power supply, and other components.

[0068] The vehicle tracking system, storage medium, and electronic device provided in this application can improve the accuracy of vehicle tracking detection and have the same effect as the above-mentioned vehicle tracking method. The multiple embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from the other embodiments. Similar or identical parts between the multiple embodiments can be referenced to each other. For the system provided in the embodiments, since it corresponds to the method provided in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0069] This article uses individual examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method of this application.

[0070] In this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A vehicle tracking method, comprising: Recording vehicle characteristic information of the target vehicle when the target vehicle enters; Perform multi-frame license plate recognition according to the vehicle feature information; If there is a target license plate number that meets the recognition threshold the earliest, the target license plate number is used as the final license plate number of the target vehicle, and the final license plate number is added as the entry license plate number of the parking space area corresponding to the target vehicle; In the case that the vehicle leaving the parking area cannot be identified, the vehicle leaving the parking area is matched with the vehicle license plate number of the vehicle entering the parking area to determine the vehicle information of the vehicle leaving the parking area.

2. The method according to claim 1, further comprising: In the case where there is a vehicle leaving the parking area, performing a license plate recognition on the vehicle leaving the parking area by using character matching; When the length of consecutive similar characters of the license plate number of the outgoing vehicle and the target license plate number of the target vehicle entering the parking area is less than a preset character length threshold, confirming that the outgoing vehicle and the target vehicle are different vehicles; When the length of continuous similar characters of the license plate number of the outgoing vehicle and the target license plate number of the target vehicle entering the parking area is greater than or equal to the preset character length threshold, it is confirmed that the outgoing vehicle and the target vehicle are the same vehicle.

3. The method according to claim 2, before performing a license plate recognition on a vehicle leaving the parking space area by using character matching, further comprises: When a vehicle enters the parking area, turning on a detection switch for reporting the exit of the parking area; When the exit reporting detection switch is not turned on, an alarm is issued to stop the vehicle exiting the parking area; After the vehicle leaves the parking area, turning on the entry reporting detection switch of the parking area; When the entry reporting detection switch is not turned on, the alarm for vehicles entering the parking area is stopped.

4. The method according to claim 3, further comprising: Determine whether a parking entry reporting detection switch of the parking area is turned on; In response to the entry reporting detection switch of the parking area being turned on, an alarm is issued to the entering vehicle and the vehicle feature information and vehicle shooting information are uploaded, and the entry reporting detection switch is turned off; or, Determine whether the exit reporting detection switch of the parking area is turned on; in response to the exit reporting detection switch of the parking area being turned on, warn the exiting vehicle and upload the vehicle feature information and the vehicle shooting information; turn off the exit reporting detection switch.

5. The method according to claim 1, in the case where the vehicle leaving the parking area cannot be identified and thus the vehicle leaving the parking area is not detected, before matching the vehicle leaving the parking area with the license plate number of the vehicle entering the parking area, further comprising: Detecting vehicle targets in the parking area, wherein the vehicle targets include a whole vehicle detection frame, a front or rear vehicle detection frame, and a license plate detection frame; When the whole vehicle detection frame, the front or rear vehicle detection frame, and the license plate detection frame are not in the parking area, the parking area is determined to be in an empty state, and the empty space count is increased by 1; When at least one of the whole vehicle detection frame, the vehicle head or parking space detection frame and the license plate detection frame is within the parking space area, the vacancy count is reset to zero; When the vacancy count meets the set vacancy threshold, the unreported exit event is reported and a vehicle detection frame is virtualized; wherein the license plate number of the exiting vehicle corresponding to the exit event is the last entry license plate number of the parking space area.

6. The method according to claim 5, further comprising: After a normal exit alarm or a supplementary report of an unreported exit event, the entry license plate number recorded in the parking area is cleared.

7. The method according to claim 5, wherein: The detecting of the vehicle target in the parking area includes: When the area ratio of the intersection parking space area of ​​the maximum inscribed rectangle of the whole vehicle detection frame and the parking space area to the whole vehicle detection frame is greater than a first threshold, confirming that the whole vehicle detection frame is within the parking space area; When the area ratio of the intersection parking space area of ​​the maximum inscribed rectangle of the front or rear detection frame and the parking space area to the front or rear detection frame is greater than a second threshold, confirming that the front or rear detection frame is within the parking space area; In the case where there is an intersection parking area between the license plate detection frame and the parking area, it is confirmed that the license plate detection frame is in the parking area.

8. A vehicle tracking system comprising: A vehicle detection module, configured to record vehicle characteristic information of a target vehicle when the target vehicle enters; A driving-in number detection module is configured to perform multi-frame license plate recognition based on the vehicle feature information; The vehicle alarm module is configured to, when there is a target license plate number that earliest meets the recognition threshold, use the target license plate number as the final license plate number of the target vehicle, and add the final license plate number as the entry license plate number of the parking space area corresponding to the target vehicle; The vehicle matching module is configured to match the outgoing vehicle with the entry license plate number of the parking area when the outgoing vehicle cannot be identified, so as to determine the vehicle information of the outgoing vehicle.

9. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the vehicle tracking method according to any one of claims 1 to 7.

10. A server comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the vehicle tracking method according to any one of claims 1 to 7 when calling the computer program in the memory.

Images