Intelligent settlement method and system for parking lot exit

Abstract

The present invention belongs to the technical field of parking lots. An intelligent settlement method and system for parking lot exit is provided. The method comprises: controlling a first camera to capture first image data of an exit channel area, determining exit vehicle information according to the first image data; judging whether there is a non-exit vehicle among the vehicles in the exit channel area according to the exit vehicle information; if so, using a second camera to capture second image data of a designated area, and identifying the license plate number of each exit vehicle in turn according to the second image data; if not, controlling a radio frequency reader to transmit a radio frequency activation signal to the exit channel area, and receiving feedback signals from each vehicle-mounted radio frequency device, parsing and obtaining the license plate number of each exit vehicle from each feedback signal; and completing the settlement of parking fees according to the license plate number. The present invention can dynamically decide to adopt an image recognition method or a radio frequency recognition method according to actual conditions, and can improve the efficiency of obtaining the license plate number of the exit vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of parking lots, and in particular to an intelligent settlement method and system for parking lot exit. Background Art

[0002] In order to improve the efficiency of parking lot settlement, the exits of parking lots are equipped with automatic license plate recognition devices, which can automatically recognize license plates through image recognition technology, and then realize automatic deduction and settlement. However, the automatic license plate recognition device based on image recognition technology can only recognize the first vehicle in the queue for payment, and other vehicles in the queue for payment still need to undergo license plate recognition and settlement one by one, resulting in low settlement efficiency and congestion at the exit of the parking lot.

[0003] Some parking lots have begun to use radio frequency-based wireless license plate automatic recognition devices, which can activate the ETC equipment and electronic license plate on the vehicle through radio frequency signals to obtain the vehicle's identity and then automatically deduct fees. However, this method is prone to misidentification and misdeduction of fees for vehicles equipped with radio frequency activation devices passing through the parking lot exit, causing more trouble for car owners.

[0004] How to reasonably solve the above technical problems to achieve more efficient parking lot settlement is a technical problem that urgently needs to be solved. Summary of the invention

[0005] In this regard, the present invention provides an intelligent settlement method, system, electronic device, computer storage medium and computer program product for parking lot exit to solve the above technical problems.

[0006] The present invention discloses an intelligent settlement method for parking lot exit, the method comprising the following steps:

[0007] Controlling the first camera to capture first image data of the exit passage area, and determining exit vehicle information according to the first image data; the exit vehicle information includes position information and direction information of each vehicle;

[0008] Determine whether there is a non-exit vehicle in the exit channel area according to the exit vehicle information;

[0009] If yes, use a second camera to capture second image data of a designated area, and identify the license plate number of each exiting vehicle in turn according to the second image data; wherein the designated area is the area located in front of the exit gate;

[0010] If not, the RF reader is controlled to transmit a RF activation signal to the exit channel area, and feedback signals from each vehicle-mounted RF device are received, and the license plate number of each exiting vehicle is obtained by parsing the feedback signals;

[0011] The parking fee is settled according to the license plate number.

[0012] In some embodiments, controlling the first camera to capture first image data of the exit passage area includes:

[0013] The second camera captures third image data of a designated area, and when a license plate number is recognized according to the third image data, controls the microphone configured on the second camera or the first camera to turn on to obtain an audio signal;

[0014] The audio signal is identified and it is determined whether it is a whistle sound, and if so, the first camera is controlled to capture the first image data of the exit channel area.

[0015] In some embodiments, judging whether there is a non-exit vehicle in the vehicles in the exit channel area according to the exit vehicle information includes:

[0016] A waiting vehicle curve is obtained by fitting according to the position information of each vehicle, a deviation angle between each vehicle and the waiting vehicle curve is calculated according to the orientation information of each vehicle, and a sum of the deviation angles is calculated;

[0017] If the sum is lower than the deviation threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area;

[0018] If the sum is higher than the deviation threshold, the position information of each vehicle is constructed as a position feature, and the orientation information of each vehicle is constructed as an orientation feature. The position feature, the orientation feature and the waiting vehicle curve are input into a convolution prediction model. The convolution prediction model outputs the probability that each vehicle is a non-exit vehicle. When any of the probabilities is not higher than the probability threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area. Otherwise, it is determined that there is a non-exit vehicle in the vehicles in the exit channel area.

[0019] In some embodiments, the deviation threshold is calculated as follows:

[0020] The curvature of the waiting vehicle curve is calculated, and the deviation threshold is calculated based on the curvature and a corresponding positive correlation function.

[0021] In some embodiments, the convolution prediction model outputs the probability that each vehicle is a non-exit vehicle, including:

[0022] The convolution prediction model outputs a first probability that each vehicle is a non-exit vehicle;

[0023] Calculating the distance between each vehicle and the exit gate according to the position information, and calculating the probability deviation coefficient according to the distance; wherein the probability deviation coefficient is a value less than or equal to 1;

[0024] The probability deviation coefficient is multiplied by the corresponding first probability to obtain a second probability that each vehicle is a non-exit vehicle.

[0025] In some embodiments, the multiplying the probability deviation coefficient by the corresponding first probability to obtain a second probability that each vehicle is a non-exit vehicle includes:

[0026] Controlling the first camera to capture fourth image data of a designated area, and identifying an edge obstacle area in front of the exit gate from the fourth image data;

[0027] Identify each first vehicle whose body is completely in the edge obstacle area according to the position information of each vehicle, set the probability deviation coefficient of each first vehicle to 1, and multiply the first probability of each first vehicle by the probability deviation coefficient to obtain a second probability that each first vehicle is a non-exit vehicle;

[0028] According to the position information of each vehicle, the second vehicle whose body part is in the edge obstacle area and the number of vehicles behind the second vehicle are identified, the probability deviation coefficient of the second vehicle is set to the first probability deviation coefficient, and a correction coefficient is determined according to the number of vehicles. The correction coefficient is multiplied by the first probability deviation coefficient to obtain the corresponding second probability deviation coefficient; the second probability deviation coefficient is then multiplied by the first probability of the second vehicle to obtain a second probability that the second vehicle is a non-exit vehicle.

[0029] The present invention also discloses an intelligent settlement system for parking lot exit, the system comprising a processing module and a storage module; the storage module stores a computer program; the processing module calls and executes the computer program stored in the storage module to implement any of the aforementioned methods.

[0030] The present invention also discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement any of the above methods.

[0031] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0032] The present invention also discloses a computer program product, which implements any of the above methods when the computer program product is run on a terminal.

[0033] The present invention decides whether to adopt an image recognition method or a radio frequency identification method according to whether there is a non-exit vehicle in the vehicles in the exit channel area, which can improve the efficiency of obtaining the license plate number of the exit vehicle and avoid the defect that the radio frequency identification method easily misidentifies the non-exit vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a flow chart of an intelligent settlement method for parking lot exit disclosed in an embodiment of the present invention;

[0036] Figure 2 is a schematic diagram of a vehicle exit state disclosed in an embodiment of the present invention;

[0037] Figure 3 The invention discloses a structural diagram of an intelligent settlement system for parking lot exit. DETAILED DESCRIPTION

[0038] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0039] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0040] like Figure 1 As shown, the embodiment of the present invention discloses an intelligent settlement method for parking lot exit, the method comprising the following steps:

[0041] Controlling the first camera to capture first image data of the exit passage area, and determining exit vehicle information according to the first image data; the exit vehicle information includes position information and direction information of each vehicle;

[0042] Determine whether there is a non-exit vehicle in the exit channel area according to the exit vehicle information;

[0043] If yes, use a second camera to capture second image data of a designated area, and identify the license plate number of each exiting vehicle in turn according to the second image data; wherein the designated area is the area located in front of the exit gate;

[0044] If not, the RF reader is controlled to transmit a RF activation signal to the exit channel area, and feedback signals from each vehicle-mounted RF device are received, and the license plate number of each exiting vehicle is obtained by parsing the feedback signals;

[0045] The parking fee is settled according to the license plate number.

[0046] The present invention arranges a camera and a radio frequency reader in the exit area of ​​the parking lot, so that a suitable identification method of the license plate number of the exiting vehicle can be selected according to the actual situation, thereby realizing rapid exit settlement. Specifically, a first camera is set to aim at the exit channel area, which is the area in front of the exit gate, and the vehicles to be exited are queued here and exit after verification and settlement by the gate; the position information and orientation information of each vehicle in the exit channel area can be extracted from the first image data captured by the first camera, and it can be judged whether there are non-exiting vehicles in the vehicles in the exit channel area. If there are no non-exiting vehicles, the license plate numbers of all vehicles in the exit channel area can be obtained at one time by radio frequency; if there are non-exiting vehicles, it is not appropriate to use the radio frequency method, but to use the conventional image recognition method to obtain the license plate numbers of the vehicles arriving in front of the exit gate one by one. In this process, the first camera can also be used to continue to monitor whether there are non-exiting vehicles in the vehicles in the exit channel area. If it is judged that there are no non-exiting vehicles, it can be switched to the radio frequency method.

[0047] Therefore, the present invention decides whether to adopt image recognition or radio frequency identification according to whether there is a non-exit vehicle in the vehicles in the exit channel area, which can improve the efficiency of obtaining the license plate number of the exit vehicle and avoid the defect of radio frequency identification that it is easy to misidentify non-exit vehicles.

[0048] It should be noted that the parking lot management system stores the license plate number of each vehicle and the corresponding deduction account, or the third-party system stores the license plate number of each vehicle and the corresponding deduction account, and the parking lot management system can obtain these data by accessing the third-party system. After identifying the license plate number, the corresponding parking fee can be deducted from the corresponding deduction account.

[0049] In some embodiments, controlling the first camera to capture first image data of the exit passage area includes:

[0050] The second camera captures third image data of a designated area, and when a license plate number is recognized according to the third image data, controls the microphone configured on the second camera or the first camera to turn on to obtain an audio signal;

[0051] The audio signal is identified and it is determined whether it is a whistle sound, and if so, the first camera is controlled to capture the first image data of the exit channel area.

[0052] In this embodiment, the radio frequency method has the defect of misidentification. It can only be used when there are many vehicles waiting to exit and there are no non-exiting vehicles mixed in. In this way, it can also achieve the effect of efficient settlement. In this regard, the present invention is set to capture the third image data of the designated area with the second camera. When the license plate number is recognized, it indicates that a vehicle is about to exit; at the same time, the microphone on the first camera or the second camera is controlled to start to pick up the sound, and analyze whether there is a vehicle horn in the picked up audio signal. If so, it indicates that there is exit congestion at the exit of the parking lot. When there is exit congestion, the first camera is controlled to capture the first image data of the exit channel area, thereby starting the above technical solution of the present invention.

[0053] Such an arrangement can, on the one hand, reduce the probability of misidentification of the radio frequency method, and on the other hand, reduce the load of the first camera's shooting and the corresponding image analysis processing.

[0054] In some embodiments, judging whether there is a non-exit vehicle in the vehicles in the exit channel area according to the exit vehicle information includes:

[0055] A waiting vehicle curve is obtained by fitting according to the position information of each vehicle, a deviation angle between each vehicle and the waiting vehicle curve is calculated according to the orientation information of each vehicle, and a sum of the deviation angles is calculated;

[0056] If the sum is lower than the deviation threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area;

[0057] If the sum is higher than the deviation threshold, the position information of each vehicle is constructed as a position feature, and the orientation information of each vehicle is constructed as an orientation feature. The position feature, the orientation feature and the waiting vehicle curve are input into a convolution prediction model. The convolution prediction model outputs the probability that each vehicle is a non-exit vehicle. When any of the probabilities is not higher than the probability threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area. Otherwise, it is determined that there is a non-exit vehicle in the vehicles in the exit channel area.

[0058] In this embodiment, Figure 2As shown, the present invention first fits the waiting vehicle curve based on the position information of each vehicle in the exit channel area, and then calculates the deviation angle between the direction of each vehicle and the waiting vehicle curve (that is, the angle between the vehicle direction and the tangent line of the waiting vehicle curve at that point), and calculates the sum of the deviation angles of all vehicles. If the sum is lower than the deviation threshold, it means that the directions of all vehicles are along the waiting vehicle curve, and at this time, it can be determined that all vehicles are exit vehicles that will leave through the parking lot exit. If the sum is higher than the deviation threshold, it means that there are vehicles (for example Figure 2 The direction of vehicle 5) in the waiting vehicle curve deviates greatly from that of the waiting vehicle curve. The vehicle may be trying to pass through the exit vehicle team but temporarily fails to leave the team, that is, a non-exit vehicle. At this time, the convolution prediction model needs to be further used for more in-depth analysis. Specifically, the position information of all vehicles is constructed as position features, and the direction information is constructed as direction features. Then, the position features, direction features and the waiting vehicle curve are input into the convolution prediction model. The convolution prediction model can predict the probability of each vehicle being a non-exit vehicle based on this. When the above probability of at least one vehicle is higher than the probability threshold, it means that there is a non-exit vehicle in the exit channel area, and the radio frequency mode should not be turned on at this time.

[0059] The convolution prediction model is preferably constructed based on Convnext-yolox, which includes Convnext-yolo5 and Convnext-yolo6, which are not specifically limited. Convnext-yolox mainly includes a backbone network, a feature extraction network, and a detection network. The specific construction and training details of the convolution prediction model are not repeated in the present invention.

[0060] In some embodiments, the deviation threshold is calculated as follows:

[0061] The curvature of the waiting vehicle curve is calculated, and the deviation threshold is calculated based on the curvature and a corresponding positive correlation function.

[0062] In this embodiment, after fitting the waiting vehicle curve, the curvature of the curve is calculated by integration. When the curvature of the waiting vehicle curve is larger, the vehicle is more likely to deviate from the waiting vehicle curve, and when the curvature of the waiting vehicle curve is smaller (for example, a straight line), the vehicle to be exited is basically parallel to the waiting vehicle curve, and the deviation is very small. In this regard, the present invention sets the relationship between the deviation threshold and the curvature of the waiting vehicle curve to be positively correlated, thereby reducing the probability of misidentification. The present invention does not limit the specific form of the above positive correlation function.

[0063] The arc of the waiting vehicle curve is calculated as follows: ,in, is the radius of curvature at a point on the waiting vehicle curve, is an infinitesimal arc length element on the curve, and a and b are the parameter values ​​of the starting and ending points of the waiting vehicle curve.

[0064] Radius of curvature The calculation method is: .

[0065] is the curvature at a certain point on the waiting vehicle curve, and the calculation formula is as follows: ;in: is the parameterized expression of the curve, is a parameter.

[0066] In addition, the curvature of the waiting vehicle curve can also be calculated using existing software tools.

[0067] In some embodiments, the convolution prediction model outputs the probability that each vehicle is a non-exit vehicle, including:

[0068] The convolution prediction model outputs a first probability that each vehicle is a non-exit vehicle;

[0069] Calculating the distance between each vehicle and the exit gate according to the position information, and calculating the probability deviation coefficient according to the distance; wherein the probability deviation coefficient is a value less than or equal to 1;

[0070] The probability deviation coefficient is multiplied by the corresponding first probability to obtain a second probability that each vehicle is a non-exit vehicle.

[0071] In this embodiment, the convolution prediction model can predict the probability that a vehicle is a non-exit vehicle based on the degree of fit between the vehicle in the exit channel area and the convoy formation, but the deviation angle between the vehicle direction and the waiting vehicle curve may be too large due to various factors, such as when the vehicle has just entered the exit channel area, the vehicle direction has not been adjusted, and the vehicle changes the vehicle direction to observe the situation ahead. To avoid misjudgment, the present invention further determines the probability deviation coefficient based on the distance between each vehicle and the exit gate, and uses the probability deviation coefficient to adjust the first probability of each vehicle output by the convolution prediction model to a more accurate second probability.

[0072] Among them, the probability deviation coefficient should be positively correlated with the above distance. When the vehicle is closer to the exit gate, that is, the smaller the above distance is, the smaller the probability that the vehicle is a non-exit vehicle is, the smaller the probability deviation coefficient is set to 1.0; when the vehicle is farther from the exit gate, that is, the greater the above distance is, the greater the probability that the vehicle is a non-exit vehicle is, the greater the probability deviation coefficient is set to 1.2.

[0073] In some embodiments, the multiplying the probability deviation coefficient by the corresponding first probability to obtain a second probability that each vehicle is a non-exit vehicle includes:

[0074] Controlling the first camera to capture fourth image data of a designated area, and identifying an edge obstacle area in front of the exit gate from the fourth image data;

[0075] Identify each first vehicle whose body is completely in the edge obstacle area according to the position information of each vehicle, set the probability deviation coefficient of each first vehicle to 1, and multiply the first probability of each first vehicle by the probability deviation coefficient to obtain a second probability that each first vehicle is a non-exit vehicle;

[0076] According to the position information of each vehicle, the second vehicle whose body part is in the edge obstacle area and the number of vehicles behind the second vehicle are identified, the probability deviation coefficient of the second vehicle is set to the first probability deviation coefficient, and a correction coefficient is determined according to the number of vehicles. The correction coefficient is multiplied by the first probability deviation coefficient to obtain the corresponding second probability deviation coefficient; the second probability deviation coefficient is then multiplied by the first probability of the second vehicle to obtain a second probability that the second vehicle is a non-exit vehicle.

[0077] In this embodiment, the vehicles waiting to exit the exit channel at the front section can be further divided into two types: first vehicles ( Figure 2 Vehicles 1 and 2), and the second vehicle whose body part is in the edge obstacle area Figure 2 Vehicles in 3).

[0078] For the first vehicle (whose number is determined by the length of the edge obstacle area, the longer the length, the greater the number of corresponding first vehicles), since it is completely in the edge obstacle area, it is most difficult to return to the parking lot and not leave the parking lot. At this time, its probability deviation coefficient is set to 1, that is, the prediction result of the convolution prediction model is directly used. For the second vehicle, since its body is only partially in the edge obstacle area, it is less difficult to return to the parking lot and not leave the parking lot. At this time, the correction coefficient is determined according to the number of vehicles behind the second vehicle, and the correction coefficient is multiplied by the first probability deviation coefficient (that is, the probability deviation coefficient calculated according to the aforementioned distance) to obtain the corresponding second probability deviation coefficient, and then the second probability deviation coefficient is multiplied by the first probability predicted by the convolution prediction model to obtain the second probability that the second vehicle is a non-exit vehicle.

[0079] For vehicles other than the first vehicle and the second vehicle, the probability deviation coefficient is calculated based on the above distance. The specific calculation method refers to the above content and will not be repeated here.

[0080] like Figure 3As shown, an embodiment of the present invention also discloses an intelligent settlement system for parking lot exit, the system comprising a processing module and a storage module; a computer program is stored in the storage module; the processing module calls and executes the computer program stored in the storage module to implement the method described in the aforementioned embodiment.

[0081] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method described in the above embodiment.

[0082] An embodiment of the present invention further discloses a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the above embodiment.

[0083] The embodiment of the present invention further discloses a computer program product. When the computer program product is run on a terminal, the method described in the above embodiment is implemented.

[0084] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0085] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0086] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0087] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0088] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0089] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An intelligent settlement method for parking lot exit, characterized by: The method comprises the following steps: Controlling the first camera to capture first image data of the exit passage area, and determining exit vehicle information according to the first image data; the exit vehicle information includes position information and direction information of each vehicle; Determine whether there is a non-exit vehicle in the exit channel area according to the exit vehicle information; If yes, use a second camera to capture second image data of a designated area, and identify the license plate number of each exiting vehicle in turn according to the second image data; wherein the designated area is the area located in front of the exit gate; If not, the RF reader is controlled to transmit a RF activation signal to the exit channel area, and feedback signals from each vehicle-mounted RF device are received, and the license plate number of each exiting vehicle is obtained by parsing the feedback signals; The parking fee is settled according to the license plate number.

2. The intelligent settlement method for parking lot exit according to claim 1, characterized in that: Controlling the first camera to capture first image data of the exit passage area includes: The second camera captures third image data of a designated area, and when a license plate number is recognized according to the third image data, controls the microphone configured on the second camera or the first camera to turn on to obtain an audio signal; The audio signal is identified and it is determined whether it is a whistle sound, and if so, the first camera is controlled to capture the first image data of the exit channel area.

3. The intelligent settlement method for parking lot exit according to claim 1, characterized in that: Judging whether there is a non-exit vehicle in the vehicles in the exit channel area according to the exit vehicle information includes: A waiting vehicle curve is obtained by fitting according to the position information of each vehicle, a deviation angle between each vehicle and the waiting vehicle curve is calculated according to the orientation information of each vehicle, and a sum of the deviation angles is calculated; If the sum is lower than the deviation threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area; If the sum is higher than the deviation threshold, the position information of each vehicle is constructed as a position feature, and the orientation information of each vehicle is constructed as an orientation feature. The position feature, the orientation feature and the waiting vehicle curve are input into a convolution prediction model. The convolution prediction model outputs the probability that each vehicle is a non-exit vehicle. When any of the probabilities is not higher than the probability threshold, it is determined that there is no non-exit vehicle in the vehicles in the exit channel area. Otherwise, it is determined that there is a non-exit vehicle in the vehicles in the exit channel area.

4. The intelligent settlement method for parking lot exit according to claim 3 is characterized by: The deviation threshold is calculated as follows: The curvature of the waiting vehicle curve is calculated, and the deviation threshold is calculated based on the curvature and a corresponding positive correlation function.

5. The intelligent settlement method for parking lot exit according to claim 3 is characterized by: The convolution prediction model outputs the probability that each vehicle is a non-exit vehicle, including: The convolution prediction model outputs a first probability that each vehicle is a non-exit vehicle; Calculating the distance between each vehicle and the exit gate according to the position information, and calculating the probability deviation coefficient according to the distance; wherein the probability deviation coefficient is a value less than or equal to 1; The probability deviation coefficient is multiplied by the corresponding first probability to obtain a second probability that each vehicle is a non-exit vehicle.

6. The intelligent settlement method for parking lot exit according to claim 5, characterized in that: The second probability of each vehicle being a non-exit vehicle is obtained by multiplying the probability deviation coefficient with the corresponding first probability, including: Controlling the first camera to capture fourth image data of a designated area, and identifying an edge obstacle area in front of the exit gate from the fourth image data; Identify each first vehicle whose body is completely in the edge obstacle area according to the position information of each vehicle, set the probability deviation coefficient of each first vehicle to 1, and multiply the first probability of each first vehicle by the probability deviation coefficient to obtain a second probability that each first vehicle is a non-exit vehicle; According to the position information of each vehicle, the second vehicle whose body part is in the edge obstacle area and the number of vehicles behind the second vehicle are identified, the probability deviation coefficient of the second vehicle is set to the first probability deviation coefficient, and a correction coefficient is determined according to the number of vehicles. The correction coefficient is multiplied by the first probability deviation coefficient to obtain the corresponding second probability deviation coefficient; the second probability deviation coefficient is then multiplied by the first probability of the second vehicle to obtain a second probability that the second vehicle is a non-exit vehicle.

7. An intelligent settlement system for parking lot exit, the system comprising a processing module and a storage module; the storage module stores a computer program; the characteristics are: The processing module calls and executes the computer program stored in the storage module to implement the method according to any one of claims 1 to 6.

8. An electronic device comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 6.

9. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that: When the computer program product runs on a terminal, the method according to any one of claims 1 to 6 is implemented.

Images