Parking management system based on edge calculation and video monitoring

Abstract

The invention discloses a parking management system based on edge calculation and video monitoring, and belongs to the technical field of parking management, and the system specifically comprises the steps: obtaining a parking region, a lane exit position and vehicle parking time, calibrating a vehicle which is not in the parking region as an undetermined vehicle, and dividing a target vehicle and a non-target vehicle according to the parking time; selecting any target vehicle and acquiring a time node h when the target vehicle leaves the parking area for the first time, extracting a monitoring image sequence of a lane exit in a past time period T, identifying the number of vehicles and drawing a curve Z, and calculating an average vehicle moving speed and a congestion coefficient of the lane exit; and constructing alternative routes from the target vehicle to all lane exits, calculating a distance value of any alternative route and combining the congestion coefficient to obtain an evaluation score, and selecting the lane exit corresponding to the alternative route with the minimum evaluation score as a driving-out exit of the target vehicle, thereby improving the driving-out efficiency of the vehicle in the target area.

Description

Technical Field

[0001] The present invention relates to the technical field of parking management, and in particular to a parking management system based on edge computing and video surveillance. Background Art

[0002] The smart parking system, also known as the smart parking management system, uses cameras to collect vehicle license plate information for identification to record information about vehicles entering and leaving the parking lot, and records the parking positions of different vehicles, and then counts the number of remaining parking spaces in each parking road area. At this time, when a new vehicle enters the parking lot, the number of remaining parking spaces in different driving directions is displayed on the LED display, and arrows are used to guide the vehicle into the vacant parking space, which is extremely convenient.

[0003] Since the traffic entering the parking lot is characterized by gradualness and aggregation, that is, the number of vehicles gradually increases and they are concentrated in a specific period of time, and the exit efficiency of the parking lot is usually closely related to the congestion level of the access road, the exit efficiency of the parking lot will be affected by the congestion of the external roads. This influence may cause congestion when vehicles queue up inside the parking lot waiting to leave. For example, during rush hours, due to the inevitable large number of vehicles leaving the parking lot, users tend to choose exits based on their travel habits. However, due to insufficient understanding of the queue conditions at each exit and the congestion of the external roads, they may choose exits with more queued vehicles and congested roads, which further aggravates the congestion, resulting in lower efficiency of users' vehicle travel, increased travel time, and increased parking costs for users. Summary of the invention

[0004] The purpose of the present invention is to provide a parking management system based on edge computing and video surveillance to solve the following technical problems: Since the traffic entering the parking lot is characterized by gradualness and aggregation, that is, the number of vehicles gradually increases and they are concentrated in a specific period of time, and the exit efficiency of the parking lot is usually closely related to the congestion level of the access road, the exit efficiency of the parking lot will be affected by the congestion of the external roads. This influence may cause congestion when vehicles queue up inside the parking lot waiting to leave. For example, during rush hours, due to the inevitable large number of vehicles leaving the parking lot, users tend to choose exits based on their travel habits. However, due to insufficient understanding of the queue conditions at each exit and the congestion of the external roads, they may choose exits with more queued vehicles and congested roads, which further aggravates the congestion, resulting in lower efficiency of users' vehicle travel, increased travel time, and increased parking costs for users.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A parking management system based on edge computing and video surveillance, comprising: A data acquisition module is used to obtain all parking areas, lane exit positions and the dwelling time of all vehicles in the target area, mark all vehicles that are not currently in the parking area as pending vehicles, and divide the pending vehicles according to the dwelling time T to obtain target vehicles and non-target vehicles; A data analysis module is used to select any target vehicle and obtain the time node h when the target vehicle first leaves the parking area, take the time node h as the starting point, extract the monitoring image of the exit position of any lane in the target area in the past time period T and generate a monitoring image sequence, identify the number of vehicles at the exit position of the lane at any time according to the monitoring image sequence, draw a curve Z of the change of the number of vehicles over time, obtain the average vehicle moving speed of any target lane in the past time period T, and calculate the congestion coefficient C of any lane exit by combining the average vehicle moving speed and the curve Z; The result generation module is used to construct alternative routes from the target vehicle to all lane exits and mark them as S1, S2, ..., Sn, where n is the number of lane exits in the target area, calculate the distance value from any target vehicle to the vehicle exit corresponding to the alternative route, and calculate the evaluation score of any alternative route in combination with the congestion coefficient of the lane exit, and select the lane exit corresponding to the alternative route with the smallest evaluation score as the exit of the target vehicle.

[0006] As a further solution of the present invention: in the data acquisition module, the process of dividing the vehicles to be determined is: If the residence time H of any pending vehicle is greater than or equal to the preset threshold, the pending vehicle is marked as a target vehicle; if the residence time H of any pending vehicle is less than the preset threshold, the pending vehicle is marked as a non-target vehicle.

[0007] As a further solution of the present invention: in the data analysis module, the specific calculation process of the average vehicle moving speed at the lane exit is: S11, grayscale processing is performed on the monitoring images in sequence to obtain monitoring grayscale images, a rectangular coordinate system is established with the central pixel of the monitoring grayscale image as the origin, the coordinates (x, y) of all pixels in the monitoring grayscale image are generated, and the grayscale values ​​Pi (x, y) of all pixels in the monitoring grayscale image are identified, i = 1, ..., m, i represents the order of the monitoring grayscale image where the pixel is located in the sequence, and the grayscale difference d of the pixel at the same position between adjacent monitoring grayscale images is calculated by the grayscale difference formula d = |Pi+1 (x, y)-Pi (x, y)|, if d is greater than a preset threshold, it indicates that the corresponding pixel is a vehicle pixel, if d is less than or equal to the preset threshold, it indicates that the corresponding pixel is a road area pixel; S12, merge adjacent vehicle pixels, and circumscribe a minimum circumscribed rectangle around the merged vehicle pixel area to obtain a pixel rectangle of any vehicle in the monitoring grayscale image, extract the pixel rectangle corresponding to any target vehicle, select the midpoint pixel of the rectangle from the pixel rectangle and mark it as a reference point, select the position information of the reference point in each frame of the monitoring image in the monitoring image sequence, and calculate the average speed of the target vehicle in each frame of the monitoring image according to the position change between adjacent frames and mark it as V, take the average speed V as the speed of the target vehicle, and calculate according to the calculation formula , where N is the total number of vehicles and Vi is the average speed of any vehicle; the average vehicle moving speed at the vehicle exit is calculated.

[0008] As a further solution of the present invention: in the data analysis module, the specific calculation process of the crowding coefficient is: ; where t∈[hT,h].

[0009] As a further solution of the present invention: in the result generation module, the specific generation process of the alternative routes is: The position information of the target vehicle is obtained and the Dijkstra algorithm is used to generate the shortest path from the target vehicle to all lane exit positions in the target area, and the shortest path is marked as an alternative route.

[0010] As a further solution of the present invention: in the result generation module, the specific calculation formula of the evaluation score is: ; ; Wherein, D is the route length of any alternative route, α is the preset correction coefficient, li represents the projection length of any vehicle at the lane exit in the road direction, and L is the total projection length of all vehicles at any lane exit in the road direction.

[0011] As a further solution of the present invention: in the result generation module, if there are two or more alternative routes with the same evaluation scores and both are minimum values, the congestion coefficient of the road exit corresponding to each alternative route is obtained, and the lane exit corresponding to the smallest congestion coefficient is used as the exit for the target vehicle.

[0012] As a further solution of the present invention: it also includes collecting feedback data of each target vehicle; if the target vehicle confirms to select the planned route, the driving state information of the target vehicle is updated, and the lane exit of the target vehicle without a planned route is dynamically planned and adjusted.

[0013] Beneficial effects of the present invention: In actual situations, users will select corresponding vehicle exits according to their own habits. In this case, the vehicle exit selected by the user may have a large queue of vehicles, resulting in low efficiency of the user's vehicle exiting and parking. Therefore, in this case, the data analysis module selects any target vehicle and obtains the time node when the target vehicle first leaves the parking area. Taking the time node as the starting point, the monitoring image of any lane exit position of the target area in the past time period T is extracted and a monitoring image sequence is generated. The number of vehicles at the lane exit position at any time is identified according to the monitoring image sequence, and a curve Z of the number of vehicles changing with time is drawn. The monitoring image sequence of any target vehicle in the past time period T is obtained. The average vehicle moving speed of the marked lane is combined with the average vehicle moving speed and curve Z to calculate the congestion coefficient of any lane exit, which can effectively analyze the congestion that may be encountered at any lane exit in the target area, provide a basis for subsequent result generation, construct alternative routes from the target vehicle to all lane exits, calculate the distance value from any target vehicle to the vehicle exit corresponding to the alternative route, and calculate the evaluation score of any alternative route in combination with the congestion coefficient of the lane exit, and select the lane exit corresponding to the alternative route with the smallest evaluation score as the exit of the target vehicle, thereby determining the corresponding vehicle exit for each user, and improving the efficiency of vehicle exit in the target area. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below in conjunction with the accompanying drawings.

[0015] Figure 1 It is a flow chart of a parking management system based on edge computing and video surveillance of the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] See also Figure 1 As shown, the present invention is a parking management system based on edge computing and video surveillance, comprising: A data acquisition module is used to obtain all parking areas, lane exit positions and the dwelling time of all vehicles in the target area, mark all vehicles that are not currently in the parking area as pending vehicles, and divide the pending vehicles according to the dwelling time T to obtain target vehicles and non-target vehicles; A data analysis module is used to select any target vehicle and obtain the time node h when the target vehicle first leaves the parking area, take the time node h as the starting point, extract the monitoring image of the exit position of any lane in the target area in the past time period T and generate a monitoring image sequence, identify the number of vehicles at the exit position of the lane at any time according to the monitoring image sequence, draw a curve Z of the change of the number of vehicles over time, obtain the average vehicle moving speed of any target lane in the past time period T, and calculate the congestion coefficient C of any lane exit by combining the average vehicle moving speed and the curve Z; The result generation module is used to construct alternative routes from the target vehicle to all lane exits and mark them as S1, S2, ..., Sn, where n is the number of lane exits in the target area, calculate the distance value from any target vehicle to the vehicle exit corresponding to the alternative route, and calculate the evaluation score of any alternative route in combination with the congestion coefficient of the lane exit, and select the lane exit corresponding to the alternative route with the smallest evaluation score as the exit of the target vehicle.

[0018] In actual situations, users will select the corresponding vehicle exit according to their own habits. In this case, the vehicle exit selected by the user may have a large queue of vehicles, resulting in low efficiency in the user's vehicle exiting and parking. Therefore, in this case, by selecting any target vehicle and obtaining the time node when the target vehicle first leaves the parking area, taking the time node as the starting point, extracting the monitoring image of the exit position of any lane in the target area within the past time period T and generating a monitoring image sequence, identifying the number of vehicles at the exit position of the lane at any time according to the monitoring image sequence, and drawing a curve Z of the change of the number of vehicles over time, the number of vehicles at any target lane in the past time period T is obtained. The average vehicle moving speed is combined with the average vehicle moving speed and the curve to calculate the congestion coefficient of any lane exit, which can effectively analyze the congestion that the target vehicle may encounter during the exit process, provide a basis for subsequent result generation, construct alternative routes from the target vehicle to all lane exits, and calculate the distance value from any target vehicle to the vehicle exit corresponding to the alternative route. The evaluation score of any alternative route is calculated in combination with the congestion coefficient of the lane exit, and the lane exit corresponding to the alternative route with the smallest evaluation score is selected as the exit of the target vehicle, thereby determining the corresponding vehicle exit for each user, and improving the efficiency of vehicle exit in the target area.

[0019] In a preferred situation of this embodiment, in the data acquisition module, the process of dividing the pending vehicles is: If the residence time H of any pending vehicle is greater than or equal to the preset threshold, the pending vehicle is marked as a target vehicle; if the residence time H of any pending vehicle is less than the preset threshold, the pending vehicle is marked as a non-target vehicle.

[0020] In another preferred situation of this embodiment, in the data analysis module, the specific calculation process of the average vehicle moving speed at the lane exit is: S11, grayscale processing is performed on the monitoring images in sequence to obtain monitoring grayscale images, a rectangular coordinate system is established with the central pixel of the monitoring grayscale image as the origin, the coordinates (x, y) of all pixels in the monitoring grayscale image are generated, and the grayscale values ​​Pi (x, y) of all pixels in the monitoring grayscale image are identified, i = 1, ..., m, i represents the order of the monitoring grayscale image where the pixel is located in the sequence, and the grayscale difference d of the pixel at the same position between adjacent monitoring grayscale images is calculated by the grayscale difference formula d = |Pi+1 (x, y)-Pi (x, y)|. If d is greater than a preset threshold, it indicates that the corresponding pixel is a vehicle pixel. If d is less than or equal to the preset threshold, it indicates that the corresponding pixel is a road surface area pixel. S12, merge adjacent vehicle pixels, and circumscribe a minimum circumscribed rectangle around the merged vehicle pixel area to obtain a pixel rectangle of any vehicle in the monitoring grayscale image, extract the pixel rectangle corresponding to any target vehicle, select the midpoint pixel of the rectangle from the pixel rectangle and mark it as a reference point, select the position information of the reference point in each frame of the monitoring image in the monitoring image sequence, and calculate the average speed of the target vehicle in each frame of the monitoring image according to the position change between adjacent frames and mark it as V, take the average speed V as the speed of the target vehicle, and calculate according to the calculation formula , where N is the total number of vehicles and Vi is the average speed of any vehicle; the average vehicle moving speed at the vehicle exit is calculated.

[0021] In another preferred embodiment of the present invention, in the data analysis module, the specific calculation process of the crowding coefficient is as follows: ; where t∈[hT,h].

[0022] In another preferred situation of this embodiment, in the result generation module, the specific generation process of the alternative routes is: The position information of the target vehicle is obtained, and the Dijkstra algorithm is used to generate the shortest path from the current position to the exit position of each lane, and the shortest path is marked as an alternative route.

[0023] In another preferred embodiment of the present invention, in the result generation module, the specific calculation formula of the evaluation score is: ; ; Wherein, D is the route length of any alternative route, α is the preset correction coefficient, li represents the projection length of any vehicle at the lane exit in the road direction, and L is the total projection length of all vehicles at any lane exit in the road direction.

[0024] In another preferred situation of the present embodiment, in the result generation module, if there are two or more alternative routes with the same evaluation scores and both are minimum values, the congestion coefficient of the road exit corresponding to each alternative route is obtained, and the lane exit corresponding to the smallest congestion coefficient is used as the exit for the target vehicle.

[0025] Because when the evaluation scores of multiple alternative routes are the same and are all minimum values, by comparing the congestion coefficients of the corresponding road exits, the least congested route can be selected, thereby improving traffic efficiency and reducing waiting time, which helps to optimize the driving experience of the target vehicle.

[0026] In another preferred situation of the present embodiment, feedback data of each target vehicle is collected; if the target vehicle confirms to select the planned route, the driving status information of the target vehicle is updated, and the lane exit is dynamically planned and adjusted for the target vehicle with an unplanned route.

[0027] It can be understood that when the current target vehicle determines the planned route, the congestion coefficient of the lane exit corresponding to the planned route will be affected by the current target vehicle selection, so as to dynamically adjust the lane exit for the target vehicle with an unplanned route. The specific process can be understood that when planning for the target vehicle with an unplanned route, it is necessary to consider a target vehicle in the target area that selects a lane entrance but does not reach the lane exit. Therefore, the calculation formula for the evaluation score of the subsequent target vehicle is modified to ; Where X represents the number of target vehicles that have selected the planned route and correspond to a specific lane exit in the feedback data.

[0028] 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 parking management system based on edge computing and video surveillance, characterized in that: include: A data acquisition module is used to obtain all parking areas, lane exit positions and the dwelling time of all vehicles in the target area, mark all vehicles that are not currently in the parking area as pending vehicles, and divide the pending vehicles according to the dwelling time T to obtain target vehicles and non-target vehicles; A data analysis module is used to select any target vehicle and obtain the time node h when the target vehicle first leaves the parking area, take the time node h as the starting point, extract the monitoring image of the exit position of any lane in the target area in the past time period T and generate a monitoring image sequence, identify the number of vehicles at the exit position of the lane at any time according to the monitoring image sequence, draw a curve Z of the change of the number of vehicles over time, obtain the average vehicle moving speed of any target lane in the past time period T, and calculate the congestion coefficient C of any lane exit by combining the average vehicle moving speed and the curve Z; The result generation module is used to construct alternative routes from the target vehicle to all lane exits and mark them as S1, S2, ..., Sn, where n is the number of lane exits in the target area, calculate the distance value from any target vehicle to the vehicle exit corresponding to the alternative route, and calculate the evaluation score of any alternative route in combination with the congestion coefficient of the lane exit, and select the lane exit corresponding to the alternative route with the smallest evaluation score as the exit of the target vehicle.

2. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the data acquisition module, the process of dividing the pending vehicles is as follows: If the residence time H of any pending vehicle is greater than or equal to the preset threshold, the pending vehicle is marked as a target vehicle; if the residence time H of any pending vehicle is less than the preset threshold, the pending vehicle is marked as a non-target vehicle.

3. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the data analysis module, the specific calculation process of the average vehicle moving speed at the lane exit is: S11, grayscale processing is performed on the monitoring images in sequence to obtain monitoring grayscale images, a rectangular coordinate system is established with the central pixel of the monitoring grayscale image as the origin, the coordinates (x, y) of all pixels in the monitoring grayscale image are generated, and the grayscale values ​​Pi (x, y) of all pixels in the monitoring grayscale image are identified, i = 1, ..., m, i represents the order of the monitoring grayscale image where the pixel is located in the sequence, and the grayscale difference d of the pixel at the same position between adjacent monitoring grayscale images is calculated by the grayscale difference formula d = |Pi+1 (x, y)-Pi (x, y)|, if d is greater than a preset threshold, it indicates that the corresponding pixel is a vehicle pixel, if d is less than or equal to the preset threshold, it indicates that the corresponding pixel is a road area pixel; S12, merge adjacent vehicle pixels, and circumscribe a minimum circumscribed rectangle around the merged vehicle pixel area to obtain a pixel rectangle of any vehicle in the monitoring grayscale image, extract the pixel rectangle corresponding to any target vehicle, select the midpoint pixel of the rectangle from the pixel rectangle and mark it as a reference point, select the position information of the reference point in each frame of the monitoring image in the monitoring image sequence, and calculate the average speed of the target vehicle in each frame of the monitoring image according to the position change between adjacent frames and mark it as V, take the average speed V as the speed of the target vehicle, and calculate according to the calculation formula , where N is the total number of vehicles and Vi is the average speed of any vehicle; the average vehicle moving speed at the vehicle exit is calculated.

4. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the data analysis module, the specific calculation process of the crowding coefficient is: , where t∈[hT,h].

5. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the result generation module, the specific generation process of the alternative routes is as follows: The position information of the target vehicle is obtained and the Dijkstra algorithm is used to generate the shortest path from the target vehicle to all lane exit positions in the target area, and the shortest path is marked as an alternative route.

6. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the result generation module, the specific calculation formula of the evaluation score is: ; ; Wherein, D is the route length of any alternative route, α is the preset correction coefficient, li represents the projection length of any vehicle at the lane exit in the road direction, and L is the total projection length of all vehicles at any lane exit in the road direction.

7. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: In the result generation module, if there are two or more alternative routes with the same evaluation scores and both are minimum values, the congestion coefficient of the road exit corresponding to each alternative route is obtained, and the lane exit corresponding to the minimum congestion coefficient is used as the exit of the target vehicle.

8. A parking management system based on edge computing and video surveillance according to claim 1, characterized in that: It also includes collecting feedback data from each target vehicle; if the target vehicle confirms to select the planned route, the driving status information of the target vehicle is updated, and the lane exit of the target vehicle without a planned route is dynamically planned and adjusted.

Images