Motor vehicle illegal parking identification method and system based on unmanned aerial vehicle cluster

Abstract

The invention discloses a motor vehicle illegal parking identification method and system based on an unmanned aerial vehicle cluster, and the method comprises the following steps: controlling the flight of the unmanned aerial vehicle cluster, and enabling the unmanned aerial vehicle cluster to inspect a designated region according to a set cruise height and route; in the unmanned aerial vehicle cluster inspection process, ground images are obtained in real time through imaging equipment carried on an unmanned aerial vehicle holder; identifying the motor vehicles in the ground image, determining the position coordinates of each motor vehicle and counting the total number of the identified motor vehicles; according to a set illegal parking judgment rule, illegal parking detection is carried out on the recognized motor vehicle, voting judgment is carried out based on an unmanned aerial vehicle cluster, and if it is considered that the motor vehicle has an illegal parking behavior, warning information is automatically generated. According to the invention, the illegal parking motor vehicle can be quickly identified based on the unmanned aerial vehicle, so that the detection efficiency and accuracy can be greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of machine vision, and in particular to a method and system for identifying illegally parked motor vehicles based on drone clusters. Background Art

[0002] With the acceleration of urbanization and the continuous increase in the number of motor vehicles, the problem of illegal parking of motor vehicles often occurs. At present, the traditional method of identifying and detecting illegal parking of vehicles mainly relies on manual patrols and fixed cameras, which has problems such as low detection efficiency, limited coverage, and susceptibility to environmental influences. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a method and system for identifying illegally parked motor vehicles based on drone swarms, which can quickly identify illegally parked motor vehicles based on drones, thereby greatly improving detection efficiency and accuracy.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] On the one hand, a method for identifying illegal parking of motor vehicles based on drone clusters is provided, which comprises the following steps:

[0006] Control the flight of the drone swarm so that each drone in the drone swarm inspects the same designated area according to the corresponding cruising altitude and route;

[0007] During each drone inspection, the imaging device on the drone gimbal acquires ground images in real time;

[0008] Identify the motor vehicles in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images;

[0009] Based on the voting of the drone swarm, each motor vehicle identified in all ground images is tested for illegal parking to determine whether each motor vehicle is an illegally parked vehicle;

[0010] When at least one motor vehicle is determined to be an illegally parked vehicle, a warning message is automatically generated.

[0011] Preferably, each drone is controlled to fly according to formula (1):

[0012]

[0013] Among them, P(t) is the position of the UAV after flying for t time, P0 is the initial position, V is the flight speed of the UAV, and a is the acceleration.

[0014] Preferably, the imaging device acquires ground images at a downward angle of 45°-90°.

[0015] Preferably, after the motor vehicle is identified, the position coordinates of the motor vehicle are determined according to formula (2):

[0016]

[0017] Where L(x,y) is the recognition result of a point in the ground image; I(x+i,y+j) is the pixel value at position (x+i,y+j) in the ground image; W ij is the weight matrix of image recognition, n and m are the number of rows and columns of the weight matrix respectively.

[0018] Preferably, based on the voting of the drone cluster, each motor vehicle identified in all ground images is subjected to illegal parking detection to determine whether each motor vehicle is an illegally parked vehicle, including the following steps:

[0019] According to the set illegal parking determination rules, the same identified motor vehicle is detected for illegal parking, wherein the illegal parking determination rules include judging whether the motor vehicle is parked in the parking area according to the position coordinates of the motor vehicle; if the motor vehicle is not parked in the parking area, it is considered that the motor vehicle has illegal parking behavior, otherwise, it is not considered that the motor vehicle has illegal parking behavior;

[0020] All drones vote to determine whether the same identified motor vehicle is an illegally parked vehicle.

[0021] Preferably, the illegal parking determination rule includes determining whether the motor vehicle is parked in a parking area based on the position coordinates of the motor vehicle.

[0022] Preferably, all drones vote according to formula (3) to determine whether the same identified motor vehicle is an illegally parked vehicle:

[0023]

[0024] Where V(x) represents the total number of votes for the identified motor vehicle x, U represents the total number of drones participating in the vote, and v i (x) represents the vote of the i-th drone on the motor vehicle x. If it is considered that the motor vehicle x has illegally parked, then v i (x)=1, otherwise v i (x) = 0, V threshold represents the voting threshold for determining that motor vehicle x is an illegally parked vehicle; when V(x) is greater than or equal to the set voting threshold V threshold When , motor vehicle x is determined to be an illegally parked vehicle.

[0025] Preferably, the method for identifying illegally parked motor vehicles also includes: storing all ground images containing illegally parked motor vehicles and illegal parking information in a cloud database, and each ground image containing illegally parked motor vehicles and each illegal parking information is used as a data point, and the data points are analyzed to obtain data analysis results.

[0026] Preferably, the data points are analyzed by formula (4) to obtain data analysis results:

[0027]

[0028] Where D(t) is the data analysis result at time t, f i (t) is the function representation of the ith data point, M is the total number of data points, and λ is the data attenuation coefficient.

[0029] On the other hand, a motor vehicle illegal parking identification system is also provided, which is used to implement the above motor vehicle illegal parking identification method, and comprises:

[0030] A number of drones, which are organized into a drone swarm;

[0031] A plurality of imaging devices, which are carried on the drones and used to obtain ground images, and each drone in the drone cluster is equipped with an imaging device;

[0032] a motor vehicle identification unit, which is used to identify the motor vehicles in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images;

[0033] An illegal parking detection unit is used to perform illegal parking detection on each motor vehicle identified in all ground images to determine whether each motor vehicle is an illegally parked vehicle;

[0034] A drone swarm voting unit, which is used to determine whether the identified motor vehicle is an illegally parked vehicle;

[0035] A monitoring alarm unit, which is used to automatically generate a warning message when at least one motor vehicle is determined to be an illegally parked vehicle;

[0036] And a data analysis unit, which is used to receive all ground images containing illegally parked motor vehicles and illegal parking information, and analyze the data points to obtain data analysis results.

[0037] The present invention can acquire ground images in real time based on drones, and in combination with image recognition algorithms, automatically identify and detect illegal parking behaviors of motor vehicles, and upload the detection results to the background management system, which can greatly improve the efficiency and accuracy of illegal parking detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flowchart of the steps of the method for identifying illegal parking of motor vehicles based on drone clusters in the present invention;

[0039] Figure 2 The image of the motor vehicle identified in the present invention;

[0040] Figure 3 It is a structural schematic diagram of the motor vehicle illegal parking identification system based on drone cluster in the present invention. DETAILED DESCRIPTION

[0041] 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.

[0042] Embodiment 1:

[0043] like Figure 1 As shown, this embodiment provides a method for identifying illegal parking of motor vehicles based on drone clusters, which includes the following steps:

[0044] S1. Control the flight of the drone cluster so that each drone in the drone cluster inspects the same designated area according to the corresponding cruising altitude and route; the cruising altitude and route of each drone during the inspection can be different;

[0045] Specifically, in this embodiment, the flight of each drone is controlled according to formula (1):

[0046]

[0047] Among them, P(t) is the position of the drone after flying for t time, P0 is the initial position, V is the flight speed of the drone, and a is the acceleration. This formula ensures that the drone flies according to the predetermined path and altitude to cover the target detection area; and the initial position P0, flight speed V, flight time t and acceleration a of each drone can be the same or different;

[0048] S2. During each drone inspection, the ground image is acquired in real time by the imaging device (including a high-resolution visible light camera) mounted on the drone gimbal; in this embodiment, the imaging device acquires the ground image at a 45°-90° downward angle to ensure that the monitoring area is covered and a clear and complete image is obtained;

[0049] S3, identify the motor vehicle in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images. In this embodiment, various image recognition modules and yolov8 target detection algorithms, dense target detection algorithms, etc. can be used to automatically identify motor vehicles in ground images. Various target recognition algorithms can be used to detect ground images to determine whether there are motor vehicles (such as Figure 2 As shown), further, after the motor vehicle is identified, the position coordinates of the motor vehicle are determined according to formula (2):

[0050]

[0051] Wherein, L(x, y) is the recognition result of the pixel point (x, y) in the ground image, that is, whether the pixel point (x, y) is a motor vehicle; I(x+i, y+j) is the pixel value of the pixel point (x+i, y+j) in the ground image; W ij is the weight matrix of image recognition, n and m are the number of rows and columns of the weight matrix respectively; therefore, the position coordinates of the motor vehicle in the image can be determined by performing a convolution operation on the image pixel value and the weight matrix according to formula (2), and the total number of recognized motor vehicles can be counted;

[0052] S4. Based on the voting of the drone cluster, each motor vehicle identified in all ground images is detected for illegal parking, and it is determined whether each motor vehicle is an illegally parked vehicle, so as to improve the accuracy of detection and reduce the false alarm rate. The specific steps include the following:

[0053] Each drone performs illegal parking detection on the same identified motor vehicle according to the set illegal parking judgment rules. The illegal parking judgment rules include judging whether the motor vehicle is parked in the parking area according to the position coordinates of the motor vehicle; if the motor vehicle is not parked in the parking area, it is considered that the motor vehicle has illegal parking behavior, otherwise, it is not considered that the motor vehicle has illegal parking behavior; the illegal parking judgment rules used by each drone are the same to ensure the consistency and independence of the judgment results of each drone;

[0054] All drones vote according to formula (3) to determine whether the same identified motor vehicle is an illegally parked vehicle:

[0055]

[0056] Where V(x) represents the total number of votes for the identified motor vehicle x, U represents the total number of drones participating in the vote, and v i (x) represents the vote of the i-th drone on the motor vehicle x. If it is considered that the motor vehicle x has illegally parked, then v i (x)=1, otherwise v i(x) = 0; V threshold represents the voting threshold for determining that motor vehicle x is an illegally parked vehicle; when V(x) is greater than or equal to the set voting threshold V threshold When , motor vehicle x is determined to be an illegally parked vehicle;

[0057] Repeat the above steps of drone illegal parking detection and drone voting to determine whether each identified motor vehicle is an illegally parked vehicle;

[0058] S5. When at least one motor vehicle is determined to be illegally parked, a warning message is automatically generated, and the illegal parking information is uploaded to the backend management system for reference by relevant departments;

[0059] The illegal parking information includes one or more of the ground image where the illegally parked motor vehicle is located, the position coordinates of the illegally parked motor vehicle, the license plate number of the illegally parked motor vehicle, and the motor vehicle density distribution map;

[0060] S6. All ground images containing illegally parked motor vehicles and illegal parking information are stored in a cloud database, and each ground image containing illegally parked motor vehicles and each illegal parking information is used as a data point. The data points are further analyzed using formula (4) to obtain data analysis results:

[0061]

[0062] Where D(t) is the data analysis result at time t, f i (t) is the function representation of the ith data point, M is the total number of data points, and λ is the data decay coefficient; thus, the formula analyzes and predicts the detection data by combining the function representation of each data point and the time decay coefficient, providing data support for improving the detection algorithm, so as to optimize the algorithm performance and improve the detection accuracy of the system.

[0063] Therefore, this embodiment uses drones to acquire ground images in real time, combines image recognition algorithms, automatically identifies and detects illegal parking of motor vehicles, and uploads the detection results to the background management system, which can greatly improve the efficiency and accuracy of illegal parking detection and provide strong support for traffic management departments to conduct effective illegal parking monitoring and law enforcement.

[0064] Embodiment 2:

[0065] This embodiment provides a motor vehicle illegal parking recognition system, which can implement the motor vehicle illegal parking recognition method described in Example 1, such as Figure 3 As shown, it includes:

[0066] A plurality of drones 1 are formed into a drone cluster;

[0067] A plurality of imaging devices 2, which are mounted on the drones 1 and used to obtain ground images, and each drone 1 in the drone cluster is equipped with an imaging device 2;

[0068] A motor vehicle identification unit 3 is used to identify the motor vehicles in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images, and the process is the same as step S3;

[0069] The illegal parking detection unit 4 is used to perform illegal parking detection on each motor vehicle identified in all ground images to determine whether each motor vehicle is an illegally parked vehicle. The process is the same as the step in step S4 where each drone performs illegal parking detection on the same identified motor vehicle;

[0070] The drone group voting unit 5 is used to determine whether the identified motor vehicle is an illegally parked vehicle, and the process is the same as the step in step S4 where the drones vote according to formula (3);

[0071] The monitoring alarm unit 6 is used to automatically generate a warning message when determining that at least one motor vehicle is illegally parked, and upload the illegal parking information to the background management system, and the process is the same as step S5;

[0072] And a data analysis unit 7, which is used to receive all ground images containing illegally parked motor vehicles and illegal parking information, and analyze the data points through formula (4) to obtain data analysis results.

[0073] In summary, the present invention acquires ground images in real time through drones, combines image recognition algorithms, automatically identifies and detects illegal parking of motor vehicles, and uploads the detection results to the background management system, which can greatly improve the efficiency and accuracy of illegal parking detection.

[0074] It should be noted that the technical features in the above-mentioned embodiments 1-2 can be combined arbitrarily, and the combined technical solutions all belong to the protection scope of the present application. In this article, terms such as "including", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.

[0075] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for identifying illegal parking of motor vehicles based on drone clusters, characterized in that: The steps include: Control the flight of the drone swarm so that each drone in the drone swarm inspects the same designated area according to the corresponding cruising altitude and route; During each drone inspection, the imaging device on the drone gimbal acquires ground images in real time; Identify the motor vehicles in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images; Based on the voting of the drone swarm, each motor vehicle identified in all ground images is tested for illegal parking to determine whether each motor vehicle is an illegally parked vehicle; When at least one motor vehicle is determined to be an illegally parked vehicle, a warning message is automatically generated.

2. The method for identifying illegal parking of motor vehicles according to claim 1, characterized in that: Control the flight of each drone according to formula (1): Among them, P(t) is the position of the UAV after flying for t time, P0 is the initial position, V is the flight speed of the UAV, and a is the acceleration.

3. The method for identifying illegal parking of motor vehicles according to claim 1, characterized in that: The imaging device acquires ground images at a downward angle of 45°-90°.

4. The method for identifying illegal parking of motor vehicles according to claim 1, characterized in that: After the motor vehicle is identified, the position coordinates of the motor vehicle are determined according to formula (2): Where L(x,y) is the recognition result of a point in the ground image; I(x+i,y+j) is the pixel value at position (x+i,y+j) in the ground image; W ij is the weight matrix of image recognition, n and m are the number of rows and columns of the weight matrix respectively.

5. The method for identifying illegal parking of motor vehicles according to claim 1, characterized in that: Based on the voting of the drone cluster, each motor vehicle identified in all ground images is detected for illegal parking to determine whether each motor vehicle is an illegally parked vehicle, including the following steps: Perform illegal parking detection on the identified same motor vehicle according to the set illegal parking determination rules, wherein the illegal parking determination rules include determining whether the motor vehicle is parked in the parking area according to the position coordinates of the motor vehicle; If the motor vehicle is not parked in the parking area, it is considered that the motor vehicle has parked illegally. Otherwise, it is not considered that the motor vehicle has parked illegally. All drones vote to determine whether the same identified motor vehicle is an illegally parked vehicle.

6. The method for identifying illegal parking of motor vehicles according to claim 5, characterized in that: The illegal parking determination rule includes determining whether the motor vehicle is parked in a parking area based on the position coordinates of the motor vehicle.

7. The method for identifying illegal parking of motor vehicles according to claim 5, characterized in that: All drones vote according to formula (3) to determine whether the same identified motor vehicle is an illegally parked vehicle: Where V(x) represents the total number of votes for the identified motor vehicle x, U represents the total number of drones participating in the vote, and v i (x) represents the vote of the i-th drone on the motor vehicle x. If it is considered that the motor vehicle x has illegally parked, then v i (x)=1, otherwise v i (x) = 0, V threshold represents the voting threshold for determining that motor vehicle x is an illegally parked vehicle; when V(x) is greater than or equal to the set voting threshold V threshold When , motor vehicle x is determined to be an illegally parked vehicle.

8. The method for identifying illegal parking of motor vehicles according to claim 1, characterized in that: The method for identifying illegally parked motor vehicles also includes: storing all ground images containing illegally parked motor vehicles and illegal parking information in a cloud database, and each ground image containing illegally parked motor vehicles and each illegal parking information is used as a data point, and the data points are analyzed to obtain data analysis results.

9. The method for identifying illegal parking of motor vehicles according to claim 8, characterized in that: The data points are analyzed by formula (4) to obtain the data analysis results: Where D(t) is the data analysis result at time t, f i (t) is the function representation of the ith data point, M is the total number of data points, and λ is the data attenuation coefficient.

10. A motor vehicle illegal parking identification system, used to implement the motor vehicle illegal parking identification method according to any one of claims 1 to 9, characterized in that: include: A number of drones, which are organized into a drone swarm; A plurality of imaging devices, which are carried on the drones and used to obtain ground images, and each drone in the drone cluster is equipped with an imaging device; a motor vehicle identification unit, which is used to identify the motor vehicles in each ground image, determine the position coordinates of each motor vehicle in each ground image, and count the total number of motor vehicles in each ground image and / or the total number of motor vehicles in all ground images; An illegal parking detection unit is used to perform illegal parking detection on each motor vehicle identified in all ground images to determine whether each motor vehicle is an illegally parked vehicle; A drone swarm voting unit, which is used to determine whether the identified motor vehicle is an illegally parked vehicle; A monitoring alarm unit, which is used to automatically generate a warning message when at least one motor vehicle is determined to be an illegally parked vehicle; And a data analysis unit, which is used to receive all ground images containing illegally parked motor vehicles and illegal parking information, and analyze the data points to obtain data analysis results.

Images