An unmanned aerial vehicle based vehicle illegal parking identification system

Abstract

The application discloses a kind of vehicle illegal parking identification systems based on unmanned aerial vehicle, including management background, unmanned aerial vehicle with the management background signal connection, the unmanned aerial vehicle includes image real-time acquisition device, picture shooting device, edge processing subsystem and wireless communication subsystem;The edge processing subsystem includes first image recognition module, illegal parking judgment module, second image recognition module, output module.The present application is used to solve the problem that the existing technology is easy to reduce the timeliness of law enforcement due to response delay, and the cost of system configuration, maintenance and expansion is high, to reduce the delay of illegal parking identification system, improve the timeliness of law enforcement, and reduce the cost.

Description

Technical Field

[0001] This invention relates to the field of vehicle illegal parking inspection, and specifically to a vehicle illegal parking identification system based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Illegally parked motor vehicles on both sides of urban roads seriously affect the city's image and road traffic order, obstruct normal passage, and increase the safety hazard of traffic accidents. Traditionally, there are two control methods: manpower patrol by traffic police and the deployment of cameras in key areas to capture illegally parked vehicles. The former has high manpower costs and low enforcement efficiency, while the latter has a small coverage area, and large-scale monitoring requires a large investment in the purchase, installation, and maintenance of cameras.

[0003] To overcome the aforementioned problems, existing technologies propose combining drones and image recognition to detect and capture illegally parked vehicles. Utilizing the flexibility and route planning capabilities of drones, this approach can free up traffic police manpower and allow for large-scale inspections simultaneously. However, most existing technologies focus on vehicle identification and contour extraction using image recognition, with little discussion on how to determine whether a vehicle is illegally parked. That is, most existing technologies directly package and send the extracted vehicle information (including photos of the vehicle and license plate information) to a backend system, which then processes and identifies the images, and uses the results for machine screening and / or manual judgment to determine if a vehicle is illegally parked. This approach requires massive data transmission. When multiple drones are used to inspect different blocks simultaneously, it requires significant bandwidth and is prone to response delays, reducing enforcement timeliness. Furthermore, the backend needs to process a large number of raw images simultaneously, placing high demands on the backend server configuration. Increased drone numbers necessitate significant server expansion and upgrades, leading to substantial system maintenance and expansion costs. Therefore, it is necessary to optimize the system architecture of existing technologies. Summary of the Invention

[0004] This invention provides a vehicle illegal parking identification system based on unmanned aerial vehicles (UAVs) to solve the problems in existing technologies where response delays easily reduce the timeliness of law enforcement, while the system has high equipment, maintenance, and expansion costs. The invention aims to reduce the latency of the illegal parking identification system, improve the timeliness of law enforcement, and reduce costs.

[0005] This invention is achieved through the following technical solution:

[0006] A vehicle illegal parking identification system based on drones includes a management backend and a drone connected to the management backend by a signal. The drone includes a real-time image acquisition device, an image capturing device, an edge processing subsystem, and a wireless communication subsystem.

[0007] The real-time image acquisition device is used to acquire images of the road edge along the flight path of the UAV in real time, and identify vehicles parked on the road edge, defining them as vehicles to be investigated.

[0008] The image capturing device is used to capture top-down images, front images, and rear images of the vehicle to be inspected;

[0009] The wireless communication subsystem is used to realize wireless communication between the edge processing module and the management backend;

[0010] The edge processing subsystem includes:

[0011] The first image recognition module identifies the vehicle outline and parking frame markings based on a top-down image of the vehicle to be inspected.

[0012] The illegal parking detection module determines whether the vehicle to be investigated is illegally parked based on the recognition result of the first image recognition module; if so, it is defined as an illegally parked vehicle.

[0013] The second image recognition module extracts license plate information based on the front and rear images of the illegally parked vehicle;

[0014] The output module is used to package the location information, license plate information, top-down image, front image, and rear image of illegally parked vehicles into the wireless communication subsystem.

[0015] To address the issues of reduced enforcement timeliness due to response delays in existing technologies, coupled with high system setup, maintenance, and expansion costs, this invention proposes a vehicle illegal parking identification system based on unmanned aerial vehicles (UAVs). This system uses a real-time image acquisition device to acquire road edge images along the UAV's flight path and identifies vehicles parked on the road edge as vehicles to be investigated. It then captures overhead, front, and rear images of the vehicles using an image capture device. An edge processing subsystem processes the captured images locally to identify illegally parked vehicles. Finally, a wireless communication subsystem packages and sends all relevant information about illegally parked vehicles to a management backend. Upon receiving information about illegally parked vehicles, the management backend can take any subsequent action, such as sending a text message to remind the owner to move the vehicle, instructing nearby traffic police to remove the vehicle on-site, or issuing a ticket as an administrative penalty. No specific limitations are specified here.

[0016] This application integrates an edge processing subsystem onto a drone, utilizing edge processing technology to achieve local identification of illegally parked vehicles. It only requires sending the illegally parked vehicle information to the management backend, avoiding the bandwidth pressure and latency issues caused by transmitting large amounts of invalid video and / or image data to the management backend, thus improving the timeliness of traffic management enforcement. Furthermore, by distributing the identification process locally across each drone through the edge processing subsystem, this application significantly reduces the configuration requirements of the backend server, lowering system deployment and maintenance costs. Moreover, if the number of drones needs to be increased later in the application's lifespan, no large-scale expansion of the backend server is required, further reducing expansion costs.

[0017] In the edge processing subsystem of this solution, the first image recognition module identifies the vehicle outline and parking frame markings from the top-view image of the vehicle to be investigated. If no parking frame markings are identified outside the vehicle outline, the illegal parking judgment module determines that the vehicle to be investigated is illegally parked and defines it as an illegally parked vehicle. Then, the second image recognition module retrieves the front and rear images of the vehicle, extracts the license plate information, and sends it to the output module. The output module packages the location information, license plate information, top-view image, front image, and rear image of the illegally parked vehicle and transmits them to the wireless communication subsystem, which then sends them to the management backend. In this solution, both the first and second image recognition modules can use existing image recognition algorithms to complete their respective recognition requirements. Furthermore, if two image recognition models are built into one image recognition module, the work requirements of both image recognition modules in this application can be completed by one image recognition module. This is not difficult for those skilled in the art to implement, and therefore will not be elaborated upon here.

[0018] Furthermore, the illegal parking detection module includes:

[0019] The marked parking space determination unit is used to determine whether the outline of the vehicle is located within the parking frame marking;

[0020] The no-parking zone determination unit is used to determine whether the current road belongs to a no-parking zone;

[0021] The summarization unit is used to summarize the judgment results of the marked parking space judgment unit and the no-parking zone judgment unit; if the vehicle outline is not located within any parking frame mark and the current road is a no-parking zone, then the vehicle to be investigated is identified as an illegally parked vehicle.

[0022] In this solution, the parking space determination unit determines whether the vehicle outline identified by the first image recognition module is located within the parking frame markings. This determination process can be completed using existing image recognition or machine learning technologies. Of course, a vehicle partially crossing the lines can also be considered parked within the parking frame. Furthermore, this solution uses a no-parking zone determination unit to determine whether the current road is a no-parking area. A vehicle to be investigated must simultaneously meet the following two conditions to be identified as illegally parked by the illegal parking determination module of this solution: the current road is a no-parking area, and the vehicle to be investigated is not within the parking frame.

[0023] In this solution, the no-parking zone judgment unit can identify no-parking zones based on pre-planned drone paths, control the drone to only work when patrolling roads within no-parking zones, automatically identify them based on traffic signs when the drone enters an intersection, and even determine whether there are no-parking signs on the roadside based on overhead images (such as strictly no-parking sections painted yellow).

[0024] Furthermore, if the marked parking space judgment unit determines that the vehicle outline is within the parking frame marking, the marked parking space judgment unit is also used to determine whether the vehicle outline partially obscures the parking frame marking: if so, the summarizing unit determines that the current vehicle to be investigated is parked improperly, and the output module transmits the license plate information of the improperly parked vehicle to the wireless communication module.

[0025] Although some roads have marked parking spaces on the side, some drivers still park improperly, with parts of their vehicles outside the parking lines. This interferes with normal traffic and endangers road safety, and current technology offers no solution. To overcome this deficiency, this solution uses a marked parking space detection unit to determine whether the vehicle's outline partially obscures the parking line markings, thus identifying improper parking. If improperly parked, the license plate information is sent to a wireless communication module and transmitted to a management backend. The backend then sends a reminder or suggestion to the driver, such as via SMS, to remind them to park properly.

[0026] Furthermore, the illegal parking judgment module also includes an exemption judgment unit, which is used to obtain the license plate color and image information of the illegally parked vehicles identified in the summary unit, and to judge the type of illegally parked vehicle based on the license plate color and image information; if the illegally parked vehicle belongs to the exempted vehicle, then the definition of it as an illegally parked vehicle is cancelled.

[0027] Different cities or regions may have different definitions of vehicles that can be exempted from illegal parking penalties. Therefore, this solution can flexibly set the license plate color and image information of exempted vehicles, and filter illegally parked vehicles through the exemption judgment unit to remove exempted vehicles locally, thereby avoiding useless data transmission and avoiding interference with special vehicles such as ambulances, rescue vehicles, and law enforcement vehicles.

[0028] The exemption judgment unit of this scheme first performs a preliminary screening based on the license plate color; for example, license plates with white backgrounds and black lettering are directly excluded from the definition of illegally parked vehicles. Secondly, a second screening is performed based on image information to remove special-operation vehicles such as ambulances and engineering vehicles. In the second screening process, machine learning technology is used to identify whether the current vehicle's image information belongs to the designated special-operation vehicle; the aforementioned image information can be any one or more of the following: top-down view, front view, and rear view.

[0029] Furthermore, the illegal parking judgment module also includes a temporary parking judgment unit, which is used to determine whether the illegally parked vehicle identified in the summary unit is a temporary parking vehicle: if so, the definition of it as an illegally parked vehicle is cancelled.

[0030] Existing drone-based illegal parking detection technologies mostly report vehicles immediately upon detection of parking in no-parking zones, failing to consider temporary parking situations such as passenger pick-up / drop-off, urination / delivery, etc., lacking human consideration. Therefore, this solution incorporates a temporary parking judgment unit into the illegal parking judgment module. For vehicles already identified as illegally parked, it determines whether their parking is temporary. If determined to be temporary, the illegal parking definition is removed, allowing the temporary parking to occur, thus making traffic enforcement more humane and reasonable.

[0031] Furthermore, the temporary parking determination unit includes a third image recognition module; the third image recognition module is used to extract facial information from the image of the vehicle's front. This solution uses the third image recognition module to obtain the image of the vehicle's front and recognize facial information within the windshield area. If a facial image is recognized, it can be determined that the vehicle is currently in a temporary parking situation.

[0032] Furthermore, the temporary stop determination unit also includes:

[0033] The classification subunit is used to classify illegally parked vehicles that do not qualify as exempt vehicles into gasoline vehicles and new energy vehicles based on the color of the license plate.

[0034] The oil tanker temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as oil tankers are temporary parking vehicles.

[0035] The new energy vehicle temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as new energy vehicles are temporary parking vehicles.

[0036] During their research, the inventors discovered that relying on images of the vehicle's front to extract facial information is prone to errors due to various factors such as light intensity, lighting angle, windshield reflection, and window tinting. This results in low accuracy and significant recognition errors, affecting the accuracy of temporary parking determination. Therefore, this solution provides a more feasible method for determining temporary parking. First, a classification subunit identifies the license plate color, categorizing illegally parked vehicles that are not exempt into gasoline vehicles and new energy vehicles. Then, separate temporary parking determination subunits for gasoline vehicles and new energy vehicles are used to determine whether each type of vehicle qualifies for temporary parking. By differentiating between gasoline and new energy vehicles, this solution allows for more targeted temporary parking determination techniques based on their different power sources, significantly improving the accuracy of temporary parking determination.

[0037] Furthermore, the temporary stop determination unit also includes:

[0038] The thermal infrared imaging subunit is used to acquire local thermal infrared images of illegally parked vehicles;

[0039] The temperature extraction subunit is used to extract vehicle body temperature data of a specified area in a local thermal infrared image;

[0040] An ambient temperature monitoring subunit is used to acquire ambient temperature data.

[0041] The oil tanker temporary parking judgment subunit determines whether a vehicle classified as an oil tanker is a temporary parking vehicle by comparing ambient temperature data with the body temperature data of the engine hood.

[0042] The new energy vehicle temporary parking judgment subunit determines whether illegally parked vehicles classified as new energy vehicles are temporary parking vehicles by comparing ambient temperature data with vehicle body temperature data on the tire surface.

[0043] For gasoline vehicles: The thermal infrared imaging subunit acquires thermal infrared images of illegally parked vehicles, including the hood, and the temperature extraction subunit extracts the temperature of the hood area. This temperature is then compared with the ambient temperature data. If the temperature of the hood area is more than 3°C higher than the ambient temperature, the gasoline vehicle is considered to have either not been turned off or had just been turned off, and is therefore considered a temporarily parked vehicle.

[0044] For new energy vehicles: The thermal infrared image of the illegally parked vehicle, including the tires, is obtained through the thermal infrared imaging subunit, and the tire temperature is extracted through the temperature extraction subunit. The temperature is then compared with the ambient temperature data. If the tire temperature is more than 2°C higher than the ambient temperature, the new energy vehicle can be considered to be a temporarily parked vehicle.

[0045] This solution utilizes the respective power characteristics of gasoline vehicles and new energy vehicles, and makes temporary stop judgments by measuring the temperature of different parts of the vehicle body, thus overcoming the problem of errors that can easily occur when relying on in-vehicle human image recognition technology to judge the temporary stop of the vehicle.

[0046] Furthermore, the temporary parking determination unit also includes a timing subunit; the timing subunit is used to start timing from the time the temporary parking vehicle is determined until a set duration.

[0047] For vehicles that are temporarily parked, a timing subunit begins to start timing once the vehicle is identified as such by this application. During the timing period, the drone can remain hovering or continue subsequent patrol operations. After the set time has elapsed, the drone returns to the location of the corresponding temporarily parked vehicle to determine if the vehicle is still there. If the temporarily parked vehicle remains in place, it can be identified as an illegally parked vehicle.

[0048] Furthermore, the vehicle illegal parking identification system also includes several rest subsystems set up along the flight path of the drone for temporary parking of the drone; during the timing process of the timing subunit, the drone docks at the nearest rest subsystem.

[0049] During the timing subunit's timing process, keeping the drone hovering consumes a significant amount of energy; continuing subsequent patrol operations can easily lead to overlapping of multiple tasks, placing high demands on program design and edge processing module performance, which is detrimental to cost control. Therefore, this solution sets up several rest subsystems along the patrol path in key monitoring areas. During the timing subunit's timing process, the drone docks at the nearest rest subsystem, thereby reducing energy consumption, avoiding overlapping of multiple tasks, and lowering control complexity and edge processing costs. Furthermore, the drone can recharge while docking at a rest subsystem, extending its single patrol endurance.

[0050] Compared with the prior art, the present invention has at least the following advantages and beneficial effects:

[0051] 1. This invention discloses a vehicle illegal parking identification system based on drones. It utilizes edge processing technology to achieve local drone identification of illegally parked vehicles. It only needs to send the information of illegally parked vehicles to the management backend, avoiding the bandwidth pressure and delay response problems caused by transmitting a large amount of invalid video and / or image data to the management backend, which is conducive to improving the enforcement timeliness of traffic management departments.

[0052] 2. The present invention provides a vehicle illegal parking identification system based on drones. By distributing the identification process to each drone for local processing through an edge processing subsystem, the system can significantly reduce the configuration requirements of the backend server and reduce the system deployment and maintenance costs. Furthermore, if the number of drones needs to be increased in the later stages of use, there is no need to carry out large-scale expansion of the backend server, which helps to reduce expansion costs.

[0053] 3. The present invention provides a vehicle illegal parking identification system based on drones, which uses a marked parking space judgment unit to determine whether the vehicle to be investigated is parked improperly and can proactively remind the owner of the improperly parked vehicle, filling the gap in the existing technology.

[0054] 4. The present invention provides a vehicle illegal parking identification system based on drones. The system excludes exempt vehicles such as law enforcement vehicles, ambulances, and rescue vehicles through an exemption judgment unit on the drone, thereby avoiding invalid data transmission and interference with special vehicles such as ambulances, rescue vehicles, and law enforcement vehicles.

[0055] 5. The present invention provides a vehicle illegal parking identification system based on drones, which also incorporates a temporary parking judgment unit in the illegal parking judgment module. For illegally parked vehicles that have been identified as illegally parked, the system determines whether they are temporarily parked. If they are determined to be temporarily parked vehicles, the system cancels the definition of illegally parked vehicles and allows the temporary parking behavior to occur, thereby making road traffic law enforcement more humane and reasonable.

[0056] 6. The present invention provides a vehicle illegal parking identification system based on drones, which utilizes the respective power characteristics of gasoline vehicles and new energy vehicles. By measuring the temperature of different parts of the vehicle body, it makes a temporary parking judgment, overcoming the problem that the image recognition technology of people inside the vehicle is prone to errors when judging the temporary parking of the vehicle. Attached Figure Description

[0057] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0058] Figure 1 This is a system schematic diagram of a specific embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of the edge processing subsystem in a specific embodiment of the present invention;

[0060] Figure 3 This is a schematic diagram of the working method of a specific embodiment of the present invention;

[0061] Figure 4 This is a schematic diagram of the rest subsystem in a specific embodiment of the present invention.

[0062] The attached diagram shows the markings and corresponding component names:

[0063] 1-Stop surface, 2-Stop plate, 3-Drive device, 4-Blocking part. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments and accompanying drawings. The illustrative embodiments and descriptions of this invention are for explaining the invention only and are not intended to limit the invention. In the description of this application, it should be understood that terms such as "front," "rear," "left," "right," "upper," "lower," "vertical," "horizontal," "high," "low," "inner," and "outer," indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the scope of protection of this application.

[0065] Example 1:

[0066] like Figure 1 and Figure 2 The illustrated vehicle illegal parking identification system based on drones includes a management backend and several drones connected to the management backend via signals. Each drone has a built-in real-time image acquisition device, an image capturing device, an edge processing subsystem, and a wireless communication subsystem. In this embodiment, the real-time image acquisition device can be a camera, and the image capturing device can be an action camera. Of course, the real-time image acquisition device and the image capturing device can also be combined into a single capturing device.

[0067] The real-time image acquisition device is used to acquire images of the road edge along the flight path of the UAV in real time, and identify vehicles parked on the road edge, defining them as vehicles to be investigated.

[0068] The image capturing device is used to capture top-down images, front images, and rear images of the vehicle to be inspected;

[0069] The wireless communication subsystem is used to realize wireless communication between the edge processing module and the management backend;

[0070] The edge processing subsystem includes:

[0071] The first image recognition module identifies the vehicle outline and parking frame markings based on a top-down image of the vehicle to be inspected.

[0072] The illegal parking detection module determines whether the vehicle to be investigated is illegally parked based on the recognition result of the first image recognition module; if so, it is defined as an illegally parked vehicle.

[0073] The second image recognition module extracts license plate information based on the front and rear images of the illegally parked vehicle;

[0074] The output module is used to package the location information, license plate information, top-down image, front image, and rear image of illegally parked vehicles into the wireless communication subsystem.

[0075] The illegal parking judgment module includes a marked parking space judgment unit, a no-parking zone judgment unit, a summary unit, an exemption judgment unit, and a temporary parking judgment unit.

[0076] The marked parking space judgment unit is used to determine whether the outline of the vehicle is located within the parking frame marking. If the marked parking space judgment unit determines that the outline of the vehicle is located within the parking frame marking, the marked parking space judgment unit is also used to determine whether the outline of the vehicle partially obscures the parking frame marking. If so, the summarizing unit determines that the vehicle to be investigated is parked improperly, and the license plate information of the improperly parked vehicle is transmitted to the wireless communication module through the output module.

[0077] The no-parking zone detection unit is used to determine whether the current road belongs to a no-parking zone. The no-parking zone detection unit can identify no-parking zones based on pre-planned drone path planning, or it can control the drone to only work when inspecting roads in no-parking zones. It can also automatically identify no-parking zones based on traffic signs at intersections when the drone enters them, and it can even determine whether there are no-parking signs on the curb based on overhead images.

[0078] The summarization unit is used to summarize the judgment results of the marked parking space judgment unit and the no-parking zone judgment unit; if the vehicle outline is not located within any parking frame mark and the current road is a no-parking zone, then the vehicle to be investigated is identified as an illegally parked vehicle.

[0079] The exemption judgment unit is used to obtain the license plate color and image information of illegally parked vehicles identified in the summary unit, and to determine the type of illegally parked vehicle based on the license plate color and image information; if the illegally parked vehicle belongs to the exempted vehicle category, the definition of illegally parked vehicle for it is cancelled.

[0080] The temporary parking judgment unit is used to determine whether the illegally parked vehicles identified in the summary unit are temporary parking vehicles: if so, the definition of the illegally parked vehicle is canceled.

[0081] In this embodiment, the temporary parking determination unit includes a third image recognition module, a classification subunit, a temporary parking determination subunit for gasoline vehicles, a temporary parking determination subunit for new energy vehicles, a thermal infrared imaging subunit, a temperature extraction subunit, and an ambient temperature monitoring subunit; specifically:

[0082] The third image recognition module is used to extract human image information from the front of the vehicle.

[0083] The classification subunit is used to classify illegally parked vehicles that are not exempt from exemption into gasoline vehicles and new energy vehicles based on the color of the license plate.

[0084] The oil tanker temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as oil tankers are temporary parking vehicles.

[0085] The new energy vehicle temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as new energy vehicles are temporary parking vehicles.

[0086] The thermal infrared imaging subunit is used to acquire local thermal infrared images of illegally parked vehicles.

[0087] The temperature extraction subunit is used to extract vehicle body temperature data from a specified area in a local thermal infrared image.

[0088] The ambient temperature monitoring subunit is used to acquire ambient temperature data.

[0089] The temporary parking judgment subunit for fuel vehicles determines whether illegally parked vehicles classified as fuel vehicles are temporary parking vehicles by comparing ambient temperature data with vehicle body temperature data under the hood.

[0090] The new energy vehicle temporary parking judgment subunit determines whether illegally parked vehicles classified as new energy vehicles are temporary parking vehicles by comparing ambient temperature data with vehicle body temperature data on the tire surface.

[0091] In this embodiment, the first image recognition module, the second image recognition module, and the third image recognition module all perform the required image recognition functions based on existing image recognition algorithms; preferably, a lightweight image recognition model is used to facilitate edge processing.

[0092] The working method of the vehicle illegal parking recognition system in this embodiment is as follows: Figure 3 As shown, it includes the following steps:

[0093] S1. Make the drone fly along the designated path. During the flight, the drone will acquire real-time images of the road edge through the real-time image acquisition device and identify vehicles parked on the road edge. When a vehicle is found parked on the road edge, it will be defined as a vehicle to be investigated.

[0094] S2. Take top-down, front-end, and rear-end images of the vehicle to be inspected using an image capturing device, and transmit them to the edge processing subsystem.

[0095] S3. The edge processing subsystem identifies the top-view image of the vehicle to be investigated through the first image recognition module, extracts the vehicle outline, identifies whether there is a parking frame mark, and sends the recognition result to the illegal parking judgment module.

[0096] S4. The illegal parking detection module performs the following judgments through the marked parking space detection unit:

[0097] First, determine if the image contains a parking box icon:

[0098] If so, determine that the vehicle to be investigated is not illegally parked; continue to determine whether the vehicle to be investigated is parked properly (the method of judgment is whether the outline of the vehicle partially obscures the parking frame mark): if not, transmit the license plate information of the improperly parked vehicle to the wireless communication module through the output module and send it to the management backend, and then proceed to step S7; if yes, proceed directly to step S7.

[0099] If not, the no-parking zone determination unit will be used to determine whether the current road belongs to a no-parking zone:

[0100] If the vehicle is not in a no-parking zone, determine that it is not illegally parked and proceed to step S7;

[0101] If the area is a no-parking zone, the vehicle currently under investigation is defined as an illegally parked vehicle.

[0102] The exemption determination unit determines whether the illegally parked vehicle is an exempt vehicle:

[0103] If so, cancel the definition of the illegally parked vehicle and proceed to step S7;

[0104] If not, the temporary parking determination unit will determine whether the currently illegally parked vehicle is a temporary parking vehicle:

[0105] If so, cancel the definition of the illegally parked vehicle and proceed to step S7;

[0106] If not, then it is determined to be an illegally parked vehicle.

[0107] S5. The second image recognition module identifies the license plate information of the illegally parked vehicle.

[0108] S6. The output module packages the location information, license plate information, top-down image, front image, and rear image of the illegally parked vehicle into the wireless communication subsystem and sends it to the management backend.

[0109] S7. Complete the identification of illegal parking of the current vehicle and return to step S1.

[0110] In a more preferred embodiment, the method by which the temporary parking determination unit determines whether a currently illegally parked vehicle is a temporary parking vehicle includes:

[0111] The third image recognition module retrieves an image of the front of the vehicle and identifies whether the image contains human facial information.

[0112] If so, then the illegally parked vehicle is determined to be a temporary parking vehicle;

[0113] If not, obtain the license plate color through the classification sub-unit and classify the currently illegally parked vehicle as a gasoline vehicle or a new energy vehicle;

[0114] If the license plate is blue, it is classified as a gasoline vehicle. For gasoline vehicles: The thermal infrared imaging subunit acquires the thermal infrared image of the illegally parked vehicle, including the hood, and the temperature extraction subunit extracts the temperature of the hood area. Then, the gasoline vehicle temporary parking judgment subunit compares the temperature with the ambient temperature data. If the temperature of the hood area is more than 3°C higher than the ambient temperature, the gasoline vehicle is considered to be a temporarily parked vehicle; otherwise, the gasoline vehicle is not considered to be a temporarily parked vehicle.

[0115] If the license plate is green, it is classified as a new energy vehicle. For new energy vehicles: the thermal infrared imaging subunit acquires thermal infrared images of the illegally parked vehicle, including the tires, and the temperature extraction subunit extracts the tire temperature. Then, the new energy vehicle temporary parking judgment subunit compares the temperature with the ambient temperature data. If the tire temperature is more than 2°C higher than the ambient temperature, the new energy vehicle is considered to be a temporarily parked vehicle; otherwise, the new energy vehicle is not considered to be a temporarily parked vehicle.

[0116] Example 2:

[0117] A vehicle illegal parking identification system based on drones, in addition to embodiment 1, includes a timing subunit in the temporary parking judgment unit; the timing subunit is used to start timing from the time the temporary parking vehicle is judged until a set time is reached.

[0118] The vehicle illegal parking identification system also includes several rest subsystems installed along the flight path of the drone for temporary parking of the drone.

[0119] In this embodiment, during the timing process of the timing subunit, the corresponding drone docks at the nearest rest subsystem. After the timing time reaches the set temporary parking threshold, it returns to its original location and checks whether the temporarily parked vehicle is still there; if the temporarily parked vehicle is still in its original location, it is determined to be an illegally parked vehicle.

[0120] Example 3:

[0121] Based on Embodiment 2, this embodiment provides a rest subsystem specifically for the drone used in this application, such as a vehicle illegal parking identification system using a drone. Figure 4 As shown, the device includes a stop plate 2 that matches a streetlight pole or utility pole. The bottom end of the stop plate 2 is hinged to the outer wall of the streetlight pole or utility pole. A rotating shaft is fixedly connected to the hinged edge of the stop plate 2. The rotating shaft is driven by a drive device 3, thereby controlling the rotation of the stop plate 2. The inner wall of the stop plate 2 matches the streetlight pole or utility pole. The drive device 3 is a servo motor or a stepper motor.

[0122] In this embodiment, the inner wall of the stop plate 2 is provided with a stop plane 1.

[0123] Under normal conditions, the stop plate 2 is vertically attached to the outer wall of the street light pole or utility pole, and the stop plane 1 is parallel to the axis of the street light pole or utility pole.

[0124] When a resting space is needed for the drone, the drive unit 3 drives the landing plate 2 to rotate downwards to the open state, so that the landing plane 1 is horizontal, and the drone lands on the landing plane 1.

[0125] In this embodiment, the dimensions of the stop plate 2 and the stop plane 1 can be flexibly set according to the specific model and size of the drone.

[0126] In a more preferred embodiment, the landing surface 1 may have a built-in wireless charging module for charging drones that land on it.

[0127] In a more preferred embodiment, a limiting member located below the stop plate 2 can be provided on the street light pole or utility pole to limit the rotation angle of the stop plate 2; so that when the outer wall of the stop plate 2 rotates to abut the limiting member, the stop plane 1 is exactly horizontal.

[0128] In a more preferred embodiment, the landing plate 2 is an arc-shaped structure that matches the street lamp pole or utility pole, and the landing surface 1 is excavated from its arc-shaped inner wall; therefore, there are naturally formed blocking parts 4 on both sides of the landing surface 1 to reduce the risk of drones falling on it in windy weather.

[0129] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0130] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

Claims

1. A vehicle illegal parking identification system based on unmanned aerial vehicles (UAVs), comprising a management backend and a UAV signal-connected to the management backend, characterized in that, The drone includes a real-time image acquisition device, a picture shooting device, an edge processing subsystem, and a wireless communication subsystem; The real-time image acquisition device is used to acquire images of the road edge along the flight path of the UAV in real time, and identify vehicles parked on the road edge, defining them as vehicles to be investigated. The image capturing device is used to capture top-down images, front images, and rear images of the vehicle to be inspected; The wireless communication subsystem is used to realize wireless communication between the edge processing module and the management backend; The edge processing subsystem includes: The first image recognition module identifies the vehicle outline and parking frame markings based on a top-down image of the vehicle to be inspected. The illegal parking detection module determines whether the vehicle to be investigated is illegally parked based on the recognition result of the first image recognition module; if so, it is defined as an illegally parked vehicle. The second image recognition module extracts license plate information based on the front and rear images of the illegally parked vehicle; The output module is used to package the location information, license plate information, top-down image, front image and rear image of the illegally parked vehicle into the wireless communication subsystem; The illegal parking detection module includes: The marked parking space determination unit is used to determine whether the outline of the vehicle is located within the parking frame marking; The no-parking zone determination unit is used to determine whether the current road belongs to a no-parking zone; The summarization unit is used to summarize the judgment results of the marked parking space judgment unit and the no-parking zone judgment unit; if the vehicle outline is not located within any parking frame mark and the current road is a no-parking zone, then the vehicle to be investigated is identified as an illegally parked vehicle. The illegal parking judgment module further includes: a temporary parking judgment unit, used to judge whether the illegally parked vehicles identified in the summary unit are temporary parking vehicles; if so, the definition of the illegally parked vehicle is cancelled. The temporary stop determination unit includes a third image recognition module; the third image recognition module is used to extract human image information from the front image of the vehicle. The temporary stop determination unit also includes: The classification subunit is used to classify illegally parked vehicles that do not qualify as exempt vehicles into gasoline vehicles and new energy vehicles based on the color of the license plate. The oil tanker temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as oil tankers are temporary parking vehicles. The new energy vehicle temporary parking judgment subunit is used to determine whether illegally parked vehicles classified as new energy vehicles are temporary parking vehicles. The temporary stop determination unit also includes: The thermal infrared imaging subunit is used to acquire local thermal infrared images of illegally parked vehicles; The temperature extraction subunit is used to extract vehicle body temperature data of a specified area in a local thermal infrared image; An ambient temperature monitoring subunit is used to acquire ambient temperature data. The oil tanker temporary parking judgment subunit determines whether a illegally parked vehicle classified as an oil tanker is a temporary parking vehicle by comparing ambient temperature data with the vehicle body temperature data of the hood. If the temperature of the hood area is more than 3°C higher than the ambient temperature, the oil tanker is considered to be a temporary parking vehicle. The new energy vehicle temporary parking judgment subunit determines whether a illegally parked vehicle classified as a new energy vehicle is a temporary parking vehicle by comparing ambient temperature data with the vehicle body temperature data on the tire surface; if the tire temperature is more than 2°C higher than the ambient temperature, the new energy vehicle is considered to be a temporary parking vehicle.

2. The vehicle illegal parking identification system based on unmanned aerial vehicles according to claim 1, characterized in that, If the marked parking space judgment unit determines that the vehicle outline is within the parking frame marking, the marked parking space judgment unit is also used to determine whether the vehicle outline partially obscures the parking frame marking: if so, the summarizing unit determines that the current vehicle to be investigated is parked improperly, and the output module transmits the license plate information of the improperly parked vehicle to the wireless communication module.

3. The vehicle illegal parking identification system based on unmanned aerial vehicles according to claim 1, characterized in that, The illegal parking detection module also includes: The exemption judgment unit is used to obtain the license plate color and image information of illegally parked vehicles identified in the summary unit, and to determine the type of illegally parked vehicle based on the license plate color and image information; if the illegally parked vehicle belongs to the exempted vehicle category, the definition of illegally parked vehicle for it is cancelled.

4. The vehicle illegal parking identification system based on unmanned aerial vehicles according to claim 1, characterized in that, The temporary parking determination unit also includes a timing subunit; the timing subunit is used to start timing from the time the temporary parking vehicle is determined until a set duration.

5. A vehicle illegal parking identification system based on unmanned aerial vehicles (UAVs) according to claim 4, characterized in that, The vehicle illegal parking identification system also includes several rest subsystems set up along the flight path of the drone for temporary parking of the drone; during the timing process of the timing subunit, the drone docks at the nearest rest subsystem.

Images