Unmanned aerial vehicle in-road berth shooting system with intelligent parking platform and method of unmanned aerial vehicle in-road berth shooting system

Abstract

The invention discloses an unmanned aerial vehicle in-road berth shooting system with an intelligent parking platform and a method thereof, and belongs to the technical field of unmanned aerial vehicle in-road berth shooting. The system comprises a plurality of parking spaces and a street lamp post, wherein an aerial image acquisition device is further arranged on the street lamp post; the aerial image acquisition device comprises an unmanned aerial vehicle and an airport apron; a camera module, a rechargeable battery and a device management module are arranged on the unmanned aerial vehicle; the airport apron comprises a parking platform; the parking platform comprises a device body, a magnetic attraction charging mechanism and a buffering mechanism. The magnetic attraction charging mechanism comprises a charging panel, a plug-in port is formed in the charging panel, the charging panel is connected with a charging head through a support, a magnetic attraction block is arranged on the charging head, and a magnetic attraction groove is formed in the corresponding position of the device body. The buffering mechanism is arranged at the bottom of the device body and comprises a damper and an elastic piece, the damper is sleeved with the elastic piece, and a positioning sliding block is further arranged at the bottom of the device body and is in sliding connection with the supporting block. According to the invention, rapid and stable charging connection or / and data transmission is realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of unmanned aerial vehicle (UAV) in-road parking shooting, and particularly relates to an unmanned aerial vehicle in-road parking shooting system with an intelligent parking platform and a method thereof. BACKGROUND

[0002] At present, with the rapid development of unmanned aerial vehicle technology, unmanned aerial vehicles have been widely used in many fields such as aerial photography, logistics and monitoring due to their advantages of flexibility, wide field of view and automatic operation. With the maturity and application of unmanned aerial vehicle technology, a more efficient and intelligent inspection method is provided for in-road parking management. For example, in the prior art, a patent with the publication number CN108510750A discloses a method for unmanned aerial vehicle inspection of illegal parking based on a neural network model. The method realizes the detection of illegal parking behavior in a parking lot based on image recognition technology on an unmanned aerial vehicle platform. An image recognition algorithm in deep learning is applied to train a specific deep learning model, which is then applied to intelligent detection of parking information shot by the unmanned aerial vehicle, and the parking position of the vehicle, the direction of the vehicle head, and whether the vehicle blocks the road can be checked. The above-mentioned prior art also mentions that the unmanned aerial vehicle has the characteristics of strong maneuverability, high-altitude shooting and automatic cruising, which makes it particularly suitable for identifying and monitoring illegal parking behavior in open parking lots.

[0003] However, although unmanned aerial vehicles show significant technical applicability in in-road parking shooting tasks, they still have the defect of insufficient endurance in actual application. At present, most unmanned aerial vehicles use lithium batteries as a power source, which often cannot meet the demand for long-time continuous operation when performing in-road parking inspection tasks lasting more than half an hour, thereby seriously restricting the deployment effect and application potential of unmanned aerial vehicles in large-scale and long-period tasks.

[0004] Moreover, with the continuous expansion of in-road parking shooting application scenarios, the parking and charging of unmanned aerial vehicles have become increasingly prominent. At present, although there are some unmanned aerial vehicle parking aprons on the market, most of them have single functions and lack effective protection for unmanned aerial vehicles during parking and convenient charging solutions. When the unmanned aerial vehicle lands on the parking apron, the impact force during landing may cause damage to the bottom structure of the unmanned aerial vehicle, affecting its service life and flight performance. At the same time, in terms of charging, traditional charging methods often require manual plugging and unplugging of charging cables, which is tedious and inefficient, and cannot meet the development needs of unmanned aerial vehicles for automation and intelligence. SUMMARY

[0005] To overcome the above technical problems existing in the prior art, the present application proposes an unmanned aerial vehicle in-road parking shooting system with an intelligent parking platform and a method thereof, which realizes fast and stable charging connection or / and data transmission, longer endurance time, and greatly improves the intelligence level of the unmanned aerial vehicle.

[0006] The technical scheme of the present application is implemented as follows: the unmanned aerial vehicle in-lane parking area shooting system with an intelligent parking platform comprises a parking area arranged on the side of a road, the parking area comprises a plurality of parking spaces arranged in sequence along the extension direction of the road; a street lamp column is arranged at the adjacent position of the parking space, and an aerial image acquisition device is further arranged on the street lamp column; the aerial image acquisition device comprises an unmanned aerial vehicle and a parking apron; A camera module, a charging battery and a device management module are arranged on the unmanned aerial vehicle; the device management module is electrically connected with the camera module and the charging battery; the camera module is configured to shoot a motor vehicle in the parking space; after a target motor vehicle is parked in a target parking space, the unmanned aerial vehicle approaches the target motor vehicle, first moves around the target motor vehicle in a clockwise or counterclockwise direction, the camera module shoots a vehicle panoramic image of the target motor vehicle and transmits the image to the device management module; after the vehicle panoramic image is obtained, the unmanned aerial vehicle moves to the head, roof and tail of the target motor vehicle along a preset path respectively, and shoots wide-angle photos at multiple fixed points and transmits the photos to the device management module; A geomagnetic induction device is further arranged on each parking space, and the geomagnetic induction device is connected with the device management module; when the target motor vehicle enters or leaves the target parking space, the geomagnetic induction device sends an induction signal to the device management module, and the device management module controls the unmanned aerial vehicle to fly to the target parking space for shooting; The parking apron comprises a parking platform and a rain shield, and the rain shield is used to shield the top of the parking platform; the parking platform comprises a device main body, a magnetic charging mechanism and a buffer mechanism; the magnetic charging mechanism is arranged on the device main body and comprises a charging plate; the charging plate is provided with a plug-in port, the charging plate is connected with a charging head through a support, the charging head is provided with a magnetic block, and the device main body is provided with a magnetic groove at the corresponding position; the buffer mechanism is arranged at the bottom of the device main body and comprises a damper and an elastic member, the elastic member is sleeved outside the damper, the bottom of the device main body is further provided with a positioning sliding block, and the positioning sliding block is in sliding connection with a supporting block.

[0007] Further, the magnetic block is matched with the magnetic groove to realize the magnetic attraction connection between the unmanned aerial vehicle and the charging head.

[0008] Further, the elastic member is a spring, two ends of the spring are connected with the bottom of the device main body and the supporting block respectively, and the spring is used to provide a buffering force.

[0009] Further, a guide structure is arranged between the positioning sliding block and the supporting block to ensure the stable movement of the device main body during the buffering process.

[0010] Further, a platform area for parking the unmanned aerial vehicle is arranged on the device body, and the platform area corresponds to the position of the magnetic attraction charging mechanism.

[0011] Further, a plurality of buffer mechanisms are arranged on the parking platform and are evenly distributed on the bottom of the device body to provide uniform buffering effect.

[0012] Further, when the unmanned aerial vehicle photographs the target motor vehicle, the unmanned aerial vehicle moves along an arc line with the target motor vehicle as the center; the camera module photographs a plurality of point positions on the arc line in the radial direction to obtain the vehicle panoramic image. The camera range of the vehicle panoramic image extends from the front of the target motor vehicle to the rear of the target motor vehicle; when the camera module photographs a plurality of point positions, a first vehicle panoramic image is photographed at a first point position; when the unmanned aerial vehicle moves to a second point position, the camera module performs interface verification between the first vehicle panoramic image as a virtual reference and a real scene facing the second point position, and photographs a second vehicle panoramic image; the photographing is repeated until a last point position, and the image obtained at the last point position is the vehicle panoramic image.

[0013] Further, after the vehicle panoramic image is obtained, the unmanned aerial vehicle moves along a preset path from the front of the vehicle, through the top of the vehicle, to the tail of the vehicle, and photographs wide-angle photos at a plurality of fixed points; or, After the vehicle panoramic image is obtained, the unmanned aerial vehicle moves along a preset path from the tail of the vehicle, through the top of the vehicle, to the front of the vehicle, and photographs wide-angle photos at a plurality of fixed points. The device management module splices the wide-angle photos into a complete evidence photo covering the front license plate to the rear license plate of the vehicle; The device management module further includes a path planning unit for controlling the unmanned aerial vehicle to fly along a preset arc line path from the front of the vehicle, through the top of the vehicle, to the tail of the vehicle when flying and photographing, and to stay at the front of the vehicle, the top of the vehicle, and the tail of the vehicle respectively to photograph the wide-angle photos.

[0014] Further, after the unmanned aerial vehicle moves along the arc line with the target motor vehicle as the center and obtains the vehicle panoramic image, the unmanned aerial vehicle moves along a preset path from the tail of the target motor vehicle to the top of the target motor vehicle and then to the front of the target motor vehicle, photographs wide-angle photos at a plurality of fixed points, and transmits the wide-angle photos to the device management module; or, After the unmanned aerial vehicle moves along the arc line with the target motor vehicle as the center and obtains the vehicle panoramic image, the unmanned aerial vehicle moves along a preset path from the front of the target motor vehicle to the top of the target motor vehicle and then to the tail of the target motor vehicle, photographs wide-angle photos at a plurality of fixed points, and transmits the wide-angle photos to the device management module.

[0015] Further, the street lamp post is further provided with a lighting lamp, a solar panel, a battery and a main control module; the main control module is electrically connected with the lighting lamp, the solar panel and the battery respectively, and the battery is used for supplying power to each power consumption module; The plug-in port is provided with a male pole which is electrically connected with the battery through the main control module; the charging head is provided with a female pole which is electrically connected with the charging battery through the device management module; when the unmanned aerial vehicle is magnetically connected with the charging head, the main control module and the device management module are connected through the cooperation of the male pole and the female pole for charging and / or data transmission; The device management module is internally provided with an image processing unit and a first storage unit, and the vehicle surrounding view image photographed by the unmanned aerial vehicle is directly stored into the first storage unit; the image processing unit is used for splicing the wide-angle photo into the complete evidence photo which is stored into the first storage unit again; the main control module is internally provided with a second storage unit which is used for storing the vehicle surrounding view image and the complete evidence photo output from the first storage unit.

[0016] Further, the photographing range of the unmanned aerial vehicle is provided with 1-4 continuously arranged parking spaces.

[0017] The unmanned aerial vehicle road parking space photographing method with the intelligent parking platform is applied to the unmanned aerial vehicle road parking space photographing system as described above, and the unmanned aerial vehicle road parking space photographing method comprises the following steps: Step S1: when the target motor vehicle enters the target parking space, the geomagnetic induction device sends an induction signal to the device management module, and the device management module controls the unmanned aerial vehicle to fly to the target parking space for photographing; Step S2: when the target motor vehicle is parked, the unmanned aerial vehicle moves along a circular arc line with the target motor vehicle as the center; the camera module on the unmanned aerial vehicle performs multiple point photographing in the radial direction on the orbit, obtains the vehicle surrounding view image and stores it into the first storage unit; Step S3: after obtaining the vehicle surrounding view image, the unmanned aerial vehicle moves to the head, roof and tail of the target motor vehicle according to the preset path respectively, and multiple wide-angle photos are taken at multiple fixed points and transmitted to the device management module; the image processing unit in the device management module splices the wide-angle photos into a complete evidence photo, and stores the complete evidence photo into the first storage unit; Step S4: The operator remotely operates the controller or the drone's own navigation system to control the drone to fly towards the parking platform. The drone gradually approaches the platform area on the main body of the device. As the drone lands, its bottom approaches the magnetic block on the charging head. Under the action of magnetic force, the drone falls into the magnetic slot and is magnetically connected to the charging head. At this time, the male and female poles are connected to perform charging and / or data transmission. At the same time, the first storage unit transmits the vehicle circumferential image and the complete evidence photo to the second storage unit of the main control module. Step S5: When the target vehicle leaves the target parking space, the geomagnetic sensor sends a sensing signal to the device management module. The device management module controls the drone to fly to the target parking space to take a single-point photo, and repeats steps S2-S3.

[0018] Furthermore, in step S4, at the instant the drone lands and contacts the platform area, the drone generates an impact force on the main body of the device, the main body of the device moves downward, compresses the spring, the spring deforms, and converts the kinetic energy of the drone into elastic potential energy; at the same time, the damping starts to work; the positioning slider slides along the support block, and under the action of the guide structure, the main body of the device remains stable during the buffering process and will not deviate or tilt. Once the drone is fully charged and / or the data is transmitted, the magnetic connection is disconnected, and the drone takes off and leaves the platform area. At this point, the spring gradually returns to its original shape under its own elastic force, driving the main body of the device back to its initial position, ready for the next drone landing.

[0019] Furthermore, in step S2, when the camera module takes pictures at multiple points, it takes a picture of the first vehicle's perimeter at the first point; when the drone moves to the second point, the camera module uses the first vehicle perimeter image as a reference virtual scene and performs docking and verification with the real scene facing the second point to take a picture of the second vehicle perimeter image; the shooting process is repeated until the last point, and the image obtained at the last point is the vehicle perimeter image.

[0020] Furthermore, marking lines are set on the ground of each parking space according to the junction of the front, roof and rear of the vehicle in the target parking space; Furthermore, in step S3, the image processing unit performs the following operations: Step S31: Perform noise reduction and distortion correction on the wide-angle photos taken at the front, roof, and rear of the vehicle, respectively; Step S32: Extract feature points from adjacent wide-angle photos. The feature points include the marking lines set on the ground of the parking space. Based on the marking lines and other image features, feature matching is performed to determine the stitching position relationship of adjacent photos. Step S33, according to the determined splicing position relation, a plurality of wide-angle photos are spliced into a continuous image covering from the front to the rear of the vehicle through an image fusion algorithm; Step S34, vehicle features, front and rear license plate information and the parking space number marked on the parking space in the spliced image are recognized and extracted, the parking space number is associated with the vehicle information, and a complete evidence photo containing the parking space number, the vehicle and the license plate is generated; Step S35, the complete evidence photo is subjected to brightness equalization and edge smoothing processing.

[0021] Further, in step S31, the denoising can be realized by a Gaussian filter or a median filter algorithm; the distortion correction is based on the internal parameter matrix and distortion coefficient of camera factory calibration, and the image is corrected through mathematical transformation (such as Brown model).

[0022] Further, in step S32, the shape of the marking line is at least one of vertical line, horizontal line, L shape and T shape.

[0023] Further, in step S34, the recognition adopts a pre-trained deep learning model (such as a target detection model based on YOLO or CNN) for detection and recognition of vehicles, license plates and parking space numbers. By introducing the marking line as a positioning reference, the splicing alignment process of multi-angle photos is simplified. Combined with the systematic image processing process, not only clear and coherent panoramic images can be efficiently generated, but also parking spaces and vehicle information can be automatically associated to form an evidence chain closed loop.

[0024] Further, the controller comprises a first wireless communication unit, the unmanned aerial vehicle comprises a second wireless communication unit; the first wireless communication unit and the second wireless communication unit are in communication connection; the wireless communication unit is Bluetooth or Wi-Fi.

[0025] Further, the camera module is an image sensor, such as CMOS or CCD.

[0026] The beneficial effects of the present application are:

[0027] (1) In terms of structure and function, the magnetic charging mechanism in the present application is cleverly and practically arranged, so that the endurance is significantly improved. The magnetic attraction block on the charging head is matched with the magnetic attraction slot on the device main body, so that the unmanned aerial vehicle and the charging head can be quickly and stably connected by magnetic attraction, ensuring that the unmanned aerial vehicle is fixed during charging, avoiding interruption of charging due to shaking or displacement, and greatly improving the reliability and stability of charging.

[0028] (2) The cooperation of the damping and elastic members can effectively absorb the impact force generated when the unmanned aerial vehicle lands. The elastic member (such as a spring) deforms when subjected to pressure, converting kinetic energy into elastic potential energy, which is then slowly released, acting as a buffer; the damping can consume part of the energy, reducing vibration and rebound, making the unmanned aerial vehicle land more smoothly. The sliding connection of the positioning sliding block and the supporting block and the possible guiding structure ensure the stable movement of the device body during the buffering process, prevent it from deviating or tilting, and further improve the buffering effect and the stability of the overall structure.

[0029] (3) The unmanned aerial vehicle carries a camera module, which can take pictures around the target motor vehicle to obtain a vehicle panoramic image. The camera range of the vehicle panoramic image extends from the front of the target motor vehicle to the rear of the target motor vehicle, and is associated with multiple parking spaces, with a wide camera range. At the same time, based on the unmanned aerial vehicle, multiple point hovering photography can be performed without configuring multiple camera modules, effectively reducing production costs. Moreover, the camera module also obtains wide-angle photos of multiple fixed-point photography through a preset path and transmits them to the device management module, ensuring that the camera module can collect continuous vehicle image information from the front license plate to the rear license plate in multiple directions; using the complementary characteristics of multiple-angle wide-angle photos for splicing processing can fundamentally solve the problem of missing vehicle information caused by single shooting angle or limited shooting range, ultimately generating a parking record containing a complete evidence chain, and effectively avoiding parking fee disputes caused by incomplete evidence photos. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 Structure diagram of the parking area on the side of the road of the present application;

[0031] Figure 2 Structure diagram of the target motor vehicle parked in the parking area on the side of the road of the present application;

[0032] Figure 3 Top view of the unmanned aerial vehicle photographing the target motor vehicle of the present application;

[0033] Figure 4 Principle diagram of the unmanned aerial vehicle photographing the target motor vehicle from the tail, roof to the head of the present application;

[0034] Figure 5 Principle diagram of the unmanned aerial vehicle photographing the target motor vehicle from the head, roof to the tail of the present application;

[0035] Figure 6 Top view of a single target motor vehicle parked in a target parking space of the present application;

[0036] Figure 7 Partial structure diagram of the street lamp post of the present application;

[0037] Figure 8 Structure diagram of the aerial image acquisition device of the present application;

[0038] Figure 9 Structure diagram of the aerial image acquisition device of the present application after omitting part of components;

[0039] Figure 10 Enlarged view of A of Figure 9

[0040] Figure 11 Another angle structure diagram of the aerial image acquisition device of the present application after omitting part of components;

[0041] Figure 12 Enlarged view of B of Figure 11

[0042] Figure 13 Structure diagram of the unmanned aerial vehicle landing on the charging plate of the present application.

[0043] Marking description: 1, unmanned aerial vehicle; 2, magnetic attraction charging mechanism; 21, charging plate; 23, plug-in port; 24, charging head; 25, support; 26, magnetic attraction block; 27, magnetic attraction groove; 3, buffer mechanism; 31, damping; 32, elastic piece; 33, positioning sliding block; 34, supporting block; 4, parking space; 41, marking line; 5, street lamp column; 51, apron; 52, illuminating lamp; 53, solar panel; 54, rain shield; 55, monitoring equipment; 6, controller. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0045] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application.

[0046] As​​Figures 1-13 As shown, this embodiment provides a drone-based on-street parking space photography system with an intelligent parking platform, including a parking area located on the side of the road, the parking area including multiple parking spaces 4 arranged sequentially along the road extension direction; a street lamp post 5 is located adjacent to the parking space 4, and an aerial image acquisition device is also provided on the street lamp post 5; the aerial image acquisition device includes a drone 1 and a helipad 51. The drone 1 is equipped with a camera module, a rechargeable battery, and a device management module. The device management module is electrically connected to the camera module and the rechargeable battery. The camera module is configured to photograph motor vehicles in parking space 4. When a target motor vehicle is parked in the target parking space, the drone 1 approaches the target motor vehicle and first circles around it in a clockwise or counterclockwise direction. The camera module takes a circumferential view image of the target motor vehicle and transmits it to the device management module. After obtaining the circumferential view image, the drone 1 moves to the front, roof, and rear of the target motor vehicle according to a preset path and takes wide-angle photos at multiple fixed points and transmits them to the device management module. Each of the parking spaces 4 is also equipped with a geomagnetic induction device, which is connected to the device management module. When the target vehicle enters or leaves the target parking space, the geomagnetic induction device sends an induction signal to the device management module, which then controls the drone 1 to fly to the target parking space to take pictures. The apron 51 includes a parking platform and a rain shelter 54, the rain shelter 54 being used to shield the top of the parking platform; the parking platform includes a main body, a magnetic charging mechanism 2, and a buffer mechanism 3; the magnetic charging mechanism 2 is disposed on the main body and includes a charging plate 21; the charging plate 21 is provided with a power outlet 23, and the charging plate 21 is connected to a charging head 24 via a bracket 25, the charging head 24 is provided with a magnetic block 26, and the main body is provided with a corresponding magnetic groove 27; the buffer mechanism 3 is disposed at the bottom of the main body and includes a damper 31 and an elastic element 32, the elastic element 32 being sleeved on the outside of the damper 31, and the bottom of the main body is also provided with a positioning slider 33, the positioning slider 33 being slidably connected to a support block 34; The rain shield 54 allows the drone 1 to charge and be used normally in rainy weather, enhancing the system's outdoor adaptability and stability.

[0047] Firstly, in terms of structure and function, the magnetic charging mechanism 2 in this embodiment is cleverly and practically designed, significantly improving battery life. The magnetic block 26 on the charging head 24 is compatible with the magnetic slot 27 on the main body of the device, enabling a quick and stable magnetic connection between the drone 1 and the charging head 24. This ensures that the drone 1 remains in a fixed position during charging, preventing charging interruptions due to shaking or displacement, and greatly improving the reliability and stability of charging. Simultaneously, the combined design of the power port 23 and the charging plate 21 provides a stable power input path for charging the drone 1, ensuring a smooth charging process.

[0048] Secondly, the design of the buffer mechanism 3 is a major highlight of this embodiment. The combined use of damping 31 and elastic element 32 can effectively absorb the impact force generated when the drone 1 lands. The elastic element 32 (such as a spring) deforms under pressure, converting kinetic energy into elastic potential energy, and then slowly releases it, playing a buffering role; the damping 31 can consume some energy, reduce vibration and rebound, and make the drone 1 land more smoothly. The sliding connection between the positioning slider 33 and the support block 34, as well as the possible guide structure, ensures the stable movement of the main body of the device during the buffering process, preventing it from shifting or tilting, further improving the buffering effect and the stability of the overall structure. Multiple buffer mechanisms 3 are evenly distributed at the bottom of the main body of the device, which can provide uniform buffering force and protect the drone 1 and the parking platform in all directions.

[0049] Moreover, from a user experience perspective, the parking platform provides a safe and stable parking and charging environment for the drone 1. During landing, the buffer mechanism 3 effectively reduces impact, lowers the risk of damage, and extends the drone's lifespan. The stable magnetic charging connection also reduces the need for manual intervention, improves charging efficiency, and allows the drone 1 to recover its power more quickly and be ready for the next mission.

[0050] Furthermore, the drone 1 is equipped with a camera module, enabling it to circle around and photograph the target vehicle, obtaining a 360-degree view of the vehicle. The camera range of this 360-degree view extends from the front to the rear of the target vehicle, encompassing multiple parking spaces 4, providing a wide field of view. Simultaneously, because the drone 1 can perform hovering photography at multiple points, there is no need to configure multiple camera modules, effectively reducing production costs.

[0051] Specifically, the magnetic block 26 is adapted to the magnetic slot 27 to achieve magnetic connection between the drone 1 and the charging head 24.

[0052] Specifically, the elastic element 32 is a spring, and the two ends of the spring are respectively connected to the bottom of the device body and the support block 34. The spring is used to provide a buffering force.

[0053] Specifically, a guide structure is provided between the positioning slider 33 and the support block 34 to ensure the stable movement of the main body of the device during the buffering process.

[0054] Specifically, the main body of the device is provided with a platform area for parking the drone 1, and the platform area corresponds to the position of the magnetic charging mechanism 2.

[0055] Specifically, the stopping platform is provided with a plurality of buffer mechanisms 3, which are evenly distributed at the bottom of the main body of the device to provide a uniform buffering effect.

[0056] Specifically, when the drone 1 takes pictures of the target vehicle, the drone 1 travels around the target vehicle along an arc with the target vehicle as the center; the camera module takes pictures at multiple points in the radial direction along the travel trajectory to obtain a circumferential view of the vehicle. The camera range of the vehicle circumference image extends from the front of the target vehicle to the rear of the target vehicle. When the camera module takes pictures at multiple points, it takes a first vehicle circumference image at the first point. When the drone 1 moves to the second point, the camera module uses the first vehicle circumference image as a reference virtual scene and performs docking and verification with the real scene facing the second point to take a second vehicle circumference image. The shooting is repeated until the last point, and the image obtained at the last point is the vehicle circumference image.

[0057] Specifically, after obtaining the vehicle's circumferential view image, the drone 1 moves along a preset path, starting from the front of the vehicle, passing the roof, and reaching the rear, taking wide-angle photos at multiple fixed points; or, After obtaining the vehicle's circumferential view image, the UAV 1 moves from the rear of the vehicle, over the roof, to the front of the vehicle, following a preset path, and takes wide-angle photos at multiple fixed points. The device management module stitches the wide-angle photos into a complete evidence photo covering the license plate from the front of the vehicle to the rear of the vehicle. The device management module also includes a path planning unit, which controls the UAV 1 to fly along a preset arc path from the front of the vehicle, through the roof of the vehicle, to the rear of the vehicle during flight shooting, and to stop at the front, roof, and rear of the vehicle to take the wide-angle photo.

[0058] like Figure 4 As shown, after the drone 1 circles the target vehicle counterclockwise and obtains a circumferential view of the vehicle, the drone 1 then moves from the rear of the target vehicle to the roof and then to the front of the vehicle according to a preset path, taking wide-angle photos at multiple fixed points and transmitting them to the device management module; or, like Figure 5As shown, after the drone 1 moves clockwise around the target vehicle and obtains a circumferential view of the vehicle, the drone 1 then moves from the front of the target vehicle to the roof and then to the rear of the vehicle according to a preset path, taking wide-angle photos at multiple fixed points and transmitting them to the device management module.

[0059] Specifically, the street lamp post 5 is also equipped with a lighting lamp 52, a solar power panel 53, a storage battery and a main control module; the main control module is electrically connected to the lighting lamp 52, the solar power panel 53 and the storage battery respectively, and the storage battery is used to supply power to each power-consuming module; The power socket 23 is provided with a male terminal, which is electrically connected to the battery through the main control module; the charging head 24 is provided with a female terminal, which is electrically connected to the rechargeable battery through the device management module; when the drone 1 is magnetically connected to the charging head 24, the main control module and the device management module are connected through the cooperation of the male and female terminals to perform charging and / or data transmission. The device management module has a built-in image processing unit and a first storage unit. The vehicle circumferential view image captured by the UAV 1 is directly stored in the first storage unit. The image processing unit is used to stitch the wide-angle photo into the complete evidence photo, and the complete evidence photo is then stored in the first storage unit. The main control module has a built-in second storage unit, which is used to store the vehicle circumferential view image and the complete evidence photo output from the first storage unit.

[0060] Specifically, the drone 1 has a shooting range of 1-4 consecutively arranged parking spaces 4.

[0061] Specifically, the operation process of the magnetic charging mechanism 2 and the buffer mechanism 3 is as follows: Firstly, regarding magnetic charging, the magnetic force between the magnetic block 26 and the magnetic slot 27 enables the drone 1 to connect quickly and accurately with the charging head 24. When the drone 1 lands in a suitable position, the magnetic force makes the two fit tightly together, ensuring the stable progress of the charging process. After the power port 23 is connected to an external power source, the current is transmitted to the drone 1 through the charging plate 21 and the charging head 24 to complete the charging operation. Secondly, regarding adaptive buffering: when the UAV 1 lands and impacts the main body of the device, the elastic element 32 spring is first compressed, undergoes elastic deformation, and absorbs part of the kinetic energy; the damper 31 dissipates energy through internal friction and other means, reducing vibration and rebound; the sliding connection between the positioning slider 33 and the support block 34, as well as the guide structure, ensures the linear motion and stability of the main body of the device during the buffering process, enabling the buffering mechanism 3 to evenly distribute the impact force and protect the UAV 1 and the platform area from damage; through the synergistic effect of the elastic element 32 and the damper 31, adaptive buffering of the impact force of the UAV 1 landing is achieved.

[0062] This embodiment also provides a method for shooting images of a parking space within a single UAV channel using an intelligent parking platform, applied to the aforementioned UAV system for shooting images of a parking space within a single UAV channel. The method includes the following steps: Step S1: When the target vehicle enters the target parking space, the geomagnetic induction device sends the induction signal to the device management module, which controls the drone 1 to fly to the target parking space to take pictures; Step S2: After the target vehicle is parked, the UAV 1 circles around the target vehicle along an arc; the camera module on the UAV 1 takes multiple pictures at multiple points in the radial direction along the circumference of the vehicle to obtain a circumferential view image of the vehicle and stores it in the first storage unit. Step S3: After obtaining the vehicle's circumferential view image, the UAV 1 moves to the front, roof, and rear of the target vehicle according to a preset path, and takes wide-angle photos at multiple fixed points and transmits them to the device management module; the image processing unit in the device management module stitches the wide-angle photos into a complete evidence photo, and then stores the complete evidence photo in the first storage unit. Step S4: The operator remotely operates the controller 6 or the drone 1's own navigation system to control the drone 1 to fly towards the parking platform. The drone 1 gradually approaches the platform area on the main body of the device. As the drone 1 lands, its bottom approaches the magnetic block 26 on the charging head 24. Under the action of magnetic force, the drone 1 falls into the magnetic slot 27 and is magnetically connected to the charging head 24. At this time, the male and female poles are connected to charge and / or transmit data. At the same time, the first storage unit transmits the vehicle circumferential view image and the complete evidence photo to the second storage unit of the main control module. Step S5: When the target vehicle leaves the target parking space, the geomagnetic sensor sends a sensing signal to the device management module. The device management module controls the drone 1 to fly to the target parking space to take a single-point photo, and repeats steps S2-S3.

[0063] Specifically, in step S4, at the instant the UAV 1 lands and contacts the platform area, the UAV 1 generates an impact force on the main body of the device, the main body of the device moves downward, compresses the spring, the spring deforms, and converts the kinetic energy of the UAV 1 into elastic potential energy; at the same time, the damping 31 starts to work; the positioning slider 33 slides along the support block 34, and under the action of the guide structure, the main body of the device remains stable during the buffering process and will not deviate or tilt. Once the drone 1 has finished charging and / or transmitting data, the magnetic connection is disconnected, and the drone 1 takes off and leaves the platform area. At this time, the spring gradually returns to its original shape under its own elastic force, driving the main body of the device back to the initial position, waiting for the next landing of the drone 1.

[0064] Specifically, in step S2, when the camera module takes pictures at multiple points, it takes a picture of the first vehicle's perimeter at the first point; when the drone 1 moves to the second point, the camera module uses the first vehicle perimeter image as a reference virtual scene and performs docking and verification with the real scene facing the second point to take a picture of the second vehicle perimeter image; the shooting process is repeated until the last point, and the image obtained at the last point is the vehicle perimeter image.

[0065] Specifically, marking lines 41 are set on the ground of each parking space 4 according to the junction of the front, roof and rear of the vehicle in the target parking space; More specifically, in step S3, the image processing unit performs the following operations: Step S31: Perform noise reduction and distortion correction on the wide-angle photos taken at the front, roof, and rear of the vehicle, respectively; Step S32: Extract feature points from adjacent wide-angle photos. The feature points include the marking lines 41 set on the ground of the parking space 4. Based on the marking lines 41 and other image features, perform feature matching to determine the stitching position relationship of adjacent photos. Step S33: Based on the determined stitching position relationship, the multiple wide-angle photos are stitched together into a continuous image covering the front to the rear of the vehicle using an image fusion algorithm; Step S34: Identify and extract vehicle features, front and rear license plate information, and parking space number marked on the parking space 4 from the stitched image, associate the parking space number with the vehicle information, and generate a complete evidence photo containing the parking space number, vehicle and license plate. Step S35: Perform brightness equalization and edge smoothing processing on the complete evidence photo.

[0066] Specifically, in step S31, the noise reduction can be achieved through Gaussian filtering or median filtering algorithms; the distortion correction is based on the intrinsic parameter matrix and distortion coefficients calibrated at the camera factory, and the image is corrected through mathematical transformation (such as the Brownian model).

[0067] Specifically, in step S32, the shape of the marking line 41 is at least one of the following: vertical line, horizontal line, L-shape, and T-shape.

[0068] Specifically, in step S34, the identification uses a pre-trained deep learning model (such as a YOLO- or CNN-based object detection model) to detect and identify vehicles, license plates, and parking space numbers. By introducing marker line 41 as a positioning reference, the process of stitching and aligning multi-angle photos is simplified. Combined with a systematic image processing workflow, not only can clear and coherent panoramic images be generated efficiently, but parking space 4 and vehicle information can also be automatically associated to form a closed-loop chain of evidence.

[0069] Specifically, the controller 6 includes a first wireless communication unit, and the drone 1 includes a second wireless communication unit; the first wireless communication unit and the second wireless communication unit are communicatively connected; the wireless communication unit is Bluetooth or Wi-Fi.

[0070] Specifically, the camera module is an image sensor, such as CMOS or CCD.

[0071] Specifically, the lamp post 5 is also equipped with a monitoring device 55, which is installed below the lighting lamp 52.

[0072] Based on the disclosure and teachings of the foregoing specification, those skilled in the art can make changes and modifications to the above embodiments. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and some modifications and changes to the present invention should also fall within the protection scope of the claims of the present invention. Furthermore, although some specific terms are used in this specification, these terms are only for convenience of explanation and do not constitute any limitation on the present invention.

Claims

1. A drone-based on-street parking space photography system with an intelligent parking platform, comprising a parking area located on the side of a road, the parking area including multiple parking spaces arranged sequentially along the road's extension direction; a street lamp post is located adjacent to each parking space, characterized in that, The lamppost is also equipped with an aerial image acquisition device; the aerial image acquisition device includes a drone and an airstrip. The drone is equipped with a camera module, a rechargeable battery, and a device management module. The device management module is electrically connected to the camera module and the rechargeable battery. The camera module is configured to photograph motor vehicles in parking spaces. When a target motor vehicle is parked in a target parking space, the drone approaches the target motor vehicle and first circles around it in a clockwise or counterclockwise direction. The camera module takes a circumferential view image of the target motor vehicle and transmits it to the device management module. After obtaining the circumferential view image, the drone moves to the front, roof, and rear of the target motor vehicle according to a preset path and takes wide-angle photos at multiple fixed points, which are then transmitted to the device management module. Each of the parking spaces is also equipped with a geomagnetic induction device, which is connected to the device management module. When the target vehicle enters or leaves the target parking space, the geomagnetic induction device sends a sensing signal to the device management module, which then controls the drone to fly to the target parking space to take pictures. The apron includes a parking platform and a rain shelter, the rain shelter being used to cover the top of the parking platform; the parking platform includes a main body, a magnetic charging mechanism, and a buffer mechanism; the magnetic charging mechanism is mounted on the main body and includes a charging plate; the charging plate has a power port, and a charging head is connected to the charging plate via a bracket; the charging head has a magnetic block, and a corresponding magnetic groove is provided on the main body; the buffer mechanism is located at the bottom of the main body and includes a damper and an elastic element, the elastic element being sleeved outside the damper; a positioning slider is also provided at the bottom of the main body, the positioning slider being slidably connected to a support block.

2. The UAV-based in-street parking space photography system according to claim 1, characterized in that, The magnetic block is adapted to the magnetic slot to achieve a magnetic connection between the drone and the charging head.

3. The UAV-based in-street parking space photography system according to claim 1, characterized in that, The elastic element is a spring, and the two ends of the spring are respectively connected to the bottom of the device body and the support block. The spring is used to provide a buffering force. The main body of the device is provided with a platform area for parking drones, and the platform area corresponds to the position of the magnetic charging mechanism.

4. The UAV-based in-street parking space shooting system according to claim 1, characterized in that, A guide structure is provided between the positioning slider and the support block.

5. The UAV-based in-street parking space shooting system according to claim 1, characterized in that, The stopping platform is equipped with multiple buffer mechanisms, which are evenly distributed at the bottom of the main body of the device.

6. The UAV-based in-street parking space photography system according to claim 1, characterized in that, When the drone photographs the target vehicle, the drone travels around the target vehicle along an arc with the target vehicle as the center; the camera module takes multiple pictures at multiple points in the radial direction of the vehicle along the trajectory to obtain a circumferential view of the vehicle. The camera range of the vehicle circumference image extends from the front of the target vehicle to the rear of the target vehicle. When the camera module takes pictures at multiple points, it takes a first vehicle circumference image at the first point. When the drone moves to the second point, the camera module uses the first vehicle circumference image as a reference virtual scene and performs docking and verification with the real scene facing the second point to take a second vehicle circumference image. The shooting is repeated until the last point, and the image obtained at the last point is the vehicle circumference image.

7. The UAV-based in-street parking space photography system according to claim 1, characterized in that, After obtaining the vehicle's circumferential image, the drone moves along a preset path, starting from the front of the vehicle, passing the roof, and moving to the rear, taking wide-angle photos at multiple fixed points. or, After obtaining the vehicle's circumferential image, the drone moves along a preset path, starting from the rear of the vehicle, passing the roof, and moving to the front, taking wide-angle photos at multiple fixed points. The device management module stitches the wide-angle photos into a complete evidence photo covering the license plate from the front of the vehicle to the rear of the vehicle. The device management module also includes a path planning unit, which controls the drone to fly along a preset arc path from the front of the vehicle, through the roof, to the rear of the vehicle during flight photography, and to stop at the front, roof, and rear positions to take wide-angle photos.

8. The UAV-based in-street parking space photography system according to claim 7, characterized in that, The streetlight pole is also equipped with a lighting lamp, a solar panel, a battery, and a main control module; the main control module is electrically connected to the lighting lamp, the solar panel, and the battery respectively. The power socket has a male terminal, which is electrically connected to the battery through the main control module; the charging head has a female terminal, which is electrically connected to the rechargeable battery through the device management module; when the drone is magnetically connected to the charging head, the main control module and the device management module are connected through the male and female terminals to perform charging and / or data transmission. The device management module has a built-in image processing unit and a first storage unit. The vehicle surround view image captured by the drone is directly stored in the first storage unit. The image processing unit is used to stitch the wide-angle photo into the complete evidence photo, and the complete evidence photo is then stored in the first storage unit. The main control module has a built-in second storage unit, which is used to store the vehicle surround view image and the complete evidence photo output from the first storage unit.

9. A method for photographing roadside parking spaces using a drone with an intelligent parking platform, characterized in that: The UAV roadside parking space shooting system as described in any one of claims 1 to 8, the UAV roadside parking space shooting method includes the following steps: Step S1: When the target vehicle enters the target parking space, the geomagnetic induction device sends the induction signal to the device management module, which then controls the drone to fly to the target parking space to take pictures. Step S2: After the target vehicle is parked, the drone circles around the target vehicle along an arc; the camera module on the drone takes multiple pictures at multiple points in the radial direction along the circle trajectory to obtain a circumferential view of the vehicle and stores it in the first storage unit. Step S3: After obtaining the vehicle's circumferential view image, the drone moves to the front, roof, and rear of the target vehicle according to a preset path, and takes wide-angle photos at multiple fixed points and transmits them to the device management module; the image processing unit in the device management module stitches the wide-angle photos into a complete evidence photo, and then stores the complete evidence photo in the first storage unit. Step S4: The operator remotely operates the controller or the drone's own navigation system to control the drone to fly towards the parking platform. The drone gradually approaches the platform area on the main body of the device. As the drone lands, its bottom approaches the magnetic block on the charging head. Under the action of magnetic force, the drone falls into the magnetic slot and is magnetically connected to the charging head. At this time, the male and female poles are connected to perform charging and / or data transmission. At the same time, the first storage unit transmits the vehicle circumferential image and the complete evidence photo to the second storage unit of the main control module. Step S5: When the target vehicle leaves the target parking space, the geomagnetic sensor sends a sensing signal to the device management module. The device management module controls the drone to fly to the target parking space to take a single-point photo, and repeats steps S2-S3.

10. The method for photographing unmanned aerial vehicles (UAVs) in roadside parking spaces according to claim 9, characterized in that, In step S4, at the instant the drone lands and contacts the platform area, the drone generates an impact force on the main body of the device, the main body of the device moves downward, compresses the spring, the spring deforms, and converts the kinetic energy of the drone into elastic potential energy; at the same time, the damping starts to work; the positioning slider slides along the support block, and under the action of the guide structure, the main body of the device remains stable during the buffering process and will not deviate or tilt. Once the drone is fully charged and / or the data is transmitted, disconnect the magnetic connection and the drone will take off and leave the platform area. At this point, the spring gradually returns to its original shape under its own elastic force, driving the main body of the device back to its initial position, waiting for the next drone landing.

Images