Background
With the increasing of private cars, the problem of the vehicles on two sides of a road being parked and placed randomly is very serious, and the urban traffic safety and the urban appearance management are seriously influenced. For a long time, those skilled in the art have tried to solve the problem of random parking and random playing by various methods, but still cannot effectively solve the above problem, which will be described separately below:
the utility model provides a solution is based on earth magnetism parking stall detection occupy the way parking management, and earth magnetism parking stall detection is only limited to whether there is the car in the current parking stall of detection, does not know nevertheless that the current parking stall deposits the license plate number in the vehicle, consequently can't correspond license plate and parking stall, because the roadside parking area is open parking, does not have special entry and export. In addition, because only part of car owners finish the registration in the parking system, and a considerable part of car owners do not carry on the registration in the system, when the user who has not registered in the system drives into the current parking space, the geomagnetic parking space detector of the current parking space can not obtain the identity information of the user, so that the parking management system can not automatically identify the identity of the car owner and realize automatic fee deduction. Based on this, more personnel need to be equipped to supervise or inspect on the road, for example, a method that one worker is responsible for 20 parking spaces is adopted, when a vehicle leaves a parking space, a manager carries out charging management on the vehicle which is driven out, and the method causes great labor cost, so that the labor cost becomes a heavy burden for a parking management company.
The other solution is based on the pure vision road occupation parking management, a plurality of cameras are installed on the vehicle, some cameras are used for identifying the vehicle on the parking space, other cameras are used for identifying the parking space number nameplate on the parking space, and then the license plate and the parking space are corresponding. The method has the defects that the parking space number nameplate is attached to the ground and is easily stained or shielded, so that the identification of the parking space number is influenced, and the operation of the whole system is influenced.
In summary, at present, there is no system for automatically identifying the license plate number and automatically aligning the license plate number with the parking space, so that the automation of parking charge management cannot be realized, and the dilemma of great labor management cost of the current parking management company cannot be eliminated.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides a parking space patrol system based on a positioning technology and a method thereof, and solves the problem that the existing management method depends on manpower to position license plates and parking spaces, so that the labor cost is high.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
a parking space patrol system based on a positioning technology comprises an intelligent patrol vehicle and a vehicle positioning server; the intelligent patrol vehicle is used for acquiring absolute coordinates of the intelligent patrol vehicle and relative coordinates between the intelligent patrol vehicle and the vehicle to be positioned, and uploading the two sets of coordinate data to the vehicle positioning server, and the vehicle positioning server judges whether the vehicle to be positioned is on a parking space according to the two sets of coordinate data.
The vehicle positioning server comprises a parking space map and a vehicle positioning algorithm; the parking space map comprises absolute coordinates of each parking space which is calibrated in advance, and the vehicle positioning algorithm is used for judging whether the position of the vehicle to be positioned is on the parking space, on which parking space or near which parking space.
The intelligent patrol vehicle comprises a global satellite positioning system, a milemeter, a camera, a positioning module, a power module and a communication module; the power supply module supplies power to the global satellite positioning system, the odometer, the camera, the positioning module and the communication module; the odometer is used for the absolute coordinates of the intelligent patrol vehicle; the global satellite positioning system is used for correcting the absolute coordinates of the odometer; the camera is used for photographing an image of a vehicle to be positioned; the positioning module is used for determining the absolute coordinates of the intelligent patrol vehicle and the relative coordinates of the shot vehicle and the patrol vehicle, and uploading the absolute coordinates and the relative coordinates to the vehicle positioning server.
The positioning module comprises an inspection vehicle absolute positioning module and an inspection vehicle relative position module; the patrol vehicle absolute positioning module comprises an absolute coordinate correction information module for receiving a global satellite positioning system and a module for receiving odometer relative coordinate accumulated information, when the patrol vehicle absolute positioning module cannot receive global satellite positioning system data, the received odometer relative coordinate accumulated information and the last global satellite positioning system absolute coordinate are used as current absolute positioning data of the patrol vehicle, and when the patrol vehicle absolute positioning module can receive the global satellite positioning system data, the current received global satellite positioning system data is used as the absolute positioning data of the patrol vehicle.
The odometer adopts one of the following laser odometer, inertial navigation odometer and visual odometer or the combination of the laser odometer, the inertial navigation odometer and the visual odometer.
Patrol and examine car relative positioning module include: the system comprises a license plate-to-camera distance algorithm module, a license plate-to-vehicle body distance algorithm module and a license plate absolute coordinate algorithm module; the algorithm for the distance between the license plate and the vehicle body comprises a camera-to-vehicle body distance algorithm and a license plate-to-vehicle body distance algorithm.
A parking space inspection method based on a positioning technology comprises the following steps:
step 1, a vehicle positioning server establishes a parking space map;
step 2, the intelligent inspection vehicle acquires relative coordinates between the license plate and the intelligent inspection vehicle;
step 3, the intelligent inspection vehicle obtains absolute coordinates of the intelligent inspection vehicle;
step 4, the intelligent inspection vehicle uploads the relative coordinates obtained in the step 2 and the absolute coordinates obtained in the step 3 to a vehicle positioning server;
and 5, calculating by the vehicle positioning server to obtain the absolute coordinates of the license plate, and judging whether the vehicle to be positioned is in the parking space according to the parking space map.
And the step 2 is realized by sequentially adopting a camera calibration algorithm, a coordinate inverse calculation algorithm, a depth of field calculation algorithm and a license plate position algorithm.
And the step 3 is realized by combining an odometer and a global satellite positioning system.
The invention has the advantages and positive effects that:
1. the invention overcomes the traditional prejudice that the license plate number and the parking space number are respectively identified and then are corresponding, and adopts a reverse calculation method which is not easily thought by the technical personnel in the field, namely, a method for corresponding the license plate and the parking space by firstly positioning the relative coordinate between the license plate and the patrol car, then positioning the absolute coordinate of the license plate through the absolute coordinate of the patrol car, and finally taking the absolute coordinate of the license plate to a positioning server to be matched with the parking space map coordinate of the positioning server. Through verification, the detection precision of the invention reaches the centimeter level, and the invention can completely replace a pure visual license plate and parking space corresponding method, thereby reporting the parking compliance condition of the car owner to relevant departments as the evidence of honesty check, finally realizing the unmanned office of parking management, and greatly saving the labor cost of a parking management company.
2. The invention overcomes the technical problems in the field for a long time, namely the technical problem of obtaining the absolute coordinates of the license plate and the problem of automatically corresponding the license plate and the parking space: the relative coordinates between the license plate and the inspection vehicle are obtained through a camera calibration algorithm, a coordinate back calculation algorithm, a depth of field calculation algorithm and a license plate position calculation algorithm, and then the odometer and the global satellite positioning system are skillfully combined, so that the defect that signals of the global satellite positioning system cannot be received in certain areas is overcome, the accumulated error of the odometer is corrected by the global satellite positioning system, the absolute coordinates of the inspection vehicle are obtained, and the absolute coordinates of the license plate are finally obtained, so that the first technical problem is solved, and the second technical problem is solved by establishing a parking space map and automatically matching the absolute coordinates of the license plate with the parking space map.
3. The invention organically combines the global satellite positioning system, the odometer, the camera, the absolute positioning module and the relative positioning module of the intelligent patrol vehicle with the three aspects of judging whether the vehicle is at the parking space module and the vehicle to be positioned by the positioning server, and the parts are mutually supported and interdependent after combination, thereby solving the new technical problem of automatically corresponding the license plate and the parking space and obtaining the new technical effect of accurately aligning the license plate and the parking space.
Detailed Description
The following describes the embodiments of the present invention in detail with reference to the accompanying drawings.
A parking space patrol inspection system based on a positioning technology is shown in figure 1 and comprises an intelligent patrol inspection vehicle and a vehicle positioning server; the intelligent patrol vehicle is used for acquiring absolute coordinates of the intelligent patrol vehicle and relative coordinates between the intelligent patrol vehicle and the vehicle to be positioned, and uploading the two sets of coordinate data to the vehicle positioning server, and the vehicle positioning server judges whether the vehicle to be positioned is on a parking space according to the two sets of coordinate data.
As shown in fig. 1 and 2, the vehicle positioning server includes a parking space map and a vehicle positioning algorithm. The parking space map comprises absolute coordinates of each parking space which is calibrated in advance, and the vehicle positioning algorithm judges whether the position of the vehicle to be positioned is on the parking space, on which parking space or near which parking space.
As shown in fig. 3, the intelligent patrol vehicle comprises a global satellite positioning system, a milemeter, a camera, a positioning module, a power module and a communication module. The power supply module supplies power to the global satellite positioning system, the odometer, the camera, the positioning module and the communication module; the odometer is used for the absolute coordinates of the intelligent patrol vehicle; the global satellite positioning system can adopt a GPS module or a Beidou module to correct the absolute coordinates of the odometer; the camera is used for photographing an image of a vehicle to be positioned; the positioning module is used for intelligently inspecting the absolute coordinates of the vehicle and the relative coordinates of the shot vehicle and the inspection vehicle, and uploading the absolute coordinates and the relative coordinates to the vehicle positioning server.
As shown in fig. 3, the odometer adopts one of the following laser odometer, inertial odometer and visual odometer or their combination.
As shown in fig. 4-5, the positioning module includes an absolute positioning module and a relative position module; the patrol vehicle absolute positioning module comprises an absolute coordinate correction information module for receiving a global satellite positioning system and a module for receiving odometer relative coordinate accumulated information, when the patrol vehicle absolute positioning module cannot receive global satellite positioning system data, the received odometer relative coordinate accumulated information and the last global satellite positioning system absolute coordinate are used as current absolute positioning data of the patrol vehicle, and when the patrol vehicle absolute positioning module can receive the global satellite positioning system data, the current received global satellite positioning system data is used as the absolute positioning data of the patrol vehicle.
As shown in fig. 6, the relative positioning module of the inspection vehicle comprises: the system comprises a license plate-to-camera distance algorithm module, a license plate-to-vehicle body distance algorithm module and a license plate absolute coordinate algorithm module; the algorithm for the distance between the license plate and the vehicle body is divided into a camera-to-vehicle body distance algorithm and a license plate-to-vehicle body distance algorithm.
The invention also provides a routing inspection method of the parking space routing inspection system based on the positioning technology, which has the design idea that: obtaining the relative coordinates from the license plate to the inspection vehicle through a correlation algorithm, obtaining the absolute coordinates of the inspection vehicle through a milemeter and a global satellite positioning system, and finally obtaining the absolute coordinates of the license plate through the absolute coordinates of the inspection vehicle and the relative coordinates from the license plate to the inspection vehicle.
A positioning method based on the roadside parking space patrol inspection system comprises the following steps:
step 1, a vehicle positioning server establishes a parking space map.
The parking space map is a parking space map which is established in a positioning server database after each parking space is calibrated in advance. The parking space map records the position of each parking space in the world coordinate system, also called absolute coordinates.
The parking space calibration method can be various, and the method for calibrating the parking space in the embodiment comprises the following steps: firstly, a movable calibration vehicle and at least 3 reflectors for positioning the position of the vehicle are arranged in the environment, and the position of the movable calibration vehicle under a world coordinate system is known at any time; secondly, at least 2 light reflecting plates are distributed on each parking space (the two light reflecting plates are distributed on each parking space according to a certain rule), and the two light reflecting plates are used for calculating the relative distance between the light reflecting plates and a calibrated vehicle; and thirdly, obtaining the coordinates of the parking space reflector under a world coordinate system through calibrating the absolute coordinates of the vehicle and the relative coordinates from the parking space reflector to the calibrated vehicle, thereby obtaining the coordinates of each parking space under the world coordinate system, and completing the calibration of each parking space.
And 2, the intelligent inspection vehicle acquires relative coordinates between the license plate and the intelligent inspection vehicle.
The key point and difficulty of the patent is to obtain and calculate the relative coordinates from the license plate to the inspection vehicle.
The relative coordinates between the license plate and the inspection vehicle are obtained through a camera calibration algorithm, a coordinate back calculation algorithm, a depth of field calculation algorithm and a license plate position calculation algorithm, and the specific method is as follows:
1. distance (depth of field) algorithm from license plate to camera:
(1) camera (video camera) calibration algorithm:
internal reference: shooting and collecting more than 10 images from different angles as data, and adopting a plane calibration method; calibrating camera internal parameters, wherein an internal parameter matrix is a 3 multiplied by 3 matrix;
external ginseng: namely the coordinates of the camera in a vehicle body coordinate system (the center of the vehicle body is the original point, the direction of the vehicle head is the positive direction of the axis), the height from the camera to the ground, the depression angle beta and the included angle theta between the direction of the camera and the positive direction of the x axis of the vehicle body;
(2) and (3) coordinate back calculation algorithm: the distance from the license plate to the camera is calculated by the image information shot by the camera and the camera calibration parameters (different camera calibration parameters are different).
Setting an internal reference matrix as:
pixel (u, v) in the matrix, then the calculation formula of this pixel in camera coordinates (note: camera coordinate system takes lens optical center as origin, principal axis is Z-axis, x-axis is parallel to image plane x-axis) (plane algorithm expressed by camera Z-axis ═ 1):
through the algorithm, the coordinates of the four vertexes of the license plate in the camera are calculated, and the size of the license plate in a camera coordinate system can be calculated.
(3) The depth of field calculation method comprises the following steps:
setting the length of an actual license plate as length and the width as width, wherein the length of the license plate in a camera coordinate system is L and the width is W; because the license plate and the camera have angular offset, the position of the license plate is adjusted and the depth of field is calculated according to a depression angle beta between the camera in the external reference and the ground and an included angle theta between the orientation of the camera and the positive direction of the x axis of the vehicle body, as shown in fig. 7.
The depth of field formula of the vehicle license plate after being straightened in the camera coordinates is as follows:
2. calculating the position of the license plate in the vehicle body coordinate system:
(1) the coordinates of the camera relative to the center of the vehicle body can be calculated through external parameters calibrated by the camera;
(2) according to the external reference: the depression angle beta between the camera and the ground and the depth of field can be calculated by a trigonometric function to obtain a base, namely the horizontal distance between the license plate and the camera: z is a radical of0
The license plate center is located at (x) in the coordinate system of the camerac,yc,zc);
Wherein the center x of the license plate0、y0The calculation method of (2): calculating the center of the rectangle according to the positions of four vertexes of the license plate under a camera coordinate system;
since the camera coordinates are left-handed coordinates, the conversion to the right-handed coordinates is (z)0,x0,-y0);
And then calculating the position of the license plate position relative to the vehicle body center through the relative positions of the vehicle body center and the camera calibrated by the external reference, and setting: the position of the vehicle camera in the vehicle body coordinate system is (x)c,yc,zc)
The specific conversion formula is as follows:
and 3, the intelligent patrol vehicle acquires absolute coordinates of the intelligent patrol vehicle.
The step skillfully combines the odometer and the global satellite positioning system, not only makes up for the defect that signals of the global satellite positioning system cannot be received in certain areas, but also corrects the accumulated error of the odometer by utilizing the global satellite positioning system and acquires the absolute coordinates of the inspection vehicle. The function of converting the coordinates of the license plate under a vehicle body coordinate system into the coordinates under a world coordinate system is realized, and the specific method comprises the following steps:
setting: the GPS position of the vehicle body center is (x, y, gamma)w
The position of the license plate under the central coordinate of the vehicle body is (x, y, z)c
The conversion calculation formula is:
obtaining the coordinates of the license plate in a world coordinate system as (x)G,yG) Namely the absolute coordinates of the intelligent patrol car.
And 4, uploading the relative coordinates obtained in the step 2 and the absolute coordinates obtained in the step 3 to a vehicle positioning server by the intelligent patrol vehicle.
And 5, calculating by the positioning server to obtain the absolute coordinates of the license plate, and judging whether the vehicle to be positioned is in the parking space according to the parking space map.
Nothing in this specification is said to apply to the prior art.
It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.