Idle parking space detection method and device and electronic equipment

Abstract

The invention relates to an idle parking space detection method and device and electronic equipment, and belongs to the technical field of automatic parking, and the detection method comprises the steps: obtaining the parking space area and the parking space width of a to-be-detected parking space through the angular point data of the to-be-detected parking space, determining a proportion threshold value according to the length and the width of a current driving vehicle and the length and the width of a parking lot calibration parking space, a target ratio is obtained according to the parking space area, the parking space width, the vehicle data and the parking space data of the parking lot, if the target ratio is larger than or equal to a proportion threshold value, the to-be-detected parking space is determined to be an idle parking space, and if the target ratio is smaller than the proportion threshold value, the to-be-detected parking space is determined to be parked. And meanwhile, more hardware and computing power support are not needed, the identification efficiency is effectively improved, and the detection cost is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of automatic parking, and in particular to a method, device and electronic equipment for detecting an empty parking space. Background Art

[0002] In the 2020s, automatic parking is a powerful part of smart travel and assisted driving. In terms of robustness and safety, automatic parking needs to be divided into visual perception and radar ranging to run simultaneously in order to park safely, accurately and efficiently.

[0003] In the automatic parking system, how to accurately and efficiently locate the vacant parking space is a key issue. Since the visual perception algorithm based on parking space detection cannot determine whether the parking space is vacant when detecting the corner points of the parking space, the general treatment is to add obstacle detection or ultrasonic radar to assist in detecting whether the parking space is vacant. However, the disadvantages of these two methods are: adding obstacle detection requires a lot of computing power and takes a long time; when using ultrasonic detection, the results of multiple sensors need to be fused, and the fusion algorithm is relatively complex; in summary, although both methods can accurately identify vacant parking spaces, there are problems of high detection cost and low efficiency. Summary of the invention

[0004] To this end, the present invention provides a method, device and electronic equipment for detecting an empty parking space to solve the problems of high detection cost and low efficiency in the prior art.

[0005] In a first aspect, a method for detecting an empty parking space is provided, the method comprising:

[0006] Obtaining the parking space area and parking space width of the parking space to be measured according to the corner point data of the parking space to be measured; the corner point data includes the coordinates of the four corner points of the parking space to be measured;

[0007] A ratio threshold is obtained based on vehicle data and parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located;

[0008] Obtaining a target ratio according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data;

[0009] If the target ratio is greater than or equal to the ratio threshold, the parking space to be tested is determined to be an empty parking space;

[0010] If the target ratio is less than the ratio threshold, the parking space to be tested is determined to be parked.

[0011] Furthermore, obtaining the ratio threshold value based on the vehicle data and the parking lot space data includes:

[0012] According to the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot, a proportional threshold is obtained by a threshold formula, and the threshold formula is:

[0013]

[0014] Among them, k is the ratio threshold; H d is the length of the current driving vehicle; H t is the width of the current driving vehicle; W d W is the length of the parking lot. t Mark the width of the parking spaces in the parking lot.

[0015] Furthermore, the detection method further comprises:

[0016] If the length and width of the calibrated parking space in the parking lot where the current driving vehicle is located cannot be obtained, the length and width of the standard parking space will be used as the length and width of the calibrated parking space in the parking lot respectively.

[0017] Further, obtaining the parking area of ​​the parking space to be measured based on the corner point data of the parking space to be measured includes:

[0018] Get a bird's-eye view of the current frame through the four cameras of the current driving vehicle;

[0019] Identify four corner points of the parking space to be measured in the bird's-eye view by a deep learning algorithm;

[0020] The coordinate system is established with the centroid of the current driving vehicle as the coordinate origin, and the coordinates of the four corner points are A(x1,y1), B(x2,y2), C(x3,y3), and D(x4,y4);

[0021] The parking space area is obtained by using the coordinate area formula according to the coordinates of the four corner points. The coordinate area formula is:

[0022] S=|x1(y2-y4)+x2(y4-y1)+x3(y1-y2)| / 2;

[0023] Wherein, S is the parking space area to be measured.

[0024] Further, obtaining the parking space width of the parking space to be measured based on the corner point data of the parking space to be measured includes:

[0025] The four corner points of the parking space to be measured are connected to form a quadrilateral. If the quadrilateral is a regular quadrilateral, the short side is taken as the width; if the quadrilateral is not a regular quadrilateral, the distance between the two long sides is taken as the width.

[0026] Further, obtaining the target ratio according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data includes:

[0027] If the width of the currently driven vehicle in the vehicle data is greater than the width of the parking space, the target ratio is obtained by a target ratio formula, which is:

[0028]

[0029] Among them, k m is the target ratio; S is the parking space area to be measured; H t W is the length of the parking lot. t Mark the width of parking spaces in the parking lot;

[0030] If the width of the currently driven vehicle in the vehicle data is less than or equal to the width of the parking space, the parking space to be measured is determined to be parked.

[0031] Furthermore, after the parking space to be tested is determined to be an empty parking space, the method further includes:

[0032] Get a bird's-eye view of the current frame through the four cameras of the current driving vehicle;

[0033] Extracting the boundary of the drivable area in the bird's-eye view by a deep learning algorithm to obtain a drivable area for the vehicle;

[0034] If the parking space to be tested is within the drivable area, then the parking space to be tested is a parking space;

[0035] If the parking space to be measured is outside the drivable area, then the parking space to be measured is a non-parking space.

[0036] Furthermore, the step of extracting the boundary of the drivable area in the bird's-eye view by a deep learning algorithm to obtain the drivable area for the vehicle includes:

[0037] The deep learning algorithm adopts the Canny edge detection algorithm.

[0038] In a second aspect, a device for detecting an empty parking space is provided, the device comprising:

[0039] The parking space data acquisition module to be measured is used to obtain the parking space area and parking space width of the parking space to be measured according to the corner point data of the parking space to be measured; the corner point data includes the coordinates of the four corner points of the parking space to be measured;

[0040] A threshold module, used to obtain a proportional threshold according to vehicle data and parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located;

[0041] A target ratio module, used for obtaining a target ratio according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data;

[0042] an empty parking space determination module, configured to determine the parking space to be tested as an empty parking space if the target ratio is greater than or equal to the ratio threshold;

[0043] It is also used to determine that the parking space to be tested is parked if the target ratio is less than the ratio threshold.

[0044] In a third aspect, an electronic device is provided, including:

[0045] at least one processor; and

[0046] a memory communicatively connected to the at least one processor; wherein,

[0047] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform any of the above-mentioned detection methods.

[0048] The present invention adopts the above technical solution and has at least the following beneficial effects:

[0049] Provided are a method, device and electronic device for detecting an empty parking space. The parking space area and parking space width of the parking space to be measured are obtained through the corner point data of the parking space to be measured. A ratio threshold is determined according to the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot. A target ratio is obtained according to the parking space area, parking space width, vehicle data and parking space data in the parking lot. If the target ratio is greater than or equal to the ratio threshold, the parking space to be measured is determined to be an empty parking space. If the target ratio is less than the ratio threshold, the parking space to be measured is determined to be parked. The empty parking space recognition method of the present invention can accurately recognize empty parking spaces without requiring more hardware and computing power support, effectively improving recognition efficiency and reducing detection costs.

[0050] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 is a flow chart of a method for detecting an empty parking space shown in an exemplary embodiment of the present invention;

[0053] Figure 2is a flow chart of a method for detecting an empty parking space shown in another exemplary embodiment of the present invention;

[0054] Figure 3 is a schematic block diagram of a device for detecting an empty parking space according to an exemplary embodiment of the present invention;

[0055] Figure 4 It is a schematic block diagram of an electronic device shown in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] Problems with existing technologies: In an automatic parking system, how to accurately and efficiently locate an empty parking space is a key issue. Since the visual perception algorithm based on parking space detection cannot determine whether a parking space is an empty space when detecting the corner points of a parking space, the general treatment is to add obstacle detection or ultrasonic radar to assist in detecting whether a parking space is an empty space. However, the disadvantages of these two methods are: (1) Adding obstacle detection requires a lot of computing power and takes a long time; (2) When using ultrasonic detection, the results of multiple sensors need to be fused, and the fusion algorithm is relatively complex.

[0058] The embodiments of the present application provide a method, device and electronic device for detecting vacant parking spaces, which can accurately identify vacant parking spaces without the need for more hardware and computing power support, effectively improve the recognition efficiency, reduce the detection cost, enhance the energy-saving performance of the assisted driving system, effectively improve the endurance of the electric vehicle, and provide users with a good riding experience.

[0059] The detection method and device in this application are described below through specific examples.

[0060] See also Figure 1 , Figure 1 is a flow chart of a method for detecting an empty parking space according to an exemplary embodiment of the present invention, see Figure 1 , the method comprising:

[0061] Step S11, obtaining the parking area and parking width of the parking space to be measured according to the corner point data of the parking space to be measured; the corner point data includes the coordinates of the four corner points of the parking space to be measured;

[0062] Step S12, obtaining a ratio threshold value according to the vehicle data and the parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located;

[0063] Step S13, obtaining a target ratio according to the parking space area, parking space width, vehicle data and parking lot parking space data;

[0064] Step S14: If the target ratio is greater than or equal to the ratio threshold, the parking space to be tested is determined to be an empty parking space; if the target ratio is less than the ratio threshold, the parking space to be tested is determined to be parked.

[0065] It should be noted that the technical solution provided in this embodiment can realize free parking space detection in the form of a small program or in the form of an independent application in specific practice, and the applicable scenarios include but are not limited to: automatic parking.

[0066] Specifically, the vacant parking spaces are detected one by one starting from the parking space closest to the currently driven vehicle. When an vacant parking space is detected, the detection is stopped, and the automatic parking system controls the vehicle to automatically park in the vacant parking space.

[0067] It can be understood that the detection method provided in this embodiment obtains the parking space area and parking space width of the parking space to be measured through the corner point data of the parking space to be measured, determines the ratio threshold according to the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot, and obtains the target ratio according to the parking space area, parking space width, vehicle data and parking lot parking space data. If the target ratio is greater than or equal to the ratio threshold, the parking space to be measured is determined to be an empty parking space. If the target ratio is less than the ratio threshold, the parking space to be measured is determined to be parked. The empty parking space identification method of the present invention can accurately identify empty parking spaces without the need for more hardware and computing power support, effectively improving the identification efficiency and reducing the detection cost.

[0068] In specific practice, step S11 "obtaining the parking space area of ​​the parking space to be measured based on the corner point data of the parking space to be measured" includes: obtaining a bird's-eye view of the current frame through the four-way camera of the current driving vehicle; identifying the four corner points of the parking space to be measured in the bird's-eye view through a deep learning algorithm; establishing a coordinate system with the centroid of the current driving vehicle as the coordinate origin to obtain the coordinates of the four corner points, which are A(x1, y1), B(x2, y2), C(x3, y3), and D(x4, y4); obtaining the parking space area through the coordinate area formula based on the coordinates of the four corner points, and the coordinate area formula is: S=|x1(y2-y4)+x2(y4-y1)+x3(y1-y2)| / 2; wherein S is the area of ​​the parking space to be measured.

[0069] It should be noted that the four-way surround view images of the current driving vehicle are processed by Around View Monitor (AVM) to generate a bird's-eye view (BEV); the BEV images are respectively input into the deep learning network for parking corner point detection, the parking corner points are detected, and a coordinate system is established with the centroid of the current driving vehicle as the coordinate origin.

[0070] In one embodiment, the parking space width of the parking space to be measured is obtained according to the corner point data of the parking space to be measured, including: connecting the four corner points of the parking space to be measured to form a quadrilateral, if the quadrilateral is a regular quadrilateral, taking the short side as the width; if the quadrilateral is not a regular quadrilateral, taking the distance between the two long sides as the width.

[0071] It should be noted that if the corner points of the parking space identified by the parking lot are not a parallelogram after connecting the corner points, the parking space is determined as an abnormal parking space, and the next parking space to be tested is monitored.

[0072] In specific practice, step S12 "obtaining a proportional threshold value based on vehicle data and parking space data" includes: obtaining a proportional threshold value through a threshold formula based on the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot, and the threshold formula is:

[0073]

[0074] Among them, k is the ratio threshold; H d is the length of the current driving vehicle; H t is the width of the current driving vehicle; W d W is the length of the parking space marked in the parking lot; t Mark the width of the parking spaces in the parking lot.

[0075] It should be noted that the parking space data of the parking lot is obtained from the parking lot management system when entering the parking lot.

[0076] In one embodiment, if the length and width of the calibrated parking space in the parking lot where the current driving vehicle is located cannot be obtained, the length and width of the standard parking space are used as the length and width of the calibrated parking space in the parking lot respectively.

[0077] It should be noted that the length and width of a standard parking space are the length and width stipulated by the state.

[0078] In specific practice, step S13 "obtaining a target ratio based on the parking space area, parking space width, vehicle data and parking lot parking space data" includes: if the width of the currently driven vehicle in the vehicle data is greater than the parking space width, then obtaining the target ratio by a target ratio formula, the target ratio formula is:

[0079]

[0080] Among them, k m is the target ratio; S is the parking space area to be measured; H t W is the length of the parking space marked in the parking lot; t The width of the parking space is marked for the parking lot; if the width of the currently driven vehicle in the vehicle data is less than or equal to the width of the parking space, the parking space to be tested is determined to be parked.

[0081] It should be noted that when the width of the current driving vehicle is equal to the width of the parking space, the difficulty coefficient of automatic parking will increase. Therefore, the width of the current driving vehicle is multiplied by 120% to obtain the corrected width. If the corrected width is less than or equal to the width of the parking space, the parking space to be tested is determined to be parked.

[0082] In specific practice, after the parking space to be tested is determined to be an empty parking space, a bird's-eye view of the current frame is obtained through the four-way cameras of the currently driven vehicle; the boundary of the drivable area in the bird's-eye view is extracted through a deep learning algorithm to obtain the vehicle's drivable area; if the parking space to be tested is within the drivable area, the parking space to be tested is a parking space; if the parking space to be tested is outside the drivable area, the parking space to be tested is a non-parking space.

[0083] It should be noted that although some parking spaces are detected as vacant, they are in a state where they cannot be reached. In this case, they are marked as non-parking spaces and the next parking space is detected.

[0084] Specifically, the boundary of the drivable area in the bird's-eye view is extracted through a deep learning algorithm to obtain the vehicle drivable area, wherein the deep learning algorithm adopts the Canny edge detection algorithm.

[0085] See also Figure 2 , Figure 2 is a flow chart of a method for detecting an empty parking space shown in another exemplary embodiment of the present invention, see Figure 2 The method includes: performing AVM stitching according to the image information sent back by four cameras to obtain a bird's-eye view of the current frame; identifying the parking space corner points and the vehicle drivable area in the bird's-eye view of the current frame through a deep learning algorithm; calculating the area of ​​all parking spaces in each frame of the image: since the four parking space corner points are irregular polygons, for any polygon, if the coordinates of each vertex A1(x1,y1), A2(x2,y2),…,An(xn,yn) are known, the area S1 of the polygon can be obtained;

[0086] Calculate the standard parking space area S2; the ratio of the parking space area S1 to the standard parking space area S2 is K;

[0087] The calculated ratio threshold is 0.6. If K≤0.6, there is a car parked in the parking space to be tested. If K>0.6, the parking space to be tested is an empty parking space.

[0088] Extract the boundary of the vehicle's drivable area to obtain the vehicle's drivable area. The Canny edge detection algorithm is a multi-stage edge detection algorithm that detects edges in an image through a series of steps. The Canny edge detection algorithm includes three steps: grayscale conversion and Gaussian filtering, gradient calculation and non-maximum suppression, double thresholding, and contour tracking.

[0089] Feature fusion combines the extracted boundary coordinates of the drivable area and the obtained empty parking space corner coordinates to determine whether the corresponding parking space corner is within the drivable area of ​​the vehicle. If the parking space is within the drivable area, it is considered a parking space, otherwise it is a non-parking space.

[0090] See also Figure 3 , Figure 3 is a schematic block diagram of a detection device for an empty parking space according to an exemplary embodiment of the present invention, see Figure 3 , the vacant parking space detection device 100 comprises:

[0091] The parking space data acquisition module 101 is used to obtain the parking space area and parking space width of the parking space to be measured according to the corner point data of the parking space to be measured; the corner point data includes the coordinates of the four corner points of the parking space to be measured;

[0092] The threshold module 102 is used to obtain a ratio threshold according to the vehicle data and the parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located;

[0093] A target ratio module 103, for obtaining a target ratio according to a parking space area, a parking space width, vehicle data and parking lot parking space data;

[0094] The empty parking space determination module 104 is used to determine the parking space to be tested as an empty parking space if the target ratio is greater than or equal to the ratio threshold; and is also used to determine the parking space to be tested as a parked space if the target ratio is less than the ratio threshold.

[0095] It should be noted that the technical solution provided in this embodiment can be applied to scenarios including, but not limited to, automatic parking in specific practice.

[0096] It can be understood that the device provided in this embodiment obtains the parking space area and parking space width of the parking space to be measured through the corner point data of the parking space to be measured, determines the ratio threshold according to the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot, and obtains the target ratio according to the parking space area, parking space width, vehicle data and parking lot parking space data. If the target ratio is greater than or equal to the ratio threshold, the parking space to be measured is determined to be an empty parking space. If the target ratio is less than the ratio threshold, the parking space to be measured is determined to be parked. The empty parking space identification method of the present invention can accurately identify empty parking spaces without the need for more hardware and computing power support, effectively improving the identification efficiency and reducing the detection cost.

[0097] See also Figure 4 , Figure 4 is a schematic block diagram of an electronic device according to an exemplary embodiment of the present invention, see Figure 4 , an electronic device 200, comprising:

[0098] at least one processor 202; and

[0099] A memory 201 is communicatively connected to at least one processor 202; wherein,

[0100] The memory 201 stores instructions that can be executed by at least one processor 202. The instructions are executed by at least one processor 202 so that the at least one processor 202 can perform any of the above detection methods.

[0101] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

[0102] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0103] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for detecting an empty parking space, characterized in that: include: Obtaining the parking space area and parking space width of the parking space to be measured according to the corner point data of the parking space to be measured; The corner point data includes the coordinates of the four corner points of the parking space to be measured; A ratio threshold is obtained based on vehicle data and parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located; Obtaining a target ratio according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data; If the target ratio is greater than or equal to the ratio threshold, the parking space to be tested is determined to be an empty parking space; If the target ratio is less than the ratio threshold, the parking space to be tested is determined to be parked.

2. The detection method according to claim 1, characterized in that: The ratio threshold is obtained based on the vehicle data and the parking lot space data, including: According to the length and width of the currently driven vehicle and the length and width of the calibrated parking space in the parking lot, a proportional threshold is obtained by a threshold formula, and the threshold formula is: Among them, k is the ratio threshold; H d is the length of the current driving vehicle; H t is the width of the current driving vehicle; W d W is the length of the parking lot. t Mark the width of the parking spaces in the parking lot.

3. The detection method according to claim 2, characterized in that: The detection method further comprises: If the length and width of the calibrated parking space in the parking lot where the current driving vehicle is located cannot be obtained, the length and width of the standard parking space will be used as the length and width of the calibrated parking space in the parking lot respectively.

4. The detection method according to claim 3, characterized in that: The step of obtaining the parking area of ​​the parking space to be measured based on the corner point data of the parking space to be measured includes: Get a bird's-eye view of the current frame through the four cameras of the current driving vehicle; Identify four corner points of the parking space to be measured in the bird's-eye view by a deep learning algorithm; The coordinate system is established with the centroid of the current driving vehicle as the coordinate origin, and the coordinates of the four corner points are A(x1,y1), B(x2,y2), C(x3,y3), and D(x4,y4); The parking space area is obtained by using the coordinate area formula according to the coordinates of the four corner points. The coordinate area formula is: S=|x1(y2-y4)+x2(y4-y1)+x3(y1-y2)| / 2; Wherein, S is the parking space area to be measured.

5. The detection method according to claim 4, characterized in that: The step of obtaining the parking space width of the parking space to be measured according to the corner point data of the parking space to be measured comprises: The four corner points of the parking space to be tested are connected to form a quadrilateral. If the quadrilateral is a regular quadrilateral, the short side is taken as the width; if the quadrilateral is not a regular quadrilateral, the distance between the two long sides is taken as the width.

6. The detection method according to claim 3, characterized in that: The target ratio is obtained according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data, including: If the width of the currently driven vehicle in the vehicle data is greater than the width of the parking space, the target ratio is obtained by a target ratio formula, which is: Among them, k m is the target ratio; S is the parking space area to be measured; H t W is the length of the parking space marked in the parking lot; t Mark the width of parking spaces for parking lots; If the width of the currently driven vehicle in the vehicle data is less than or equal to the width of the parking space, the parking space to be measured is determined to be parked.

7. The detection method according to claim 1, characterized in that: After the parking space to be tested is determined to be an empty parking space, the method further includes: Get a bird's-eye view of the current frame through the four cameras of the current driving vehicle; Extracting the boundary of the drivable area in the bird's-eye view by a deep learning algorithm to obtain a drivable area for the vehicle; If the parking space to be tested is within the drivable area, then the parking space to be tested is a parking space; If the parking space to be measured is outside the drivable area, then the parking space to be measured is a non-parking space.

8. The detection method according to claim 7, characterized in that: The method of extracting the boundary of the drivable area in the bird's-eye view by a deep learning algorithm to obtain the drivable area for the vehicle comprises: The deep learning algorithm adopts the Canny edge detection algorithm.

9. A device for detecting an empty parking space, characterized in that: The detection device comprises: The parking space data acquisition module to be measured is used to obtain the parking space area and parking space width of the parking space to be measured according to the corner point data of the parking space to be measured; the corner point data includes the coordinates of the four corner points of the parking space to be measured; A threshold module, used to obtain a proportional threshold according to vehicle data and parking lot space data; the vehicle data includes the length and width of the currently driven vehicle; the parking lot space data is the length and width of the calibrated parking space in the parking lot where the currently driven vehicle is located; A target ratio module, used for obtaining a target ratio according to the parking space area, the parking space width, the vehicle data and the parking lot parking space data; an empty parking space determination module, configured to determine the parking space to be tested as an empty parking space if the target ratio is greater than or equal to the ratio threshold; It is also used to determine that the parking space to be tested is parked if the target ratio is less than the ratio threshold.

10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the detection method according to any one of claims 1 to 8.

Images