Parking space detection method and device, electronic equipment and storage medium

Abstract

The invention relates to a parking space detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a look-around image of a vehicle, and enabling the look-around image to represent a panoramic top view with the vehicle as the center; angular point information and entrance line information are obtained based on the all-round view image, the angular point information represents related information of angular points in the all-round view image, and the entrance line information represents related information of an entrance line of a parking space in the all-round view image; determining a first entrance line and two first angular points which are matched with the first entrance line and have different positions based on the angular point information and the entrance line information; and determining a detection result of the parking space according to the first angular point and / or the first entrance line. According to the embodiment of the invention, parking space detection can be accurately and quickly completed through the parking space entrance line and the parking space angular point.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving, and in particular to a parking space detection method and device, an electronic device, and a storage medium. Background Art

[0002] The automatic parking system can obtain complex environmental information around the vehicle through various sensors, obtain effective empty parking space information through relevant methods, and plan the parking path based on parking space information, environmental information, etc., so that the vehicle can automatically and accurately park in the parking space. The system can effectively avoid collision accidents caused by human operation errors, reduce the problem of drivers being unable to park in complex and crowded environments due to insufficient parking experience, and reduce the problem of parking space waste or traffic congestion caused by improper parking, which can enhance parking safety and driving comfort.

[0003] One of the key issues to be solved in the automatic parking system is how to correctly detect empty parking spaces. There are currently two methods that can be fully automated. One is to determine empty parking spaces by identifying the free space between adjacent vehicles. However, when there are no adjacent vehicles, parking space detection cannot be achieved, and its application is limited. The other is to detect parking spaces through image features, including detecting parking space lines and detecting parking space corners. The existing methods based on parking space lines require complex post-processing, which is time-consuming and inaccurate. The methods based on parking space corners also generate parking spaces by post-processing, and cannot classify the occupancy status of parking spaces. The methods that directly detect complete parking spaces have complex models, are not suitable for low-computing power platforms, and are not particularly accurate than other methods. Therefore, how to accurately and quickly detect parking spaces has become a problem that needs to be solved urgently. Summary of the invention

[0004] In view of this, the present disclosure proposes a parking space detection method and device, an electronic device and a storage medium, which can accurately and quickly complete the parking space detection through the parking space entrance line and the parking space corner point.

[0005] According to one aspect of the present disclosure, a parking space detection method is provided, the method comprising: acquiring a surround view image of a vehicle, the surround view image representing a panoramic overhead view centered on the vehicle; acquiring corner point information and entrance line information based on the surround view image, the corner point information representing relevant information of the corner points in the surround view image, and the entrance line information representing relevant information of the entrance line of the parking space in the surround view image; determining a first entrance line and two first corner points of different positions matching the first entrance line based on the corner point information and the entrance line information; and determining a detection result of the parking space according to the first corner point and / or the first entrance line.

[0006] In a possible implementation, acquiring the surround view image of the vehicle includes: acquiring a plurality of fisheye lens images by means of a plurality of image acquisition devices installed on the vehicle, the plurality of image acquisition devices corresponding to different orientations, and the plurality of image acquisition devices being able to cover the surroundings of the vehicle; performing dedistortion processing on the plurality of fisheye lens images to obtain a plurality of corrected images; performing top-down change processing on the plurality of corrected images to obtain a plurality of corresponding top-down images; and performing splicing and fusing the plurality of top-down images to obtain the surround view image.

[0007] In a possible implementation, the stitching and fusing of the multiple overhead view images to obtain the surround view image includes: weighting the multiple overhead view images, and stitching and fusing the weighted multiple overhead view images to obtain the surround view image, wherein the weight of the overhead view images corresponding to the left and right positions of the vehicle is greater than the weight of the overhead view images corresponding to the front and rear positions of the vehicle.

[0008] In one possible implementation, the corner point information includes the corner point position, corner point confidence, orientation angle, shape and corner point type; the entrance line information includes the entrance line center point position, entrance line confidence, entrance line length, entrance line angle, parking space type where the entrance line is located and whether the parking space corresponding to the entrance line is occupied, wherein the corner point type and the parking space type where the entrance line is located include right angles and non-right angles.

[0009] In a possible implementation, the determining of the first entrance line and two first corner points of different positions matching the first entrance line based on the corner point information and the entrance line information includes: selecting corner points whose corner point confidence is greater than a first preset confidence as second corner points based on the corner point information, and selecting entrance lines whose entrance line confidence is greater than a second preset confidence as second entrance lines based on the entrance line information; selecting an entrance line from the second entrance lines as the first entrance line, and selecting a second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line from the second corner points based on the corner point information and the entrance line information as the third corner point; obtaining the two end point positions of the first entrance line based on the entrance line center point position, the entrance line length and the entrance line angle of the first entrance line; and selecting the corner point of the third corner point whose distance from the two end point positions of the first entrance line is less than a preset value as the first corner point.

[0010] In a possible implementation manner, determining the detection result of the parking space according to the first corner point and / or the first entrance line includes: determining the position of the parking space according to the corner point position and the orientation angle of the first corner point.

[0011] In a possible implementation manner, determining the detection result of the parking space according to the first corner point and / or the first entrance line includes: determining the type of the parking space according to the first entrance line.

[0012] In a possible implementation, determining the type of the parking space based on the first entrance line includes: when the parking space type where the entrance line of the first entrance line is located is non-right angle, determining the parking space as an inclined parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is greater than a calibration value, determining the parking space as a parallel parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is less than the calibration value, determining the parking space as a vertical parking space.

[0013] In a possible implementation manner, the method further includes: searching for preset parameters of the parking space according to the type of the parking space, where the preset parameters include a depth of the parking space.

[0014] In a possible implementation, determining the detection result of the parking space based on the first corner point and / or the first entrance line includes: determining whether the parking space is occupied based on the parking space occupancy result of the first entrance line; or determining whether the parking space is occupied based on a radar detection result.

[0015] According to one aspect of the present disclosure, a parking space detection device is provided, the device comprising: a first acquisition module, used to acquire a surround view image of a vehicle, the surround view image representing a panoramic overhead view centered on the vehicle; a second acquisition module, used to acquire corner point information and entrance line information based on the surround view image, the corner point information representing relevant information of the corner points in the surround view image, and the entrance line information representing relevant information of the entrance line of the parking space in the surround view image; a first determination module, used to determine a first entrance line and two first corner points of different positions matching the first entrance line based on the corner point information and the entrance line information; and a second determination module, used to determine a detection result of the parking space according to the first corner point and / or the first entrance line.

[0016] In a possible implementation, the first acquisition module is further used to: acquire multiple fisheye lens images through multiple image acquisition devices installed on the vehicle, the multiple image acquisition devices correspond to different orientations, and the multiple image acquisition devices can cover the surroundings of the vehicle; perform dedistortion processing on the multiple fisheye lens images to obtain multiple corrected images; perform top-down change processing on the multiple corrected images to obtain corresponding multiple top-view images; and perform splicing and fusion of the multiple top-view images to obtain the surround view image.

[0017] In a possible implementation, the stitching and fusing of the multiple overhead view images to obtain the surround view image includes: weighting the multiple overhead view images, and stitching and fusing the weighted multiple overhead view images to obtain the surround view image, wherein the weight of the overhead view images corresponding to the left and right positions of the vehicle is greater than the weight of the overhead view images corresponding to the front and rear positions of the vehicle.

[0018] In one possible implementation, the corner point information includes the corner point position, corner point confidence, orientation angle, shape and corner point type; the entrance line information includes the entrance line center point position, entrance line confidence, entrance line length, entrance line angle, parking space type where the entrance line is located and whether the parking space corresponding to the entrance line is occupied, wherein the corner point type and the parking space type where the entrance line is located include right angles and non-right angles.

[0019] In a possible implementation, the first determination module is also used to: screen out corner points whose corner point confidence is greater than a first preset confidence as second corner points based on the corner point information, and screen out entrance lines whose entrance line confidence is greater than a second preset confidence as second entrance lines based on the entrance line information; select an entrance line from the second entrance lines as the first entrance line, and select a second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line from the second corner points based on the corner point information and the entrance line information as the third corner point; obtain the two end point positions of the first entrance line based on the entrance line center point position, entrance line length and entrance line angle of the first entrance line; and use the corner point in the third corner point whose distance from the two end point positions of the first entrance line is less than a preset value as the first corner point.

[0020] In a possible implementation manner, the second determining module is further configured to determine the position of the parking space according to the corner point position and the orientation angle of the first corner point.

[0021] In a possible implementation manner, the second determination module is further configured to: determine the type of the parking space according to the first entrance line.

[0022] In a possible implementation, determining the type of the parking space based on the first entrance line includes: when the parking space type where the entrance line of the first entrance line is located is non-right angle, determining the parking space as an inclined parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is greater than a calibration value, determining the parking space as a parallel parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is less than the calibration value, determining the parking space as a vertical parking space.

[0023] In a possible implementation manner, the device further includes: a parameter search module, configured to search for preset parameters of the parking space according to the type of the parking space, wherein the preset parameters include the depth of the parking space.

[0024] In a possible implementation, the second determination module is further configured to: determine whether the parking space is occupied according to a parking space occupancy result of the first entrance line; or determine whether the parking space is occupied according to a radar detection result.

[0025] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to call the instructions stored in the memory to execute the above method.

[0026] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, and the computer program instructions implement the above method when executed by a processor.

[0027] In the disclosed embodiment, the entrance line information and parking space corner point information of the parking spaces around the vehicle are obtained through the surround view images around the vehicle, and the parking space is determined through the detected entrance line information and corner point information, so as to accurately and quickly complete the parking space detection. Specifically, after the parking space detection is completed, the parking space detection results such as the position, occupancy status, parking space type and parking space depth of the detected parking space can be obtained.

[0028] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present disclosure. Other features and aspects of the present disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0030] Figure 1 A flow chart of a parking space detection method according to an embodiment of the present disclosure is shown.

[0031] Figure 2a-2c A parking space classification diagram according to an embodiment of the present disclosure is shown.

[0032] Figure 3a-3d A parking space corner point shape classification diagram according to an embodiment of the present disclosure is shown.

[0033] Figure 4 A schematic structural diagram of a parking space detection device according to an embodiment of the present disclosure is shown.

[0034] Figure 5A block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0035] Figure 6 A block diagram of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0036] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0037] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0038] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0039] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0040] One of the key issues to be solved in the automatic parking system is how to correctly detect empty parking spaces. The relevant technologies can be divided into fully automated methods and non-automated methods. The non-automated methods can be implemented through infrastructure, for example, ultrasonic sensors are placed on the parking spaces to determine the occupancy status of the parking spaces, but they cannot be automated. The automated methods, which determine the parking space status by identifying the free space between adjacent vehicles, are limited by the usage scenario, that is, accurate detection can only be achieved when the adjacent vehicles are in the right position, but parking space detection cannot be achieved without adjacent vehicles.

[0041] In order to solve the above technical problems, the embodiments of the present disclosure propose a parking space detection method, which can accurately and quickly complete the parking space detection through the parking space entrance line and the parking space corner point. Figure 1 FIG. 2 is a flow chart showing a parking space detection method according to an embodiment of the present disclosure. Figure 1As shown, the parking space detection method may include:

[0042] Step S11, acquiring a surround view image of the vehicle.

[0043] The surround view image represents a panoramic bird's-eye view centered on the vehicle.

[0044] Step S12, acquiring corner point information and entry line information based on the surround view image.

[0045] The corner point information represents relevant information of the corner points in the surround view image, and the entrance line information represents relevant information of the entrance line of the parking space in the surround view image.

[0046] Step S13: determining a first entry line and two first corner points at different positions that match the first entry line based on the corner point information and the entry line information.

[0047] Step S14: determining a detection result of a parking space according to the first corner point and / or the first entrance line.

[0048] In the disclosed embodiment, the entrance line information and parking space corner point information of the parking spaces around the vehicle are obtained through the surround view images around the vehicle, and the parking space is determined through the detected entrance line information and corner point information, so as to accurately and quickly complete the parking space detection. Specifically, after the parking space detection is completed, the parking space detection results such as the position, occupancy status, parking space type and parking space depth of the detected parking space can be obtained.

[0049] In step S11 , a surround view image of the vehicle may be acquired, wherein the surround view image represents a panoramic bird's-eye view centered on the vehicle.

[0050] In a possible implementation, acquiring the surround view image of the vehicle includes: acquiring a plurality of fisheye lens images by means of a plurality of image acquisition devices installed on the vehicle, the plurality of image acquisition devices corresponding to different orientations, and the plurality of image acquisition devices being able to cover the surroundings of the vehicle; performing dedistortion processing on the plurality of fisheye lens images to obtain a plurality of corrected images; performing top-down change processing on the plurality of corrected images to obtain a plurality of corresponding top-down images; and performing splicing and fusing the plurality of top-down images to obtain the surround view image.

[0051] In the disclosed embodiment, multiple image acquisition devices need to be installed on the vehicle, and surround view images are acquired through the image acquisition devices, wherein the main device in the image acquisition device is the camera. Among the existing available cameras, the fisheye camera often has a larger viewing angle than other cameras. Therefore, in the disclosed embodiment, fisheye cameras are used for acquisition, that is, multiple fisheye cameras are arranged around the vehicle, and the environment in front, behind, left and right of the vehicle is photographed through the fisheye cameras, and then the captured fisheye lens images are dedistorted, changed from a bird's-eye view, and spliced ​​and fused according to relevant algorithms to obtain surround view images containing information about the vehicle's surrounding environment. In the disclosed embodiment, there is no restriction on the number of image acquisition devices, and the fisheye lens images acquired by multiple image acquisition devices can cover the surroundings of the vehicle. In one example, four image acquisition devices can be installed on the vehicle. In another example, eight image acquisition devices can be installed on the vehicle.

[0052] The lens of a fisheye camera is a lens with a focal length of 16mm (millimeter) or shorter and a wide viewing angle. It has the characteristics of short focal length and large field of view. The focal length of ordinary cameras is usually between 30 degrees and 60 degrees, which does not meet the requirements of the surround view image in the automatic parking system, so a fisheye camera with a larger field of view is used. Fisheye cameras have more lenses than ordinary cameras, and the light is refracted multiple times to obtain a larger imaging angle. However, because the images taken by fisheye cameras will be distorted and have a certain degree of distortion, the images taken by fisheye cameras need to be dedistorted.

[0053] Fisheye cameras are composed of a combination of multiple lenses, which makes their imaging model nonlinear and the captured images have large distortion. There are three main types of distortion: radial distortion, tangential distortion, and thin prism distortion. Because the effect of thin prism distortion on fisheye camera imaging is smaller than radial distortion and tangential distortion, thin prism distortion caused by defects in camera lens design or errors in processing and installation is usually ignored.

[0054] Radial distortion refers to the distortion distributed along the radius of the lens due to the shape of the lens. The reason for the distortion is that the area near the edge of the lens is more curved than the area near the center of the lens. It mainly includes barrel distortion and pincushion distortion. The distortion at the center of the imaging optical axis is 0, and the closer to the edge, the more serious the distortion. The adjustment formula for any point on the imaging surface is:

[0055]

[0056] Among them, (x, y) is the coordinate of the point on the imaging surface, (x corrected ′,y corrected ′) is the coordinate of the point after radial distortion correction, r represents the distance from the actual point to the optical center, k1, k2, k3 are the first-order, second-order and third-order coefficients respectively.

[0057] Tangential distortion refers to the distortion caused by errors in the camera installation process that make the lens and the imaging plane non-parallel. Tangential distortion will make some points on the image appear closer when observing. The adjustment formula for the fisheye camera lens plane for any point on the imaging plane is:

[0058]

[0059] Among them, (x, y) is also the coordinate of the point on the imaging surface, (x corrected ″,y corrected ″) is the coordinate of the point after tangential distortion correction, r represents the distance from the actual point to the optical center, and p1 and p2 are the tangential distortion coefficients.

[0060] Therefore, the adjustment formula for dedistorting the fisheye lens image obtained by the fisheye camera can be:

[0061]

[0062] After dedistortion, the distortion-corrected image is an oblique view with a certain angle to the ground, because the optical axis of the camera has a certain angle with the ground. If such an oblique view is stitched together, the resulting image is inconsistent with the normal visual image, and cannot intuitively and accurately present the environmental information around the vehicle. Therefore, it is necessary to perform a top-down change process on the distortion-corrected image to obtain a top-down view, and then stitch and fuse multiple images to merge them to obtain a surround view image containing the surrounding environment information.

[0063] In the disclosed embodiment, the top view change processing can be completed by inverse perspective transformation, which refers to projecting an image onto a new image plane, also known as "projection transformation". In the process of camera imaging, perspective transformation is a mapping transformation of an image matrix, and inverse perspective transformation can be regarded as the inverse process of mapping transformation under certain conditions.

[0064] In a possible implementation, the step of stitching and fusing the plurality of overhead view images to obtain the surround view image includes: weighting the plurality of overhead view images, and stitching and fusing the weighted plurality of overhead view images to obtain the surround view image.

[0065] In one example, the weight of each overhead view can be determined according to the position of the image acquisition device corresponding to each overhead view in the vehicle. For example, the weight of the overhead view image corresponding to the left and right position of the vehicle is greater than the weight of the overhead view image corresponding to the front and rear position of the vehicle. Of course, the weight of each overhead view can also be set according to other conditions, such as the coverage of the image acquisition device corresponding to each overhead view, etc., which is not limited in the embodiment of the present disclosure.

[0066] In the disclosed embodiment, the stitching and fusion can be regarded as two steps, namely image stitching and image fusion. Image stitching refers to stitching two or more images containing partially overlapping areas to obtain an image with a wider viewing angle or a 360-degree viewing angle. Under normal circumstances, the number and position of fisheye cameras on the vehicle are fixed, and the relative positions between the cameras will not change. Therefore, the method of manually selecting reference points can be used to achieve alignment. After determining the fisheye cameras, reference points are manually selected according to the positions of each camera. After obtaining the fisheye image, the common reference points in adjacent images are merged, so that the stitching of the surrounding images can be completed.

[0067] There is a seam at the junction of two adjacent images in the image obtained after image stitching, that is, the overlapping area of ​​different lenses. It is necessary to use an image fusion algorithm at its location to eliminate the image seam, so that the image can smoothly transition in the overlapping area without obvious stitching marks. In the embodiment of the present disclosure, the images are fused by a weighted fusion algorithm, which is to weight the grayscale values ​​of the overlapping areas in the images according to the distance from the seam and then add them, that is, the grayscale values ​​in different images are multiplied by their respective weights and then added.

[0068] In the actual process of parking space detection, generally speaking, the probability of a parking space being on the left or right side of the vehicle is much greater than the probability of a parking space being in front or behind the vehicle. Therefore, in the embodiment of the present disclosure, during the image fusion process, the weight of the overhead view image corresponding to the left and right sides of the vehicle can be set to be greater than the weight of the overhead view image corresponding to the front and rear directions of the vehicle.

[0069] In step S12, corner point information and entrance line information may be acquired based on the surround view image, wherein the corner point information represents relevant information of corner points in the surround view image, and the entrance line information represents relevant information of entrance lines of parking spaces in the surround view image.

[0070] In the disclosed embodiment, after obtaining the surround view image, the surround view image can be placed in the backbone network for image feature extraction to obtain image features about the parking space in the surround view image, and then the corner point information and entrance line information can be obtained through the corner point extraction branch and the entrance line extraction branch.

[0071] Figure 2a-2c A parking space classification diagram according to an embodiment of the present disclosure is shown. Figure 2a-2c As shown, the parking space can be simplified to a parallel parking space ( Figure 2a As shown), tilt parking space ( Figure 2b shown) and vertical parking spaces ( Figure 2c Through the surround image and the backbone network, the disclosed embodiment can obtain the entrance line information and corner point information for different parking spaces.

[0072] Among them, the backbone network can be a commonly used image feature extraction network, such as: VGG network model, ResNet network model, etc. The image feature extraction model here is not limited in the embodiment of the present disclosure, as long as it can normally extract the parking space features in the surround view image.

[0073] In the disclosed embodiment, after obtaining the parking space features in the surround image, the recognition result of the parking space line can be obtained through the extraction network, and then the recognition result of the parking space corner point and the parking space entrance line can be determined through the recognition result of the parking space line. In the image, because the color of the parking space line is usually white, the grayscale of the image around the parking space line will change dramatically from black to white or from white to black. Therefore, the parking space line is recognized by gradient. The gradient of the image is calculated to obtain the gradient value, and the gradient value is sorted to select the area that may be the parking space line.

[0074] Corner points are points with the maximum local standard deviation of image pixels, which are generally not easily affected by light and have large curvature. Parking space corner points are the characteristics of the intersection of adjacent parking space lines. Therefore, according to the above characteristics, the present disclosure adopts a corner point detection algorithm based on straight line intersections to identify parking space corner points. After the parking space lines are screened, only parallel lines, vertical lines or oblique lines are left among all the lines, and there must be intersections between them. By extracting these intersections, the parking space corner point information can be obtained.

[0075] In one possible implementation, the corner point information includes the corner point position, corner point confidence, orientation angle, shape and corner point type; the entrance line information includes the entrance line center point position, entrance line confidence, entrance line length, entrance line angle, parking space type where the entrance line is located and whether the parking space corresponding to the entrance line is occupied, wherein the corner point type and the parking space type where the entrance line is located include right angles and non-right angles.

[0076] Figure 3a-3d FIG. 4 shows a parking space corner point shape classification diagram according to an embodiment of the present disclosure. Figure 3a-3d As shown, the shapes included in the corner point information can be four as shown in the figure, namely, T-shaped ( Figure 3a shown), L-type ( Figure 3b shown), linear ( Figure 3c shown) and U-shaped ( Figure 3d shown).

[0077] The corner point positions in the above corner point information are the corner point coordinates. The vehicle can be used as a reference coordinate system to convert the positions of the corner points detected around it into corner point coordinates on the coordinate system. This allows the spatial positions to be converted into calculable parameters, preparing for subsequent matching and detection and reducing the computing burden.

[0078] The corner point confidence is the possibility that the point is a corner point of a parking space. The greater the corner point confidence, the greater the possibility that the point is determined to be a corner point. A preset confidence value can be set, and the preset value is used as a standard. Corner points corresponding to corner point confidences greater than the preset value are listed as possible corner points, and corner points less than the preset value are removed, thereby improving the accuracy of parking space detection.

[0079] The orientation angle needs to be determined by two corner points, that is, the orientation of two points is determined by two corner points. The corner point type is right angle type and non-right angle type. Determining the corner point type is convenient for matching with the entrance line below. For example, the corner point in the parallel parking space is right angle type, and the corner point in the inclined parking space is non-right angle type.

[0080] The center point position of the entrance line in the above entrance line information is the coordinate of the center point of the entrance line. The entrance line confidence is the possibility that the parking space line is the parking space entrance line. The greater the entrance line confidence, the greater the possibility that the entrance line is determined to be the entrance line. The entrance line angle is the inclination angle of the entrance line, and the inclination angle is the angle between the entrance line and the reference line. The parking space type where the entrance line is located, that is, whether the parking space is right-angled or non-right-angled, can be used to match the corner point.

[0081] Whether it is the angle of the entrance line or the inclination angle of the entrance line, the entrance line can be converted into a line segment in the coordinate system using the vehicle's reference coordinate system. In this way, the endpoint of the line segment can be determined by the slope of the line segment and the coordinates and length of one of its points.

[0082] In step S13, a first entry line and two first corner points at different positions that match the first entry line may be determined based on the corner point information and the entry line information.

[0083] In the disclosed embodiment, after obtaining the corner point information and the entrance line information, different entrance lines can be selected to find the corner points that match them, so that different parking spaces in the surround image can be identified. For example, after obtaining the entrance line information, the entrance lines can be sorted according to their confidences, and the entrance line with a larger confidence is preferentially determined as the first entrance line, and then the first corner point that matches the first entrance line is found, so that a mutually matching entrance line and corner point can be obtained, and the approximate information of a parking space can be determined.

[0084] In a possible implementation, the determining of the first entrance line and two first corner points of different positions matching the first entrance line based on the corner point information and the entrance line information includes: selecting corner points whose corner point confidence is greater than a first preset confidence as second corner points based on the corner point information, and selecting entrance lines whose entrance line confidence is greater than a second preset confidence as second entrance lines based on the entrance line information; selecting an entrance line from the second entrance lines as the first entrance line, and selecting a second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line from the second corner points based on the corner point information and the entrance line information as the third corner point; obtaining the two end point positions of the first entrance line based on the entrance line center point position, the entrance line length and the entrance line angle of the first entrance line; and selecting the corner point of the third corner point whose distance from the two end point positions of the first entrance line is less than a preset value as the first corner point.

[0085] In the disclosed embodiment, after obtaining the entry line information and the corner point information, the entry line information and the corner point information can be screened according to their confidences, that is, the corner point whose corner point confidence is greater than the first preset confidence is taken as the second corner point, and the entry line whose entry line confidence is greater than the second preset confidence is taken as the second entry line. The first preset confidence and the second preset confidence can be set according to the needs of the user.

[0086] After determining the second corner point and the second entrance line, you can select an entrance line from the second entrance line as the first entrance line, so that you can determine an entrance line as a reference line. According to the corner point information and the entrance line information, select a second corner point from the second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line as the third corner point. According to the entrance line information, the center point position and the length of the entrance line of the first entrance line can be determined, and the two endpoints of the entrance line can be obtained through the coordinates of the center point position of the entrance line and the length and angle of the entrance line:

[0087]

[0088] Among them, p1(x, y) and p2(x, y) are the two endpoints of the entrance line, p(x, y) is the center point of the entrance line, l is half the length of the entrance line, and θ represents the angle of the entrance line.

[0089] In the disclosed embodiment, the corner points among the third corner points whose distances to the two end points of the first entrance line are less than a preset value are taken as the first corner points. After obtaining the end points of the first entrance line, the third corner points can be traversed, and the corner points among the third corner points whose distances to the two end points are less than a preset value are taken as the first corner points. In this way, the first entrance line and the first corner point can be obtained, that is, the preliminary information of a parking space can be determined.

[0090] In step S14, the detection result of the parking space may be determined according to the first corner point and / or the first entrance line.

[0091] In the disclosed embodiment, after the preliminary information of a parking space is determined, the remaining detection results of the parking space, such as the parking space location, parking space type, parking space depth, etc., can be determined based on the information of the first corner point and the first entrance line. The above information makes subsequent automatic parking more convenient.

[0092] In a possible implementation manner, determining the detection result of the parking space according to the first corner point and / or the first entrance line includes: determining the position of the parking space according to the corner point position and the orientation angle of the first corner point.

[0093] In the disclosed embodiment, after obtaining the corner point position of the first corner point, the position of the parking space can be determined according to the positions and orientation angles of the two corner points, and the parking space position can be represented by the coordinates of the first corner point. In this way, multiple parking spaces can be distinguished.

[0094] In a possible implementation manner, determining the detection result of the parking space according to the first corner point and / or the first entrance line includes: determining the type of the parking space according to the first entrance line.

[0095] In a possible implementation, determining the type of the parking space based on the first entrance line includes: when the parking space type where the entrance line of the first entrance line is located is non-right angle, determining the parking space as an inclined parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is greater than a calibration value, determining the parking space as a parallel parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is less than the calibration value, determining the parking space as a vertical parking space.

[0096] In the disclosed embodiment, after the first entrance line and the first corner point are determined, the type of the parking space can be determined. Figure 2a-2c As shown, parking spaces can be divided into three types: inclined parking spaces, vertical parking spaces, and parallel parking spaces. In the case where the parking space type where the first entrance line is located is not at a right angle, Figure 2a-2c It can be seen that only Figure 2b It meets the non-right angle requirement. Therefore, when the first entrance line is determined to be non-right angle, it can be determined that the parking space is an inclined parking space. Figure 2a-2cAs shown, it may be a parallel parking space or a vertical parking space. Therefore, in order to further determine the type of parking space, a calibration value can be set according to the prescribed standards of the parking space. The general length standard of a parallel parking space is 6 meters and the width is 2.5 meters. Therefore, the calibration value can be set to 4 meters. When the length of the first entrance line is greater than 4 meters, the parking space is determined to be a parallel parking space. When the length of the first entrance line is less than 4 meters, the parking space is determined to be a vertical parking space.

[0097] In a possible implementation manner, the method further includes: searching for preset parameters of the parking space according to the type of the parking space, where the preset parameters include a depth of the parking space.

[0098] In the disclosed embodiment, after the type of parking space is determined, the corresponding standard can be found according to the type of parking space. The depth of the parking space indicates the distance between the entrance line and the bottom line of the parking space. For example, for a parallel parking space, the depth of the parking space can be regarded as the width of the parking space. According to the current standard size of parking spaces, the depth of the parallel parking space can be set to 2.5 meters. In this way, when the type of the parking space is determined to be a parallel parking space, the depth of the parking space can be obtained.

[0099] In a possible implementation, determining the detection result of the parking space according to the first corner point and / or the first entrance line includes: determining whether the parking space is occupied according to the parking space occupancy result of the first entrance line; or determining whether the parking space is occupied according to a radar detection result.

[0100] In the disclosed embodiment, after the location of the parking space is determined, it is also necessary to determine whether the parking space is available, that is, whether it is occupied by other vehicles. After the location of the parking space is determined, feature extraction can be performed to determine whether it is occupied by other vehicles. In addition, the location can be detected by radar detection, and the data from the sensor can be used to determine whether the parking space is occupied.

[0101] In the disclosed embodiment, the vehicle's image acquisition device can acquire fisheye images around the vehicle, and the fisheye lens characteristics are used to perform dedistortion processing to obtain a bird's-eye view around the vehicle, and then the bird's-eye views are spliced ​​and fused to obtain a surround view image of the vehicle. The above method can put the environmental information around the vehicle together, which is convenient for the subsequent identification of the parking space contained therein.

[0102] In the disclosed embodiment, after obtaining the surround view image of the vehicle, the features in the surround view image can be extracted through the backbone network, and then the corner point extraction branch and the entrance line extraction branch are used to extract the corner point information and the entrance line information in the features. The surround view image around the vehicle can be converted into the corner point information and the entrance line information related to the parking space through the above method, and the information that can be used for parking space detection can be extracted, which is convenient for the subsequent identification of the parking space.

[0103] In the disclosed embodiment, after obtaining the corner point information and entrance line information of the parking space, the entrance line information and the corner point information can be matched with each other. Each matched first entrance line and first corner point is a detected parking space. After detecting the parking space, other parameters of the parking space can also be obtained according to the entrance line information and the corner point information in the parking space, including but not limited to the parking space type, parking space depth, parking space position, and parking space status. Through the above method, the parking space corner point information and the parking space entrance line information extracted from the vehicle's surround view image can be used to accurately and quickly complete the parking space detection. Specifically, after completing the parking space detection, the parking space detection results such as the position, occupancy status, parking space type, and parking space depth of the detected parking space can be obtained.

[0104] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present disclosure will not repeat them. It can be understood by those skilled in the art that in the above-mentioned method of the specific implementation method, the specific execution order of each step should be determined according to its function and possible internal logic.

[0105] In addition, the present disclosure also provides a parking space detection device, an electronic device, a computer-readable storage medium, and a program, all of which can be used to implement any parking space detection method provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding records in the method part and will not be repeated here.

[0106] Figure 4 FIG. 2 is a schematic diagram showing the structure of a parking space detection device according to an embodiment of the present disclosure. Figure 4 As shown, the device 40 may include:

[0107] A first acquisition module 41 is used to acquire a surround view image of the vehicle, wherein the surround view image represents a panoramic bird's-eye view centered on the vehicle;

[0108] A second acquisition module 42 is used to acquire corner point information and entrance line information based on the surround view image, wherein the corner point information represents relevant information of the corner points in the surround view image, and the entrance line information represents relevant information of the entrance line of the parking space in the surround view image;

[0109] A first determining module 43, configured to determine a first entry line and two first corner points at different positions that match the first entry line based on the corner point information and the entry line information;

[0110] The second determining module 44 is configured to determine a detection result of a parking space according to the first corner point and / or the first entrance line.

[0111] In a possible implementation, the first acquisition module is further used to: acquire multiple fisheye lens images through multiple image acquisition devices installed on the vehicle, the multiple image acquisition devices correspond to different orientations, and the multiple image acquisition devices can cover the surroundings of the vehicle; perform dedistortion processing on the multiple fisheye lens images to obtain multiple corrected images; perform top-down change processing on the multiple corrected images to obtain corresponding multiple top-view images; and perform splicing and fusion of the multiple top-view images to obtain the surround view image.

[0112] In a possible implementation, the stitching and fusing of the multiple overhead view images to obtain the surround view image includes: weighting the multiple overhead view images, and stitching and fusing the weighted multiple overhead view images to obtain the surround view image, wherein the weight of the overhead view images corresponding to the left and right positions of the vehicle is greater than the weight of the overhead view images corresponding to the front and rear positions of the vehicle.

[0113] In one possible implementation, the corner point information includes the corner point position, corner point confidence, orientation angle, shape and corner point type; the entrance line information includes the entrance line center point position, entrance line confidence, entrance line length, entrance line angle, parking space type where the entrance line is located and whether the parking space corresponding to the entrance line is occupied, wherein the corner point type and the parking space type where the entrance line is located include right angles and non-right angles.

[0114] In a possible implementation, the first determination module is also used to: screen out corner points whose corner point confidence is greater than a first preset confidence as second corner points based on the corner point information, and screen out entrance lines whose entrance line confidence is greater than a second preset confidence as second entrance lines based on the entrance line information; select an entrance line from the second entrance lines as the first entrance line, and select a second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line from the second corner points based on the corner point information and the entrance line information as the third corner point; obtain the two end point positions of the first entrance line based on the entrance line center point position, entrance line length and entrance line angle of the first entrance line; and use the corner point in the third corner point whose distance from the two end point positions of the first entrance line is less than a preset value as the first corner point.

[0115] In a possible implementation manner, the second determining module is further configured to determine the position of the parking space according to the corner point position and the orientation angle of the first corner point.

[0116] In a possible implementation manner, the second determination module is further configured to: determine the type of the parking space according to the first entrance line.

[0117] In a possible implementation, determining the type of the parking space based on the first entrance line includes: when the parking space type where the entrance line of the first entrance line is located is non-right angle, determining the parking space as an inclined parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is greater than a calibration value, determining the parking space as a parallel parking space; when the parking space type where the entrance line of the first entrance line is located is right angle and the entrance line length of the first entrance line is less than the calibration value, determining the parking space as a vertical parking space.

[0118] In a possible implementation manner, the device further includes: a parameter search module, configured to search for preset parameters of the parking space according to the type of the parking space, wherein the preset parameters include the depth of the parking space.

[0119] In a possible implementation, the second determination module is further configured to: determine whether the parking space is occupied according to a parking space occupancy result of the first entrance line; or determine whether the parking space is occupied according to a radar detection result.

[0120] This method has a specific technical connection with the internal structure of the computer system, and can solve the technical problem of how to improve the hardware computing efficiency or execution effect (including reducing the amount of data storage, reducing the amount of data transmission, increasing the hardware processing speed, etc.), thereby obtaining the technical effect of improving the internal performance of the computer system in accordance with the laws of nature.

[0121] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0122] The present disclosure also provides a vehicle, which includes at least one processor and at least one storage device, wherein the storage device can be configured to store a program for executing the parking space detection method of the above method embodiment, and the processor can be configured to execute the program in the storage device, including but not limited to the program for executing the parking space detection method of the above method embodiment.

[0123] The embodiment of the present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and the computer program instructions implement the above method when executed by a processor. The computer-readable storage medium can be a volatile or non-volatile computer-readable storage medium.

[0124] An embodiment of the present disclosure further proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to call the instructions stored in the memory to execute the above method.

[0125] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.

[0126] The electronic device may be provided as a terminal, a server, or a device in other forms.

[0127] Figure 5 A block diagram of an electronic device according to an embodiment of the present disclosure is shown. For example, the electronic device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, or other terminal devices.

[0128] Reference Figure 5 , the electronic device may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0129] The processing component 802 generally controls the overall operation of the electronic device, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0130] The memory 804 is configured to store various types of data to support the operation of the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0131] The power supply component 806 provides power to various components of the electronic device. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device.

[0132] The multimedia component 808 includes a screen that provides an output interface between the electronic device and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the electronic device is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0133] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the electronic device is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0134] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0135] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the electronic device. For example, the sensor assembly 814 can detect the on / off state of the electronic device, the relative positioning of the components, such as the display and keypad of the electronic device, and the sensor assembly 814 can also detect the position change of the electronic device or a component of the electronic device, the presence or absence of user contact with the electronic device, the orientation or acceleration / deceleration of the electronic device and the temperature change of the electronic device. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include a light sensor, such as a complementary metal oxide semiconductor (CMOS) or a charge coupled device (CCD) image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0136] The communication component 816 is configured to facilitate wired or wireless communication between the electronic device and other devices. The electronic device can access a wireless network based on a communication standard, such as a wireless network (Wi-Fi), a second generation mobile communication technology (2G), a third generation mobile communication technology (3G), a fourth generation mobile communication technology (4G), a long-term evolution (LTE) of a universal mobile communication technology, a fifth generation mobile communication technology (5G), or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0137] In an exemplary embodiment, the electronic device may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above methods.

[0138] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions, which can be executed by a processor 820 of an electronic device to perform the above method.

[0139] Figure 61 shows a block diagram of an electronic device according to an embodiment of the present disclosure. For example, the electronic device can be provided as a server or a terminal device. Figure 6 , the electronic device includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions, such as applications, that can be executed by the processing component 1922. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0140] The electronic device may also include a power supply component 1926 configured to perform power management of the electronic device, a wired or wireless network interface 1950 configured to connect the electronic device to a network, and an input / output (I / O) interface 1958. The electronic device may operate based on an operating system stored in the memory 1932, such as a Microsoft Server operating system (Windows Server 2003). TM ), a graphical user interface operating system launched by Apple (Mac OS X TM ), a multi-user, multi-process computer operating system (Unix TM ), a free and open source Unix-like operating system (Linux TM ), an open source Unix-like operating system (FreeBSD TM ) or similar.

[0141] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by a processing component 1922 of an electronic device to perform the above method. The computer program product can be implemented in hardware, software, or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK), etc.

[0142] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0143] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0144] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0145] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A parking space detection method, characterized in that: The method comprises: Acquire a surround view image of the vehicle, wherein the surround view image represents a panoramic bird's-eye view centered on the vehicle; Acquire corner point information and entrance line information based on the surround view image, wherein the corner point information represents relevant information of the corner points in the surround view image, and the entrance line information represents relevant information of the entrance line of the parking space in the surround view image; Determine a first entry line and two first corner points at different positions that match the first entry line based on the corner point information and the entry line information; A detection result of the parking space is determined according to the first corner point and / or the first entrance line.

2. The method according to claim 1, characterized in that: The acquiring of the surround view image of the vehicle comprises: Capturing a plurality of fisheye lens images by means of a plurality of image acquisition devices installed on the vehicle, wherein the plurality of image acquisition devices correspond to different orientations and the plurality of image acquisition devices can cover the periphery of the vehicle; Performing dedistortion processing on the multiple fisheye lens images to obtain multiple corrected images; Performing top-view change processing on the plurality of corrected images to obtain a plurality of corresponding top-view images; The plurality of overhead images are spliced ​​and fused to obtain the panoramic image.

3. The method according to claim 2, characterized in that The step of splicing and fusing the plurality of overhead images to obtain the surround image comprises: The multiple overhead view images are weighted, and the weighted multiple overhead view images are spliced ​​and fused to obtain the surround view image, wherein the weight of the overhead view images corresponding to the left and right directions of the vehicle is greater than the weight of the overhead view images corresponding to the front and rear directions of the vehicle.

4. The method according to claim 1, characterized in that The corner point information includes the corner point position, corner point confidence, orientation angle, shape and corner point type; the entrance line information includes the entrance line center point position, entrance line confidence, entrance line length, entrance line angle, parking space type where the entrance line is located and whether the parking space corresponding to the entrance line is occupied, wherein the corner point type and the parking space type where the entrance line is located include right angles and non-right angles.

5. The method according to claim 4, characterized in that The determining, based on the corner point information and the entry line information, a first entry line and two first corner points at different positions that match the first entry line comprises: Based on the corner point information, a corner point with a corner point confidence greater than a first preset confidence is selected as a second corner point; based on the entrance line information, an entrance line with a entrance line confidence greater than a second preset confidence is selected as a second entrance line; Select an entrance line from the second entrance lines as the first entrance line, and select a second corner point whose corner point type is consistent with the parking space type of the entrance line of the first entrance line from the second corner points according to the corner point information and the entrance line information as the third corner point; Obtaining the positions of both end points of the first entry line according to the position of the entry line center point, the entry line length and the entry line angle of the first entry line; The corner points among the third corner points whose distances from the two end points of the first entry line are less than a preset value are used as the first corner points.

6. The method according to claim 4, characterized in that The detection result of determining the parking space according to the first corner point and / or the first entrance line includes: The position of the parking space is determined according to the corner point position and the orientation angle of the first corner point.

7. The method according to claim 4, characterized in that The detection result of determining the parking space according to the first corner point and / or the first entrance line includes: The type of the parking space is determined according to the first entrance line.

8. The method according to claim 7, characterized in that Determining the type of the parking space according to the first entrance line includes: When the parking space type where the entrance line of the first entrance line is located is non-perpendicular, determining that the parking space is an inclined parking space; When the parking space type where the entrance line of the first entrance line is located is a right angle and the entrance line length of the first entrance line is greater than a calibration value, determining that the parking space is a parallel parking space; When the parking space type where the entrance line of the first entrance line is located is a right angle and the entrance line length of the first entrance line is less than the calibration value, it is determined that the parking space is a vertical parking space.

9. The method according to claim 7 or 8, characterized in that: The method further comprises: The preset parameters of the parking space are searched according to the type of the parking space, where the preset parameters include the depth of the parking space.

10. The method according to claim 4, characterized in that The detection result of determining the parking space according to the first corner point and / or the first entrance line includes: Determining whether the parking space is occupied according to the parking space occupancy result of the first entrance line; or, Determine whether the parking space is occupied according to the radar detection result.

11. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 10.

12. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images