Parking performance testing method, device, server and product

Abstract

The present application provides a parking performance test method, device, server and product. After parking is completed, images collected by parking cameras located at different positions on the vehicle are obtained, and each image shows a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point; for each image, the first distance between the preset target point and the section of parking space line in the image is obtained; based on the preset distance between at least two preset target points located on the same side of the vehicle and arranged horizontally, and the first distances corresponding to at least two preset target points located on the same side of the vehicle and arranged horizontally, the inclination angle of the vehicle relative to the parking space is determined; thereby, the position of the vehicle parked in the parking space can be determined. The time required for automatic parking to the corresponding position can also be determined based on the length of parking time. If the time consumed is long, it is also a manifestation of inaccurate parking, thereby determining the parking performance of the vehicle.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and more specifically, to parking performance testing methods, devices, servers and products. Background Art

[0002] At present, parking spaces are becoming more and more dense, and drivers often encounter the problem of inconvenience in opening the car door after parking their vehicles. Therefore, as an important component of the Advanced Driver Assistance System (ADAS), the automatic parking function came into being.

[0003] After automatic parking, if the position of the vehicle in the parking space is inaccurate, for example, the wheels of the vehicle press against the parking space line, other vehicles will not be able to park in the parking space next to the vehicle, or other vehicles will scratch the vehicle when parking. Therefore, accuracy is an important performance indicator of automatic parking performance. Therefore, the parking performance of the vehicle needs to be tested. Summary of the invention

[0004] In view of this, the present application provides a parking performance testing method, device, server and product.

[0005] To achieve the above objectives, this application provides the following technical solutions:

[0006] According to a first aspect of an embodiment of the present disclosure, a parking performance test method is provided, comprising:

[0007] When the vehicle to be tested for parking performance has finished parking, images collected by parking cameras disposed at different positions on the vehicle are obtained, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, and all the images at least involve two adjacent sides of the vehicle, and at least two of the preset target points are located on the same side of the vehicle and are disposed transversely;

[0008] For each of the images, obtaining a first distance between the preset target point in the image and the section of the parking space line, so as to obtain first distances respectively corresponding to a plurality of the preset target points;

[0009] Determining the tilt angle of the vehicle relative to the parking space based on a preset distance between at least two preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged transversely;

[0010] Obtaining the parking time of the vehicle;

[0011] A first parking performance of the vehicle is determined based on the first distances, the tilt angles, and the parking time durations respectively corresponding to the plurality of preset target points.

[0012] In combination with the first aspect, in a first possible implementation manner, obtaining a first distance between the preset target point and the section of parking space line in the image includes:

[0013] Binarizing the image according to the preset parking space line color to obtain a binary image;

[0014] Performing morphological closing and opening operations on the binary image to obtain a first image;

[0015] Performing perspective transformation on the first image to obtain a bird's-eye view image;

[0016] Detecting straight lines in the overhead image, and identifying two straight lines whose distances are within a preset distance range and whose included angles are less than a preset included angle as parking space line boundaries;

[0017] The distance between the preset target point and the parking space line boundary is taken as the first distance.

[0018] In combination with the first aspect, in a second possible implementation manner, binarizing the image according to the preset parking space line color to obtain a binary image includes:

[0019] Performing histogram equalization processing on the image to obtain a high-contrast image;

[0020] Convert the high contrast image from the RGB color space to the HSV color space to obtain an HSV image;

[0021] The HSV image is binarized according to the preset parking space line color to obtain a binarized image.

[0022] In combination with the first aspect, in a third possible implementation manner, detecting the straight lines in the overhead image and identifying two straight lines whose distances are within a preset distance range and whose included angles are less than a preset included angle as parking space line boundaries includes:

[0023] Performing edge detection on the overhead image to obtain a second image;

[0024] Detecting straight lines in the second image using a Hough transform algorithm;

[0025] The parking space line boundary is screened out from the straight lines in the second image according to a preset distance range and a preset angle.

[0026] In combination with the first aspect, in a fourth possible implementation manner, the preset target points set on the left and right sides of the vehicle include: a tangent point between a left front wheel and the ground, a tangent point between a left rear wheel and the ground, a tangent point between a right front wheel and the ground, and a tangent point between a right rear wheel and the ground, and based on a preset distance between at least two of the preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two of the preset target points located on the same side of the vehicle and arranged transversely, determining the inclination angle of the vehicle relative to the parking space includes:

[0027] Calculate a first angle between the left side of the vehicle and the left parking space line of the parking space based on the vehicle wheelbase and the first distances corresponding to the tangent points of the left front wheel and the ground and the tangent points of the left rear wheel and the ground;

[0028] Calculating a second angle between the right side of the vehicle and the right parking space line of the parking space based on the vehicle wheelbase and the first distances corresponding to the tangent points of the right front wheel and the ground and the tangent points of the right rear wheel and the ground;

[0029] An average value of the first angle and the second angle is determined as the inclination angle.

[0030] In combination with the first aspect, a fifth possible implementation manner further includes:

[0031] averaging the first distances corresponding to the plurality of preset target points obtained by the vehicle parking multiple times to obtain the first average distances corresponding to the plurality of preset target points;

[0032] averaging a plurality of the tilt angles obtained by parking the vehicle multiple times to obtain an average tilt angle;

[0033] Averaging the parking time durations obtained from multiple parking operations of the vehicle to obtain an average parking time duration;

[0034] A second parking performance of the vehicle is determined based on the first average distance, the average tilt angle, and the average parking time.

[0035] In combination with the first aspect, in a sixth possible implementation manner, obtaining a first distance between the preset target point and the section of parking space line in the image includes:

[0036] Performing perspective transformation on the image to obtain a bird's-eye view image;

[0037] Binarizing the top view image according to the preset parking space line color to obtain a binary image;

[0038] Performing morphological closing and opening operations on the binary image to obtain a first image;

[0039] Detect straight lines in the first image, and identify two straight lines whose distance is within a preset distance range and whose included angle is less than a preset included angle as parking space line boundaries;

[0040] The distance between the preset target point and the parking space line boundary is taken as the first distance.

[0041] According to a second aspect of an embodiment of the present disclosure, there is provided a parking performance testing device, comprising:

[0042] A first acquisition module is used for acquiring images collected by parking cameras arranged at different positions on the vehicle when the vehicle to be tested for parking performance has finished parking, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, and all the images at least involve two adjacent sides of the vehicle, and at least two of the preset target points are located on the same side of the vehicle and are arranged horizontally;

[0043] A second acquisition module is used to acquire, for each of the images, a first distance between the preset target point in the image and the section of the parking space line, so as to obtain first distances corresponding to a plurality of the preset target points respectively;

[0044] a first determining module, configured to determine a tilt angle of the vehicle relative to the parking space based on a preset distance between at least two preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged transversely;

[0045] A third acquisition module is used to acquire the parking time of the vehicle;

[0046] The second determination module is used to determine the first parking performance of the vehicle based on the first distances, the tilt angles and the parking time durations respectively corresponding to the plurality of preset target points.

[0047] According to a third aspect of an embodiment of the present disclosure, a server is provided, characterized in that it includes:

[0048] processor;

[0049] a memory for storing instructions executable by the processor;

[0050] The processor is configured to execute the instructions to implement the parking performance testing method as described in the first aspect.

[0051] According to a fourth aspect of an embodiment of the present disclosure, there is provided a computer program product, comprising a computer program / instructions, wherein the computer program / instructions, when executed by a processor, implement the parking performance test method described in the first aspect.

[0052] It can be known from the above technical solutions that in the parking performance test method provided by the present application, after parking is completed, images collected by parking cameras arranged at different positions on the vehicle are obtained, each image shows a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, all images involve at least two adjacent sides of the vehicle, and at least two preset target points are located on the same side of the vehicle and are arranged horizontally; for each image, a first distance between the preset target point and the section of parking space line in the image is obtained to obtain first distances corresponding to a plurality of preset target points respectively; based on the preset distance between at least two preset target points located on the same side of the vehicle and arranged horizontally, and the first distances corresponding to at least two preset target points located on the same side of the vehicle and arranged horizontally, the inclination angle of the vehicle relative to the parking space is determined; thereby, the position of the vehicle parked in the parking space can be determined based on the first distances and inclination angles corresponding to a plurality of images respectively. The parking time can also be obtained. Based on the parking time, the time required for automatic parking at the corresponding position can be determined. If the time consumed is long, it is also a manifestation of inaccurate parking. Therefore, the parking performance of the vehicle can be determined by combining the first distance, tilt angle and parking time corresponding to multiple preset target points. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present application 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 embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0054] Figure 1 A structural diagram of the hardware architecture involved in the embodiments of the present application;

[0055] Figure 2a to Figure 2b A schematic diagram of a real environment coordinate system provided in an embodiment of the present application;

[0056] Figure 3 A schematic diagram of the calibration process of a parking camera provided in an embodiment of the present application;

[0057] Figure 4 A flowchart of a parking performance testing method provided in an embodiment of the present application;

[0058] Figure 5 A schematic diagram of an image captured by a parking camera provided in an embodiment of the present application;

[0059] Figure 6 is a schematic diagram of an image collected when the field of view of the parking camera is large;

[0060] Figure 7 A schematic diagram of a first distance between a preset target point and a parking space line provided in an embodiment of the present application;

[0061] Figures 8a to 8c A schematic diagram of an image processed by performing morphological closing and opening operations provided in an embodiment of the present application;

[0062] Fig. 9 For the general Figure 8c A schematic diagram of a top-view image after perspective transformation of the first image shown;

[0063] Fig.10 A schematic diagram of an implementation method of the second image provided in an embodiment of the present application;

[0064] Fig.11 A schematic diagram of a straight line detected from a second image provided by an embodiment of the present application;

[0065] Fig.12 A structural diagram of a parking performance testing device provided in an embodiment of the present application;

[0066] Fig.13 The invention is a structural block diagram of a server according to an exemplary embodiment. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0068] The embodiments of the present application provide a parking performance testing method, device, electronic device, medium and product. Before introducing the technical solution provided by the embodiments of the present application, the hardware architecture involved in the embodiments of the present application is first described.

[0069] like Figure 1 , which is a structural diagram of a hardware architecture involved in an embodiment of the present application, the hardware architecture includes: a vehicle 11 to be tested for parking performance and a server 12.

[0070] For example, a vehicle is provided with multiple cameras, and the multiple cameras can form a panoramic camera for the vehicle. For example, the parking camera mentioned in the embodiment of the present application can be a panoramic camera. The parking camera can be used to capture an image including multiple preset target points and parking space lines of the parking space.

[0071] Exemplarily, the preset target point may be the point of contact between the wheels of the vehicle and the ground. Exemplarily, the preset target point may be a mapping point of the front or rear of the vehicle. The mapping point is the point where the projection of the front or rear of the vehicle on the ground is closest to the parking space line.

[0072] Exemplarily, parking cameras are arranged above multiple preset target points located on the same side of the vehicle and arranged laterally. For example, a parking camera A is arranged above the intersection point between the left front wheel and the ground and the intersection point between the left rear wheel and the ground on the left side of the car, and a parking camera B is arranged above the intersection point between the right front wheel and the ground and the intersection point between the right rear wheel and the ground on the right side of the car.

[0073] For example, parking camera A can capture the tangent points between the left front wheel of the car and the ground, the tangent points between the left rear wheel and the ground, and a parking space line on the left side of the parking space. Parking camera B can capture the tangent points between the right front wheel of the car and the ground, the tangent points between the right rear wheel and the ground, and a parking space line on the right side of the parking space.

[0074] Exemplarily, a parking camera is respectively arranged above each preset target point; exemplarily, a parking camera is arranged above each wheel of the vehicle 11, and a parking camera is arranged on the rear of the vehicle 11; or, a parking camera is arranged above each wheel of the vehicle 11, and a parking camera is arranged on the front of the vehicle 11; or, a parking camera is arranged above each wheel of the vehicle 11, a parking camera is arranged on the front of the vehicle 11, and a parking camera is arranged on the rear of the vehicle 11.

[0075] It is understandable that in the process of testing the parking performance, it is necessary to detect the position of the vehicle parked in the parking space, so the distance between the intersection point of the wheel and the ground and the parking space line is needed; or, the distance between the intersection point of the wheel and the ground and the parking space line, as well as the distance between the mapping point of the front / rear of the vehicle and the parking space line are needed.

[0076] In the embodiment of the present application, the tangent point between the wheel and the ground and the mapping point of the front / rear of the vehicle are collectively referred to as preset target points.

[0077] Generally, if the object (wheel, front or rear of the vehicle) and the parking space line are photographed from a bird's-eye view, the deformation of the object and the parking space line will be relatively small, while if they are photographed from other angles, the deformation of the object and the parking space line will be relatively large. When using the parking camera on the vehicle to collect images, it is obviously impossible to shoot from a bird's-eye view, but because the parking camera is set above the object, the deformation of the object and the parking space line in the image collected by the parking camera is relatively small. If the deformation of the object and the parking space line in the image is relatively small, the preset target point between the object and the ground can be determined more accurately.

[0078] Exemplarily, the vehicle 11 may be any vehicle capable of automatic parking, such as a car, a bus, or a large truck.

[0079] Figure 1 The vehicle is a car as an example for explanation.

[0080] The images collected by the parking camera in the vehicle 11 can be uploaded to the server 12, so that the server 12 analyzes the images collected by the parking camera to evaluate the parking performance of the vehicle. Of course, in some implementations, the controller on the vehicle can also directly evaluate the parking performance of the vehicle.

[0081] Exemplarily, the server 12 may be a single server, or a server cluster consisting of multiple servers, or a cloud computing server center. The server 12 may include a processor, a memory, a network interface, and the like.

[0082] In an optional implementation, the hardware architecture may further include an electronic device 13 .

[0083] In an optional implementation, the image captured by the parking camera can be sent to the electronic device 13, and the electronic device 13 can display the image, so that the user of the electronic device 13 can mark the preset target point between the object contained in the image and the ground through the input device of the electronic device 13, so that the server 13 can analyze the distance between the preset target point of the object and the parking space line.

[0084] Exemplarily, the electronic device 13 can be any electronic product that can interact with a user through one or more methods such as a keyboard, touchpad, touch screen, remote control, voice interaction or handwriting device, such as a mobile phone, a laptop computer, a tablet computer, a PDA, a personal computer, a wearable device, a smart TV, a PAD, etc.

[0085] It should be noted that Figure 1 This is just an example. There are many types of electronic devices, not limited to Figure 1 Laptops, smartphones, PADs, and personal computers.

[0086] Figure 1 Since the side of the vehicle 11 is shown, Figure 1 3 parking cameras are shown in FIG. However, the installation position and number of the parking cameras included in the vehicle are not limited.

[0087] Those skilled in the art should understand that the above-mentioned electronic devices and servers are only examples, and other existing or future electronic devices or servers that are applicable to the present disclosure should also be included in the protection scope of the present disclosure and are included here by reference.

[0088] Before testing the vehicle to be tested, the parking cameras arranged on the vehicle 11 need to be calibrated, that is, a transformation matrix between points in the image captured by each parking camera and points in the real environment coordinate system is obtained.

[0089] For example, the real environment coordinate system may be a coordinate system with the two intersecting sides of the parking space as the X-axis and the Y-axis, and the intersection of the two intersecting sides as the origin. Figure 2a shown.

[0090] For example, the real environment coordinate system may be a coordinate system with the parking camera as the origin, the X-axis is the same as the vehicle's driving direction, and the Z-axis is perpendicular to the ground. Different parking cameras correspond to different coordinate systems. Figure 2b shown.

[0091] In order to obtain the transformation matrix between the coordinate system in the image captured by each parking camera and the real environment coordinate system, a calibration plate needs to be prepared.

[0092] For example, the calibration plates are placed at positions where each parking camera can capture the image. Figure 3 shown.

[0093] Figure 3 In the description, a calibration plate 31 with black and white checkerboard is taken as an example. Figure 3 5 black and white checkerboard calibration plates are shown in FIG. The embodiment of the present application does not limit the number and position of the calibration plates arranged around the vehicle 11, as long as each parking camera can capture at least one calibration plate.

[0094] In order to obtain an accurate transformation matrix, the position of the calibration plate can be moved so that each parking camera can capture images with the calibration plate at different positions and angles. For example, each parking camera can capture 10 to 20 images, and the position and angle of the calibration plate relative to the parking camera in different images may be different.

[0095] Then, for each parking camera, based on the coordinates of the calibration plate in the image captured by the parking camera and the coordinates of the calibration plate in the real environment coordinate system, the intrinsic parameters, extrinsic parameters and distortion coefficients of the parking camera are estimated to obtain a transformation matrix.

[0096] For each parking camera, since the transformation matrix H between the coordinate system of the image captured by the parking camera and the real environment coordinate system is known, the distance D between point A and point B in the real environment coordinate system can be obtained based on the transformation matrix H and the distance d between point A and point B in the image.

[0097] Illustratively, D=H*d.

[0098] The parking performance testing method provided in the embodiment of the present application is described below in conjunction with the above hardware architecture.

[0099] like Figure 4 As shown, it is a flowchart of the parking performance testing method provided in an embodiment of the present application. The method can be applied to the server 12 or to the controller on the vehicle. The method includes the following steps S41 to S44 during implementation.

[0100] Step S41: When the vehicle to be tested for parking performance has completed parking, images collected by parking cameras arranged at different positions on the vehicle are obtained, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, and all the images involve at least two adjacent sides of the vehicle, and at least two of the preset target points are located on the same side of the vehicle and are arranged horizontally.

[0101] For the description of parking cameras at different positions on the vehicle, please refer to the description of the hardware architecture, which will not be repeated here.

[0102] For the description of the preset target points, please refer to the above description of the preset target points, which will not be repeated here.

[0103] Exemplarily, each image may display a preset target point and a parking space line segment on the same side as the preset target point. For example, the image displays the tangent point between the left front wheel and the ground and a parking space line segment on the left side of the parking space.

[0104] For example, each image may display a plurality of preset target points located on the same side of the vehicle and a section of parking space line located on the same side as the plurality of preset target points. If the same side is the left side, the image may display the tangent point between the left front wheel and the ground, the tangent point between the left rear vehicle and the ground, and a section of parking space line on the left side of the parking space.

[0105] It can be understood that the parking space line in the image has a certain length, and the length of the parking space line is related to the tilt angle of the vehicle and the distance between the preset target point of the wheel of the vehicle and the parking space line.

[0106] The "lateral setting" mentioned in the embodiments of the present application refers to setting along the direction of the vehicle length or width, the purpose of which is to determine the tilt angle of the vehicle.

[0107] Combine the following Figure 5 The direction indicated by the arrow in the middle is the traveling direction of the vehicle, which explains the “lateral setting”.

[0108] Exemplarily, the adjacent two sides of the vehicle mentioned in the embodiments of the present application refer to the left side in the driving direction and the rear side in the direction perpendicular to the driving direction, or, the adjacent two sides of the vehicle mentioned in the embodiments of the present application refer to the left side in the driving direction and the front side in the direction perpendicular to the driving direction, or, the adjacent two sides of the vehicle mentioned in the embodiments of the present application refer to the right side in the driving direction and the front side in the direction perpendicular to the driving direction, or, the adjacent two sides of the vehicle mentioned in the embodiments of the present application refer to the right side in the driving direction and the rear side in the direction perpendicular to the driving direction.

[0109] The following is an example in which the adjacent two sides of the vehicle mentioned in the embodiment of the present application refer to the left side in the driving direction and the rear side in the direction perpendicular to the driving direction. Then all images include: an image containing the tangent point of the left front wheel and the ground and a section of parking line on the left side of the parking space, an image containing the tangent point of the left rear wheel and the ground and a section of parking line on the left side of the parking space, and a mapping point containing the rear of the vehicle and the ground and a section of parking line behind the parking space.

[0110] The "at least two adjacent sides" mentioned in the embodiment of the present application include: at least one side on the left side of the driving direction and the right side of the driving direction, and at least one side on the front side in the direction perpendicular to the driving direction and the rear side in the direction perpendicular to the driving direction.

[0111] Exemplarily, the tail end is located at the rear side of the vehicle; the front end is located at the front side of the vehicle.

[0112] It is understandable that the number of preset target points located on the same side of the vehicle may be two or more, which is described below with examples.

[0113] It can be understood that if the vehicle is a car, there are two wheels on the left side of the car, namely the left front wheel and the left rear wheel, so there are two preset target points located on the same side and set horizontally; if the vehicle is a large truck, there are three wheels on the left side of the large truck, namely the left front wheel, the left middle wheel and the left rear wheel, so there are three preset target points located on the same side and set horizontally.

[0114] The following describes step S41 by taking a vehicle including four wheels as an example. Figure 5 , which is a schematic diagram of an image captured by a parking camera provided in an embodiment of the present application.

[0115] Figure 5 The description is given by taking an example that each image includes a preset target point.

[0116] The multiple images include: an image 51 including the intersection point of the left front wheel and the ground and a section of parking space lines on the left side of the parking space, an image 52 including the intersection point of the left rear wheel and the ground and a section of parking space lines on the left side of the parking space, an image 53 including the intersection point of the right front wheel and the ground and a section of parking space lines on the right side of the parking space, an image 54 including the intersection point of the right rear wheel and the ground and a section of parking space lines on the right side of the parking space, and an image 55 including a preset target point between the rear end of the vehicle and the ground and a section of parking space lines behind the parking space.

[0117] Figure 5 The preset target points are represented by solid dots.

[0118] In order to indicate the relative position of each image and the vehicle, Figure 5 A top view of the vehicle 11 is also shown.

[0119] If the parking camera is set above the object, since the parking camera is set above the wheel, above the front of the car, and above the rear of the car, a corresponding relationship between the parking camera and the object (the object is the wheel, the front of the car, or the rear of the car) can be established. Assume that the parking camera A1 set above the left front wheel corresponds to the left front wheel; the parking camera A2 set above the left rear wheel corresponds to the left rear wheel; the parking camera A3 set above the right front wheel corresponds to the right front wheel; the parking camera A4 set above the right rear wheel corresponds to the right rear wheel; the parking camera A5 set above the rear of the car corresponds to the rear of the car; and the camera A6 set above the front of the car corresponds to the front of the car.

[0120] Based on the preset correspondence between the parking cameras and the objects, it can be determined to which object the preset target points contained in the images captured by each parking camera belong. For example, in the embodiment of the present application, it is not necessary to classify and analyze the preset target points contained in the images captured by the parking cameras to determine which object in the vehicle the parking image contained in the image specifically corresponds to, thereby speeding up the image analysis speed.

[0121] In an optional implementation, when the field of view of the parking camera is constant, the shooting range of the parking camera is limited. In the embodiment of the present application, the parking camera is set above the corresponding object, and the image captured by the parking camera only includes a section of the parking space line located on the same side as the preset target point. This saves time in analyzing the image to obtain a section of the parking space line located on the same side as the preset target point.

[0122] In order to enable those skilled in the art to better understand the beneficial effects of the embodiments of the present application, the following examples are given to illustrate.

[0123] like Figure 6 , which is a schematic diagram of an image captured when the field of view of the parking camera is large.

[0124] from Figure 6It can be seen that if the parking camera located above the left front wheel is not accurately set on the vehicle, the field of view of the parking camera is too large, so that the image captured by the parking camera is Figure 6 As shown, due to Figure 6 The image shown contains two parking space lines, wherein parking space line 61 is a section of parking space line in front of the parking space, and parking space line 62 is a section of parking space line on the left side of the parking space. Figure 6 The image shown is analyzed, and then the parking space line 62 is determined to be located on the same side as the tangent point between the left front wheel and the ground.

[0125] In the embodiment of the present application, by cleverly setting the position of the parking camera in the vehicle and cleverly selecting a parking camera with a corresponding field of view, the image captured by the parking camera includes a preset target point and a section of the parking space line on the same side as the preset target point. For example, the image includes the parking space line 62 and the preset target point between the left front wheel and the ground, but does not include the parking space line 61. Therefore, the time for analyzing the image is saved and the speed of the parking performance test is improved.

[0126] Step S42: for each of the images, obtaining a first distance between the preset target point in the image and the section of parking space line, so as to obtain first distances corresponding to a plurality of the preset target points respectively.

[0127] Exemplarily, for each image, a perpendicular line segment of the parking space line may be drawn by using a preset target point, and the length of the perpendicular line segment is the first distance.

[0128] In an optional implementation, the first distance is a pixel distance in the image, and the pixel distance needs to be converted into an actual distance in a real environment coordinate system.

[0129] The actual distances corresponding to the preset target points may be obtained based on the transformation matrix H corresponding to the preset target points and the first distances corresponding to the preset target points.

[0130] Exemplarily, the actual distance of the preset target point=the transformation matrix corresponding to the preset target point*the first distance corresponding to the preset target point.

[0131] The first distance is described below in conjunction with the accompanying drawings. Figure 7 , which is a schematic diagram of a first distance between a preset target point and a parking space line provided in an embodiment of the present application.

[0132] like Figure 7As shown, assuming that the multiple preset target points include the tangent point 1 between the left front wheel and the ground in image 51, the tangent point 2 between the left rear wheel and the ground in image 52, the tangent point 3 between the right front wheel and the ground in image 53, the tangent point 4 between the right rear wheel and the ground in image 54, and the mapping point 5 between the rear of the vehicle and the ground in image 55, the first distance d between the tangent point 1 and a section of the parking space line in image 51 can be obtained. 1 ’ , the first distance d between the tangent point 2 and a section of the parking space line in the image 52 3 ’ , the first distance d between the tangent point 3 and a section of the parking space line in the image 53 2 ’ , a first distance d between the tangent point 4 and a section of the parking space line in the image 54 4 ’ , a first distance d between the mapping point 5 and a section of the parking space line in the image 55 5 ’ .

[0133] In an optional implementation, the first distance is an actual distance in a real environment coordinate system.

[0134] That is, the first distance is the actual distance obtained by converting the pixel distance in the image.

[0135] Step S43: Determine the inclination angle of the vehicle relative to the parking space based on the preset distance between at least two preset target points located on the same side of the vehicle and arranged laterally, and the first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged laterally.

[0136] Exemplarily, if the lateral setting refers to setting along the vehicle length direction, the preset distance between the at least two preset target points may be the distance between the tangent point of the left front wheel and the ground and the tangent point of the left rear wheel and the ground.

[0137] Exemplarily, the distance between the tangent point between the left front wheel and the ground and the tangent point between the left rear wheel and the ground can be obtained from the image.

[0138] Exemplarily, if the lateral setting refers to setting along the vehicle length direction, the preset distance between the at least two preset target points may be the vehicle wheelbase of the left front wheel and the left rear wheel.

[0139] Exemplarily, if the lateral setting refers to setting along the vehicle width direction, the preset distance between the at least two preset target points may be the distance between the tangent point between the left front wheel and the ground and the tangent point between the right front wheel and the ground.

[0140] Exemplarily, the distance between the tangent point between the left front wheel and the ground and the tangent point between the right front wheel and the ground can be obtained from the image.

[0141] Exemplarily, if the lateral arrangement refers to arrangement along the vehicle width direction, the preset distance between the at least two preset target points may be the vehicle wheelbase between the left front wheel and the right front wheel.

[0142] like Figure 7 As shown, Figure 7 The horizontal setting is taken as an example to explain the setting along the vehicle length direction.

[0143] like Figure 7 As shown, assuming that the first distance d 1 ’ Corresponding to the actual distance d 1 Assume that the first distance d 3 ’ Corresponding to the actual distance d 3 Assume that the first distance d 2 ’ Corresponding to the actual distance d 2 Assume that the first distance d 4 ’ Corresponding to the actual distance d 4 Assume that the first distance d 5 ’ Corresponding to the actual distance d 5 .

[0144] Figure 7 The preset distance between the tangent point between the left front wheel and the ground and the tangent point between the left rear wheel and the ground is L1. The preset distance between the tangent point between the right front wheel and the ground and the tangent point between the right rear wheel and the ground is L2.

[0145] In an optional implementation, if the wheels of the vehicle are not straightened after parking, the first angle between the left side of the vehicle and the left parking space line of the parking space is different from the first angle between the left side of the vehicle and the left parking space line of the parking space. Exemplarily, step S43 specifically includes the following steps S431 to S433.

[0146] Step S431: Calculate a first angle between the left side of the vehicle and the left parking space line of the parking space based on the vehicle wheelbase and first distances corresponding to the tangent points of the left front wheel and the ground and the tangent points of the left rear wheel and the ground.

[0147] For example, if the wheelbase of the vehicle is L1, the actual distance of the first distance corresponding to the tangent point between the left front wheel and the ground is d 1 , the actual distance of the first distance corresponding to the tangent point between the left rear wheel and the ground is d 3 , then the calculation formula of the first angle is as follows:

[0148]

[0149] Step S432: Calculate a second angle between the right side of the vehicle and the right parking space line of the parking space based on the vehicle wheelbase and the first distances corresponding to the tangent points of the right front wheel and the ground and the tangent points of the right rear wheel and the ground.

[0150] Assume that the vehicle wheelbase is L2 and the actual distance of the first distance corresponding to the tangent point of the right front wheel and the ground is d 2 , the actual distance of the first distance corresponding to the tangent point between the right rear wheel and the ground is d 4 , then the calculation formula for the second angle is as follows:

[0151]

[0152] Step S433: Determine the average of the first angle and the second angle as the inclination angle.

[0153] Exemplarily, the calculation formula of the tilt angle θ is as follows:

[0154]

[0155] In an optional implementation, if the wheels of the vehicle are straightened after parking, the first angle between the left side of the vehicle and the left parking space line of the parking space is the same as the first angle between the left side of the vehicle and the left parking space line of the parking space. 1 And the actual distance d 3 , determine the vehicle's tilt angle relative to the parking space.

[0156] For example, the tilt angle

[0157] Step S44: Obtain the parking time of the vehicle.

[0158] In an optional implementation, the parking time length = the time of obtaining the instruction of parking completion minus the time of detecting the instruction of the parking space where parking is available.

[0159] Exemplarily, obtaining the parking completion instruction refers to obtaining a vehicle locking instruction or obtaining a parking instruction.

[0160] Exemplarily, the instruction to detect a parking space where the vehicle can be parked refers to an instruction to find a parking space where the vehicle can be parked.

[0161] It is understandable that the position of the vehicle parked in the parking space can be determined based on the tilt angle. The time required for automatic parking to the corresponding position of the corresponding parking space can be determined by the parking time. If it takes a long time to reach an accurate position, it is also an inaccurate performance, so in the process of determining the parking performance of the vehicle, it is also necessary to combine the parking time.

[0162] Step S45: determining a first parking performance of the vehicle based on the first distances, the tilt angles, and the parking time durations respectively corresponding to the plurality of preset target points.

[0163] In an optional implementation, if the first distance corresponding to the preset target point is a pixel distance in the image, the pixel distance needs to be converted into an actual distance in the real environment coordinate system. The specific implementation of step S45 includes: calculating the product of the transformation matrix H corresponding to the plurality of preset target points and the corresponding first distance to obtain the actual distances corresponding to the plurality of preset target points; determining the parking performance of the vehicle based on the actual distances corresponding to the plurality of preset target points, the tilt angle, and the parking time.

[0164] In an optional implementation, if the first distance corresponding to the preset target point is the actual distance, step S45 includes: determining the first parking performance of the vehicle based on the first distances respectively corresponding to the plurality of preset target points, the tilt angle, and the parking time.

[0165] In the parking performance test method provided in the embodiment of the present application, after parking is completed, images collected by parking cameras arranged at different positions on the vehicle are obtained, each image shows a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, all images involve at least two adjacent sides of the vehicle, and at least two preset target points are located on the same side of the vehicle and are arranged horizontally; for each image, a first distance between the preset target point and the section of parking space line in the image is obtained to obtain first distances corresponding to multiple preset target points respectively; based on the preset distance between at least two preset target points located on the same side of the vehicle and arranged horizontally, and the first distances corresponding to at least two preset target points located on the same side of the vehicle and arranged horizontally, the inclination angle of the vehicle relative to the parking space is determined; thereby, the position of the vehicle parked in the parking space can be determined based on the first distances and inclination angles corresponding to multiple images respectively. The parking time can also be obtained. Based on the parking time, the time required for automatic parking at the corresponding position can be determined. If the time consumed is long, it is also a manifestation of inaccurate parking. Therefore, the parking performance of the vehicle can be determined by combining the first distance, tilt angle and parking time corresponding to multiple preset target points.

[0166] In an optional implementation, there are multiple implementations of step S42, and the embodiments of the present application provide but are not limited to the following two.

[0167] The first implementation of step S42 includes the following steps A11 to A15.

[0168] Step A11: binarizing the image according to the preset parking space line color to obtain a binary image.

[0169] In an optional implementation, the color of the parking space line is generally yellow, white or blue. Therefore, the colors of the preset parking space line include: yellow, white and blue.

[0170] Exemplarily, the pixel values ​​of pixels in the image whose pixel values ​​belong to the pixel range of the preset parking space line color can be set to (255, 255, 255), and the pixel values ​​of pixels in the image whose pixel values ​​do not belong to the pixel range of the preset parking space line color can be set to (0, 0, 0) to obtain a binary image.

[0171] In an optional implementation, there are multiple implementations of step A11. The embodiment of the present application provides but is not limited to the following method, which includes the following steps B10 to B12.

[0172] Step B10: Perform histogram equalization processing on the image to obtain a high-contrast image.

[0173] Exemplarily, in an optional implementation, the image may not be subjected to histogram equalization processing.

[0174] Step B11: convert the high contrast image from the RGB color space to the HSV color space to obtain an HSV image.

[0175] For example, if the image is not subjected to histogram equalization processing, the image can be directly converted from the RGB color space to the HSV (Hue-Saturation-Value) color space to obtain an HSV image.

[0176] The purpose of executing step B10 in the embodiment of the present application is to improve the contrast of the image to obtain a high-contrast image. After the high-contrast image is converted into an HSV image belonging to the HSV color space, the contrast of the obtained HSV image is also relatively obvious, thereby improving the accuracy of subsequent parking space line extraction.

[0177] Exemplarily, the high contrast image may be converted from the RGB color space to the HSV color space based on the following formula to obtain an HSV image.

[0178]

[0179] V=C max .

[0180] Among them, R', G' and B' are the normalized RGB values ​​of the high contrast image, C max and C min are the maximum and minimum values ​​of R', G' and B' respectively, and Δ is C max and C min difference.

[0181] Step B12: binarizing the HSV image according to the preset parking space line color to obtain a binary image.

[0182] It can be understood that the HSV image includes an image of the H channel, an image of the S channel, and an image of the V channel, and color thresholds can be set for the H channel, the S channel, and the V channel respectively based on the set parking line color. For example, if the parking line color is set to include white, yellow, or blue, the color thresholds set for the H channel are: [color threshold B1, color threshold B2] (for yellow), [color threshold B3, color threshold B4] (for white), [color threshold B5, color threshold B6] (for blue); the color thresholds set for the S channel are: [color threshold C1, color threshold C2] (for yellow), [color threshold C3, color threshold C4] (for white), [color threshold C5, color threshold C6] (for blue); the color thresholds set for the V channel are: [color threshold D1, color threshold D2] (for yellow), [color threshold D3, color threshold D4] (for white), [color threshold D5, color threshold D6] (for blue).

[0183] Then, color filtering can be performed on the H channel image, the S channel image, and the V channel image based on the color threshold, that is, the pixel values ​​of pixels whose pixel values ​​belong to the corresponding color threshold are set to (255, 255, 255), and the pixel values ​​of pixels whose pixel values ​​do not belong to the corresponding color threshold are set to (0, 0, 0) to obtain a binary image.

[0184] Step A12: Perform morphological closing and opening operations on the binary image to obtain a first image.

[0185] The purpose of performing morphological closing operations on binary images is to fill small spaces within objects, connect adjacent objects (for example, fill cracks in contours), eliminate small holes, and smooth boundaries.

[0186] It is understandable that, since some parking spaces are on the ground and some are underground, the parking spaces are exposed to wind and rain, and the parking lines of the parking spaces will be worn to a certain extent, such as the edge of the parking line is broken, and the inside of the parking line (the parking line has a certain width) has cracks. Therefore, through the morphological closing operation, the broken parts of the edge of the parking line can be connected, so that the edge of the parking line is smoother, and the cracked area inside the parking line can be connected, that is, the parking line does not include independent small spaces inside, so that the parking line as a whole is a single connected area.

[0187] The purpose of performing morphological opening operation on the binary image after morphological closing operation is to eliminate small objects in the binary image, separate objects in thin places and smooth the boundaries of larger objects.

[0188] It is understandable that there may be some objects (such as garbage) with colors close to the parking line next to the parking line. In order to accurately obtain the first distance between the preset target point and the parking line, it is necessary to accurately extract the boundary of the parking line. Therefore, in order to avoid the situation where the object A with a color close to the parking line may appear next to the parking line, resulting in inaccurate extraction of the boundary of the parking line, the binary image is processed by morphological opening operation.

[0189] Since the morphological opening operation can separate objects at thin places, that is, disconnect narrow necks and eliminate thin protrusions, the parking space line and object A in the binary image can be distinguished, so that the boundary of the extracted parking space line is more accurate.

[0190] In order to enable those skilled in the art to better understand the purpose of performing morphological closing and opening operations in the embodiments of the present application, examples are given below.

[0191] like Figures 8a to 8c , which is a schematic diagram of an image processed by performing morphological closing and opening operations provided in an embodiment of the present application.

[0192] like Figure 8a The image shown is a binary image corresponding to the image including the left front wheel and the parking space line.

[0193] Figure 8b right Figure 8a The binary image shown is processed by closing operation. Figure 8b Relative to Figure 8a For example, the fracture marks inside the parking space line disappear, but the noise outside the parking space line is amplified; Figure 8c For Figure 8b The image shown is the image after the opening operation. Figure 8c Relative to Figure 8b For example, the noise outside the parking space line is partially eliminated.

[0194] Step A13: Perform perspective transformation on the first image to obtain a top-view image.

[0195] Perspective transformation refers to the process of converting the image captured by the parking camera into a top-down view, also known as bird's-eye view transformation. The resulting image is also called a bird's-eye view. Generally, if the object (wheel, front or rear of the vehicle) and the parking space line are photographed from a bird's-eye view, the deformation of the object and the parking space line will be relatively small, while if they are photographed from other angles, the deformation of the object and the parking space line will be relatively large. When using the parking camera on the vehicle to capture the image of the object and the parking space line, it is obviously impossible to shoot from a bird's-eye view. Therefore, the image captured by the parking camera can be transformed in perspective, that is, the shooting angle of the object and the parking space line is converted to minimize the distortion of the shape of the object and the parking space line in the image.

[0196] like Fig. 9 As shown, Figure 8c The diagram is a schematic diagram of a top view image after perspective transformation of the first image shown.

[0197] pass Fig. 9 The top view shown is similar to Figure 8c By comparing with the first image shown, it can be seen that the parking space line has changed from a curve to a straight line, that is, distortion correction has been performed.

[0198] Step A14: Detect straight lines in the overhead image, and identify two straight lines whose distances are within a preset distance range and whose included angles are smaller than a preset included angle as parking space line boundaries.

[0199] In an optional implementation, there are multiple implementations of step A14. The embodiment of the present application provides but is not limited to the following method, which includes the following steps C11 to C13.

[0200] Step C11: Perform edge detection on the overhead image to obtain a second image.

[0201] Exemplarily, edge detection is performed on the overhead image to identify points with obvious brightness changes in the overhead image, such as the boundaries of parking spaces. Exemplarily, the edge detection may be a Canny edge algorithm.

[0202] like Fig.10 Shown is a schematic diagram of an implementation method of the second image provided in an embodiment of the present application.

[0203] Fig.10 For the general Fig. 9 The second image is obtained after edge detection of the top view image shown. Fig.10 It can be seen from the second image that the second image contains edge lines, for example, the edge of the parking space line.

[0204] Step C12: Detect straight lines in the second image using a Hough transform algorithm.

[0205] In some embodiments, the Hough transform algorithm may be directly used to detect the straight line in the second image. In other embodiments, other straight line detection algorithms may also be used to detect the straight line in the second image.

[0206] Exemplarily, it can be understood that the boundary of the parking space line is a straight line rather than a curve, and the straight lines in the second image are detected using the Hough transform algorithm, and these straight lines may be the parking space lines.

[0207] like Fig.11 , which is a schematic diagram of a straight line detected from a second image provided in an embodiment of the present application.

[0208] from Fig.11 It can be seen that the boundary of the wheel (curve) has been removed, and the boundary of the parking space line (straight line) has been retained.

[0209] Step C13: Filtering out the parking space line boundary from the straight lines in the second image according to a preset distance range and a preset angle.

[0210] It is understandable that, since the parking space line has a certain width, the parking space line has two boundaries. Therefore, based on this feature of the parking space line, it is necessary to screen out a straight line pair from the straight lines in the second image, and the straight line pair includes two straight lines.

[0211] Exemplarily, a plurality of straight line pairs can be screened out from the straight lines in the second image; for each straight line pair, the distance and angle between two straight lines in the straight line pair are obtained; and straight line pairs whose distances fall within a preset distance range and whose angles are less than the preset angles are obtained to obtain the boundaries of the parking space lines.

[0212] In an optional implementation, the width of the parking space line is generally 10 cm to 18 cm. The width range of the parking space line in the image can be determined based on the transformation matrix between the image and the real environment coordinate system, and the width range can be determined as a preset distance range.

[0213] In an optional implementation, the distance between the straight line pairs in the real environment coordinate system = the pixel distance between the straight line pairs in the image * the transformation matrix. The width of the parking space line is generally 10 cm to 18 cm, and [10, 18] can be used as the preset distance range.

[0214] It is understandable that although the image is perspective transformed, the distortion of the parking space lines may not be completely eliminated, so the boundaries of the parking space lines in the parking space line boundary candidate image may not be parallel, so the preset angle may not be 0, for example, it may be 2°.

[0215] Step A15: taking the distance between the preset target point and the parking space line boundary as the first distance.

[0216] Exemplarily, the distance between a straight line close to the preset target point in the parking space line boundary and the preset target point is determined as the first distance.

[0217] like Figure 5 As shown, the parking space line boundary includes two straight lines, one of which is close to the preset target point and the other is far away from the preset target point. In the embodiment of the present application, the distance between the straight line close to the preset target point in the parking space line boundary and the preset target point is the first distance.

[0218] Through the above steps A11 to A15, the parking space line boundary extracted from the image is more accurate, so the calculated first distance is more accurate, making the parking performance test more accurate. The above steps A11 to A15 have strong robustness, and can accurately identify the parking space line boundary even in the case of blurred parking space line wear and dim light, and can achieve centimeter-level measurement accuracy, thereby providing an accurate parking performance evaluation indicator.

[0219] The second implementation of step S42 includes the following steps A21 to A25.

[0220] Step A21: Perform perspective transformation on the image to obtain a top-view image.

[0221] The purpose of performing perspective change on the image can be referred to the purpose of performing perspective change on the first image, which will not be described in detail here.

[0222] Step A22: binarizing the top view image according to the preset parking space line color to obtain a binary image.

[0223] The process of binarizing the overhead image can refer to the process of binarizing the image, which will not be repeated here.

[0224] Step A23: Perform morphological closing and opening operations on the binary image to obtain a first image.

[0225] For the description of step A23, please refer to the description of step A12, which will not be repeated here.

[0226] Step A24: Detect the straight lines in the first image, and identify two straight lines whose distances belong to a preset distance range and whose included angles are smaller than a preset included angle as parking space line boundaries.

[0227] For the description of step A24, please refer to the description of step A14, which will not be repeated here.

[0228] Step A25: taking the distance between the preset target point and the parking space line boundary as the first distance.

[0229] For the description of step A25, please refer to the description of step A15, which will not be repeated here.

[0230] In an optional implementation, there are multiple ways to obtain the preset target point contained in the image, and the embodiments of the present application provide but are not limited to the following two.

[0231] The first implementation method includes: for each image, extracting the outer contour of the object in the image; determining the plane where the ground in the image is located; and determining the tangent point between the outer contour and the plane where the ground is located as the tangent point.

[0232] The second implementation method includes the following steps D11 to D13.

[0233] Step D11: Perform perspective transformation on the image to obtain a bird's-eye view image.

[0234] The image is an image captured by a camera.

[0235] Step D12: controlling the display of the bird's-eye view image.

[0236] Step D13: Acquire a bird's-eye view image containing a preset target point.

[0237] For example, the user may manually mark a preset target point in the bird's-eye view image.

[0238] In an optional implementation, in order to accurately test and evaluate the performance of the vehicle, the vehicle can be automatically parked multiple times, so that multiple first distances, multiple tilt angles, and multiple parking time durations corresponding to each of the preset target points can be obtained. In this way, the parking performance of the vehicle can be generally evaluated based on the first distances, multiple tilt angles, and multiple parking time durations corresponding to the multiple preset target points obtained by the multiple automatic parkings. The specific method includes the following steps E11 to E14.

[0239] Step E11: averaging the first distances corresponding to the plurality of preset target points obtained by parking the vehicle multiple times to obtain the first average distances corresponding to the plurality of preset target points.

[0240] Assume that the multiple preset target points include: point 1 between the left front wheel and the ground in image 51, point 2 between the left rear wheel and the ground in image 52, point 3 between the right front wheel and the ground in image 53, point 4 between the right rear wheel and the ground in image 54, and mapping point 5 between the rear of the vehicle and the ground in image 55; if the vehicle is automatically parked five times, then five first distances corresponding to point 1, five first distances corresponding to point 2, five first distances corresponding to point 3, five first distances corresponding to point 4, and five first distances corresponding to mapping point 5 can be obtained, and five tilt angles can also be obtained.

[0241] For example, the average of the five first distances corresponding to the tangent point 1 can be calculated to obtain the first average distance corresponding to the tangent point 1. Find the average of the five first distances corresponding to tangent point 2 to get the first average distance corresponding to tangent point 2 The average of the five first distances corresponding to tangent point 3 is used to obtain the first average distance corresponding to tangent point 3. The average of the five first distances corresponding to the tangent point 4 is used to obtain the first average distance corresponding to the tangent point 4. The average of the five first distances corresponding to mapping point 5 is used to obtain the first average distance corresponding to mapping point 5.

[0242] Step E12: averaging the multiple tilt angles obtained when the vehicle is parked multiple times to obtain an average tilt angle.

[0243] Step E13: averaging the parking time taken by the vehicle during multiple parking operations to obtain an average parking time taken.

[0244] Step E14: Determine a second parking performance of the vehicle based on the first average distance, the average tilt angle, and the average parking time.

[0245] For example, the average tilt angle is like Both are greater than 0.1m, and If it is between -3° and 3°, the accuracy of determining the parking performance is higher.

[0246] For example, the above distance of 0.1 m and the angle of -3° to 3° are just examples and are not limiting.

[0247] The first parking performance is used to measure the quality of each parking, and the second parking performance is used to measure the stability of parking.

[0248] The method is described in detail in the embodiments disclosed in the above-mentioned application. The method of the application can be implemented by various forms of devices. Therefore, the application also discloses a device, and a specific embodiment is given below for detailed description.

[0249] like Fig.12 , which is a structural diagram of a parking performance test device provided in an embodiment of the present application, the parking performance test device includes: a first acquisition module 121, a second acquisition module 122, a first determination module 123, a third acquisition module 124 and a second determination module 125, wherein:

[0250] A first acquisition module 121 is used for acquiring images collected by parking cameras arranged at different positions on the vehicle when the vehicle to be tested for parking performance has finished parking, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, and all the images at least involve two adjacent sides of the vehicle, and at least two of the preset target points are located on the same side of the vehicle and are arranged horizontally;

[0251] A second acquisition module 122 is used to acquire, for each of the images, a first distance between the preset target point in the image and the section of the parking space line, so as to obtain first distances corresponding to a plurality of the preset target points respectively;

[0252] A first determining module 123, configured to determine a tilt angle of the vehicle relative to the parking space based on a preset distance between at least two preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged transversely;

[0253] A third acquisition module 124 is used to acquire the parking time of the vehicle;

[0254] The second determination module 125 is configured to determine a first parking performance of the vehicle based on the first distances, the tilt angles, and the parking time durations respectively corresponding to the plurality of preset target points.

[0255] In an optional implementation, the second acquisition module includes:

[0256] A first acquisition unit, configured to perform binarization processing on the image according to a preset parking space line color to obtain a binarized image;

[0257] A second acquisition unit is used to perform morphological closing and opening operations on the binary image to obtain a first image;

[0258] A third acquisition unit, configured to perform perspective transformation on the first image to obtain a bird's-eye view image;

[0259] A first recognition unit is used to detect straight lines in the overhead image, and recognize two straight lines whose distance is within a preset distance range and whose included angle is less than a preset included angle as parking space line boundaries;

[0260] The first determining unit is used to take the distance between the preset target point and the parking space line boundary as the first distance.

[0261] In an optional implementation, the first acquiring unit includes:

[0262] A first acquisition subunit, used for performing histogram equalization processing on the image to obtain a high-contrast image;

[0263] A first conversion subunit, configured to convert the high contrast image from an RGB color space to an HSV color space to obtain an HSV image;

[0264] The second acquisition subunit is used to perform binarization processing on the HSV image according to a preset parking space line color to obtain a binary image.

[0265] In an optional implementation, the first identification unit includes:

[0266] A third acquisition subunit is used to perform edge detection on the overhead image to obtain a second image;

[0267] A detection subunit, configured to detect straight lines in the second image using a Hough transform algorithm;

[0268] The screening subunit is used to screen out the parking space line boundary from the straight lines in the second image according to a preset distance range and a preset angle.

[0269] In an optional implementation, the preset target points set on the left and right sides of the vehicle include: a tangent point between a left front wheel and the ground, a tangent point between a left rear wheel and the ground, a tangent point between a right front wheel and the ground, and a tangent point between a right rear wheel and the ground, and the first determination module includes:

[0270] a first calculation unit, configured to calculate a first angle between the left side of the vehicle and the left parking space line of the parking space based on the wheelbase of the vehicle and first distances corresponding to the tangent points of the left front wheel and the ground and the tangent points of the left rear wheel and the ground, respectively;

[0271] a second calculation unit, configured to calculate a second angle between the right side of the vehicle and the right parking space line of the parking space based on the wheelbase of the vehicle and first distances corresponding to the tangent points of the right front wheel and the ground and the tangent points of the right rear wheel and the ground, respectively;

[0272] The second determining unit is used to determine the average of the first angle and the second angle as the inclination angle.

[0273] In an optional implementation, the method further includes:

[0274] A fourth acquisition module, configured to average the first distances respectively corresponding to the plurality of preset target points obtained by the vehicle parking multiple times, to obtain the first average distances respectively corresponding to the plurality of preset target points;

[0275] A fifth acquisition module, configured to average a plurality of the tilt angles obtained by parking the vehicle multiple times to obtain an average tilt angle;

[0276] A sixth acquisition module, configured to average the parking time durations acquired from multiple parking operations of the vehicle to obtain an average parking time duration;

[0277] The third determination module is configured to determine a second parking performance of the vehicle based on the first average distance, the average tilt angle, and the average parking time.

[0278] In an optional implementation, the second acquisition module includes:

[0279] A fourth acquisition unit, used for performing perspective transformation on the image to obtain a bird's-eye view image;

[0280] A fifth acquisition unit, configured to perform a binarization process on the overhead image according to a preset parking space line color to obtain a binarized image;

[0281] a sixth acquisition unit, configured to perform morphological closing and opening operations on the binary image to obtain a first image;

[0282] A second recognition unit is used to detect straight lines in the first image, and recognize two straight lines whose distance is within a preset distance range and whose included angle is less than a preset included angle as parking space line boundaries;

[0283] The third determining unit is configured to use the distance between the preset target point and the parking space line boundary as the first distance.

[0284] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0285] Fig.13 The invention is a structural block diagram of a server according to an exemplary embodiment.

[0286] The server includes, but is not limited to, a processor 131 , a memory 132 , a network interface 133 , an I / O controller 134 , and a communication bus 135 .

[0287] It should be noted that those skilled in the art can understand that Fig.13 The server structure shown in the figure does not constitute a limitation on the server, and the server may include Fig.13More or fewer components may be shown, or certain components may be combined, or the components may be arranged differently.

[0288] Combine the following Fig.13 A detailed introduction to each component of the server:

[0289] The processor 131 is the control center of the server. It uses various interfaces and lines to connect various parts of the entire server. It executes various functions of the server and processes data by running or executing software programs and / or modules stored in the memory 132, and calling data stored in the memory 132, so as to monitor the server as a whole. The processor 131 may include one or more processing units; illustratively, the processor 131 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 131.

[0290] The processor 131 may be a central processing unit (CPU), or an application specific integrated circuit ASIC (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application;

[0291] The memory 132 may include memory, such as high-speed random access memory (RAM) 1321 and read-only memory (ROM) 1322, and may also include a large-capacity storage device 1323, such as at least one disk storage, etc. Of course, the server may also include hardware required for other services.

[0292] The memory 132 is used to store instructions executable by the processor 131. The processor 131 has the function of executing the parking performance test method.

[0293] A wired or wireless network interface 133 is configured to connect the server to a network.

[0294] The processor 131, the memory 132, the network interface 133 and the I / O controller 134 may be interconnected via a communication bus 135, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc.

[0295] In an exemplary embodiment, the server 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-mentioned parking performance testing method.

[0296] In an exemplary embodiment, the present disclosure provides a storage medium including instructions, such as a memory 132 including instructions, and the instructions can be executed by a processor 131 of a server to complete the above method. Optionally, the storage medium can be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0297] In an exemplary embodiment, a computer-readable storage medium is also provided, which can be directly loaded into the internal memory of a computer, such as the above-mentioned memory 132, and contains software code. After being loaded and executed by a computer, the computer program can implement the steps shown in any embodiment of the above-mentioned parking performance test method.

[0298] In an exemplary embodiment, a computer program product is also provided, which can be directly loaded into the internal memory of a computer, such as the memory included in the server, and contains software code. After being loaded and executed by the computer, the computer program can implement the steps shown in any embodiment of the parking performance test method described above.

[0299] It should be noted that the features described in the various embodiments in this specification can be replaced or combined with each other. For the device or system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0300] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0301] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0302] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A parking performance testing method, It is characterized in that include: When the vehicle to be tested for parking performance has finished parking, images collected by parking cameras located at different positions on the vehicle are obtained, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, and all the images at least involve two adjacent sides of the vehicle, and at least two of the preset target points are located on the same side of the vehicle and are arranged transversely; the transverse arrangement refers to being arranged along the length or width of the vehicle; For each of the images, obtaining a first distance between the preset target point in the image and the section of the parking space line, so as to obtain first distances respectively corresponding to a plurality of the preset target points; Determining the tilt angle of the vehicle relative to the parking space based on a preset distance between at least two preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged transversely; Obtaining the parking time of the vehicle; A first parking performance of the vehicle is determined based on the first distances, the tilt angles, and the parking time durations respectively corresponding to the plurality of preset target points.

2. The parking performance testing method according to claim 1, It is characterized in that The obtaining a first distance between the preset target point and the section of parking space line in the image includes: Binarizing the image according to the preset parking space line color to obtain a binary image; Performing morphological closing and opening operations on the binary image to obtain a first image; Performing perspective transformation on the first image to obtain a bird's-eye view image; Detecting straight lines in the overhead image, and identifying two straight lines whose distances are within a preset distance range and whose included angles are less than a preset included angle as parking space line boundaries; The distance between the preset target point and the parking space line boundary is taken as the first distance.

3. The parking performance testing method according to claim 2, It is characterized in that The binarization process is performed on the image according to the preset parking space line color to obtain a binarized image, including: Performing histogram equalization processing on the image to obtain a high-contrast image; Convert the high contrast image from the RGB color space to the HSV color space to obtain an HSV image; The HSV image is binarized according to the preset parking space line color to obtain a binarized image.

4. The parking performance testing method according to claim 2 or 3, It is characterized in that The detecting straight lines in the overhead image and identifying two straight lines whose distances are within a preset distance range and whose included angles are less than a preset included angle as parking space line boundaries includes: Performing edge detection on the overhead image to obtain a second image; Detecting straight lines in the second image using a Hough transform algorithm; The parking space line boundary is screened out from the straight lines in the second image according to a preset distance range and a preset angle.

5. The parking performance testing method according to any one of claims 1 to 3, It is characterized in that The preset target points arranged on the left and right sides of the vehicle include: a tangent point between a left front wheel and the ground, a tangent point between a left rear wheel and the ground, a tangent point between a right front wheel and the ground, and a tangent point between a right rear wheel and the ground. Based on a preset distance between at least two of the preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two of the preset target points located on the same side of the vehicle and arranged transversely, determining an inclination angle of the vehicle relative to a parking space includes: Calculate a first angle between the left side of the vehicle and the left parking space line of the parking space based on the vehicle wheelbase and the first distances corresponding to the tangent points of the left front wheel and the ground and the tangent points of the left rear wheel and the ground; Calculating a second angle between the right side of the vehicle and the right parking space line of the parking space based on the vehicle wheelbase and the first distances corresponding to the tangent points of the right front wheel and the ground and the tangent points of the right rear wheel and the ground; An average value of the first angle and the second angle is determined as the inclination angle.

6. The parking performance testing method according to any one of claims 1 to 3, It is characterized in that Also includes: averaging the first distances corresponding to the plurality of preset target points obtained by the vehicle parking multiple times to obtain the first average distances corresponding to the plurality of preset target points; averaging a plurality of the tilt angles obtained by parking the vehicle multiple times to obtain an average tilt angle; Averaging the parking time durations obtained from multiple parking operations of the vehicle to obtain an average parking time duration; A second parking performance of the vehicle is determined based on the first average distance, the average tilt angle, and the average parking time.

7. The parking performance testing method according to claim 1, It is characterized in that The obtaining a first distance between the preset target point and the section of parking space line in the image includes: Performing perspective transformation on the image to obtain a bird's-eye view image; Binarizing the top view image according to the preset parking space line color to obtain a binary image; Performing morphological closing and opening operations on the binary image to obtain a first image; Detect straight lines in the first image, and identify two straight lines whose distance is within a preset distance range and whose included angle is less than a preset included angle as parking space line boundaries; The distance between the preset target point and the parking space line boundary is taken as the first distance.

8. A parking performance testing device, It is characterized in that include: A first acquisition module is used for acquiring images collected by parking cameras arranged at different positions on the vehicle when the vehicle to be tested for parking performance has finished parking, each of the images showing a preset target point located on one side of the vehicle and a section of parking space line on the same side as the preset target point, all the images at least involve two adjacent sides of the vehicle, at least two of the preset target points are located on the same side of the vehicle and are arranged transversely; the transverse arrangement refers to being arranged along the length or width of the vehicle; A second acquisition module is used to acquire, for each of the images, a first distance between the preset target point in the image and the section of the parking space line, so as to obtain first distances corresponding to a plurality of the preset target points respectively; a first determining module, configured to determine a tilt angle of the vehicle relative to the parking space based on a preset distance between at least two preset target points located on the same side of the vehicle and arranged transversely, and first distances respectively corresponding to at least two preset target points located on the same side of the vehicle and arranged transversely; A third acquisition module is used to acquire the parking time of the vehicle; The second determination module is used to determine the first parking performance of the vehicle based on the first distances, the tilt angles and the parking time durations respectively corresponding to the plurality of preset target points.

9. A server, It is characterized in that include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the parking performance testing method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instructions are executed by a processor, the parking performance testing method according to any one of claims 1 to 7 is implemented.

Images