Parking space identification method, identification system, electronic device and storage medium

Abstract

The present invention provides a parking space recognition method, recognition system, electronic device, and storage medium. The method includes: acquiring an image containing a parking space, and identifying all straight line segments in the image based on edge extraction and line detection; splicing the straight line segments to form a parking space frame, and obtaining multiple first frame points of the parking space frame; taking any first frame point as the center, capturing a corresponding frame correction area in the image, performing corner point detection on the frame correction area, and obtaining a second frame point detected by the corner point detection; voting on the first and second frame points based on the image characteristics of the parking space to obtain a final frame point; and determining the parking space based on the multiple final frame points obtained by voting. The present invention integrates line detection and corner point detection to identify parking spaces, thereby improving the accuracy of parking space detection and recognition.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and more specifically, to a parking space recognition method, recognition system, electronic equipment and storage medium. Background Art

[0002] Automatic parking is a process in which the vehicle to be parked uses on-board sensors to collect information about the vehicle's surroundings, transmits the collected information to the perception module for analysis to obtain the target point, and the planning module calculates the parking trajectory based on the target point analyzed by the perception module. Finally, the control module controls the vehicle to park in the parking space.

[0003] Because cameras are relatively inexpensive and capture a wealth of information about the vehicle's surroundings, most current perception and recognition algorithms rely on vision. Existing vision algorithms, such as those based on deep learning, require high hardware requirements and take a long time to recognize parking spaces. Traditional algorithms, on the other hand, are easy to implement and require less hardware, which helps reduce costs. Because parking spaces are composed of straight lines, most existing vision-based parking space detection algorithms primarily rely on straight-line detection. However, these algorithms can be susceptible to misidentification due to complex site conditions (e.g., shadows, worn parking space lines, and other auxiliary lines). Summary of the Invention

[0004] The present invention addresses the problem of misidentification in parking space detection that relies on straight line detection in the prior art and provides a parking space recognition method, a recognition system, an electronic device and a storage medium.

[0005] According to a first aspect of the present invention, a parking space identification method is provided, comprising:

[0006] Acquire an image containing a parking space, and identify all straight line segments in the image based on edge extraction and line detection;

[0007] splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining a plurality of first parking frame points of the parking space frame;

[0008] Taking any first vehicle frame point as the center, screenshot a corresponding vehicle frame correction area in the image, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point for corner point detection;

[0009] Voting on the first and second parking frame points based on the image characteristics of the parking space to obtain a final parking frame point;

[0010] The parking space is determined based on the multiple final parking frame points obtained by voting.

[0011] On the basis of the above technical solution, the present invention can also make the following improvements.

[0012] Optionally, acquiring an image containing a parking space includes:

[0013] Acquire an image containing the parking space based on camera photography, and convert the image from an original perspective to a bird's-eye view;

[0014] The method of identifying all straight line segments in the image based on edge extraction and line detection includes:

[0015] The noise points around the parking space lines in the top-view image are processed based on adaptive binarization and morphology, and all straight line segments in the image are detected based on edge extraction and line detection methods.

[0016] Optionally, the step of splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining a plurality of first parking frame points of the parking space frame, includes:

[0017] According to the length, width and angle of the standard parking space, all straight line segments are grouped and screened, and the parking space is spliced, wherein the intersection of the long side and the short side of the parking space frame is the first parking frame point.

[0018] Optionally, performing corner detection on the vehicle frame correction area to obtain a second vehicle frame point obtained by corner point detection includes:

[0019] Gray-scaling the vehicle frame correction area and eliminating noise based on Gaussian filtering;

[0020] Perform Harris corner detection on the vehicle frame correction area where noise is eliminated to obtain the second vehicle frame point.

[0021] Optionally, voting on the first vehicle frame point and the second vehicle frame point according to image characteristics of the parking space to determine a final vehicle frame point includes:

[0022] Extracting eight neighboring pixels of the first vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel and the first vehicle frame point conforms to a pattern of parking space changes; if so, voting for any neighboring pixel; otherwise, not voting; and obtaining the number of votes for the eight neighboring pixels corresponding to the first vehicle frame point;

[0023] Extracting eight neighboring pixels of the second vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel and the first vehicle frame point conforms to a pattern of parking space changes; if so, voting for any neighboring pixel; otherwise, not voting; and obtaining the number of votes for the eight neighboring pixels corresponding to the first vehicle frame point;

[0024] The first or second car frame point with the most tickets will be taken as the final car frame point.

[0025] According to a second aspect of the present invention, there is provided a parking space recognition system, comprising:

[0026] A recognition module is used to obtain an image containing a parking space and identify all straight line segments in the image based on edge extraction and line detection;

[0027] a first acquisition module, configured to splice the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and to acquire a plurality of first parking frame points of the parking space frame;

[0028] a second acquisition module, configured to capture a corresponding vehicle frame correction area in the image with any first vehicle frame point as the center, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point detected by corner point detection;

[0029] a first determining module, configured to vote on the first vehicle frame point and the second vehicle frame point according to image characteristics of the parking space to determine a final vehicle frame point;

[0030] The second determination module is used to determine a parking space based on multiple final vehicle frame points obtained through voting.

[0031] Optionally, the first acquisition module is configured to splice the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and to acquire a plurality of first parking frame points of the parking space frame, including:

[0032] According to the length, width and angle of the standard parking space, all straight line segments are grouped and screened, and the parking space is spliced, wherein the intersection of the long side and the short side of the parking space frame is the first parking frame point.

[0033] According to a third aspect of the present invention, an electronic device is provided, comprising a memory and a processor, wherein the processor is configured to implement the steps of a parking space identification method when executing a computer management program stored in the memory.

[0034] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer management program is stored. When the computer management program is executed by a processor, the steps of the parking space identification method are implemented.

[0035] The present invention provides a parking space recognition method, recognition system, electronic device and storage medium, which intercept a portion of the area of ​​a straight line detection parking space for corner point detection, correct the misidentification of the straight line detection, and identify the parking space by integrating the straight line detection and corner point detection, thereby improving the accuracy of parking space detection and recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flow chart of a parking space identification method provided by the present invention;

[0037] Figure 2 This is a schematic diagram of the first car frame point;

[0038] Figure 3 This is a schematic diagram of the frame correction area;

[0039] Figure 4 Schematic diagram of detected corner points;

[0040] Figure 5 A schematic structural diagram of a parking space recognition system provided by the present invention;

[0041] Figure 6 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0042] Figure 7 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION

[0043] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0044] In order to solve the problem of misidentification in the current parking space detection that relies on straight line detection, the present invention provides a parking space recognition method that integrates straight line detection and corner point detection, intercepts part of the area of ​​the straight line detection parking space for corner point detection, and corrects misidentification.

[0045] Figure 1 A flow chart of a parking space identification method provided by the present invention is as follows: Figure 1 As shown, the parking space recognition method includes the following steps:

[0046] S1, acquiring an image containing a parking space, and identifying all straight line segments in the image based on edge extraction and line detection.

[0047] As an embodiment, the obtaining of an image containing a parking space includes: obtaining an image containing a parking space based on camera shooting, and converting the image from an original perspective to a bird's-eye view; the identifying of all straight line segments in the image based on edge extraction and line detection includes: processing noise points around the parking space line in the image from a bird's-eye view based on adaptive binarization and morphological processing, and detecting all straight line segments in the image based on edge extraction and line detection methods.

[0048] It is understandable that the camera is calibrated, an image containing a parking space is captured using the calibrated camera, the image is dedistorted, and the image is converted from an original perspective to a bird's-eye view.

[0049] Because the parking space lines have little contrast with the ground and other objects are present in the image, directly fitting the parking space lines is ineffective. Therefore, the image needs to be enhanced to segment the parking space lines. Adaptive binarization and morphological processing are used to smooth the noise around the parking space lines. Edge extraction and line detection are then used to detect the parking space line segments. Due to factors such as shadows or wear on the parking space lines, the entire parking space edge may be discontinuous, appearing as intermittent straight line segments. Therefore, in the embodiments of the present invention, edge extraction and line detection methods are used to extract all straight line segments in the image.

[0050] S2: splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining a plurality of first parking frame points of the parking space frame.

[0051] As an embodiment, the step of splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining multiple first frame points of the parking space frame, includes grouping and filtering all straight line segments according to the length, width and angle of the standard parking space, and splicing the parking space, wherein the intersection of the long side and the short side of the parking space frame is the first frame point.

[0052] It is understandable that the detected straight line segments are grouped and filtered, and the parking space lines are spliced ​​according to the angle and length of the standard parking space, and the intersection of the long side and the short side of the car frame is set as the car frame point, such as Figure 2 As shown, the vehicle frame point obtained by the intersection of the long side and the short side of the vehicle frame is called the first vehicle frame point.

[0053] S3, taking any first vehicle frame point as the center, screenshot a corresponding vehicle frame correction area in the image, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point detected by corner point detection.

[0054] It is understandable that, according to step S2, multiple first vehicle frame points are obtained. Since there is a certain probability of misidentification in the straight line detection of parking spaces, in this step, for each first vehicle frame point, a correction area is intercepted in the color image according to the length of a line width with the vehicle frame point as the center point, such as Figure 3 As shown. Grayscale the corrected area, and then select Gaussian filter to eliminate noise. Harris corner detection is performed on the corrected area with noise eliminated, and the detected corner points are retained, which are called the second car frame points, as shown Figure 4 As shown, for each first vehicle frame point, there is a corresponding second vehicle frame point.

[0055] S4: Voting on the first vehicle frame point and the second vehicle frame point according to image characteristics of the parking space to obtain a final vehicle frame point.

[0056] It is understandable that, for the first vehicle frame point and the second vehicle frame point, it is necessary to determine which vehicle frame point is more accurate. The present invention uses a voting method to determine the final vehicle frame point.

[0057] As an embodiment, voting on the first vehicle frame point and the second vehicle frame point based on image characteristics of the parking space to determine the final vehicle frame point includes: extracting eight neighboring pixels of the first vehicle frame point, sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a pattern of parking space changes; if so, voting for any neighboring pixel point; otherwise, not voting; obtaining the number of votes from the eight neighboring pixels corresponding to the first vehicle frame point; extracting eight neighboring pixels of the second vehicle frame point, sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a pattern of parking space changes; if so, voting for any neighboring pixel point; otherwise, not voting; obtaining the number of votes from the eight neighboring pixels corresponding to the first vehicle frame point; and selecting the first vehicle frame point or the second vehicle frame point with the greater number of votes as the final vehicle frame point.

[0058] It can be understood that, for the first vehicle frame point, 8 neighborhood points of the first vehicle frame point are obtained with the first vehicle frame point as the center, as shown in Table 1 below.

[0059] Table 1 Eight neighborhood pixels of the first car frame point

[0060] A B C D E F G H I

[0061] In Table 1, E is the first car frame point. The eight neighboring points A, B, C, D, F, G, H, and I, centered around E, are obtained. The pixel difference between any neighboring pixel and the first car frame point is sequentially determined to see if it conforms to the parking space variation pattern. If AE ≤ 0, then the pixel difference between A and E conforms to the parking space variation pattern, and point A is voted for. Similarly, if BE ≤ 0, CE ≥ 0, DE ≤ 0, FE ≥ 0, GE ≥ 0, HE ≥ 0, and IE ≥ 0, then all points conform to the parking space variation pattern, and the corresponding point is voted for. If any point does not conform to the parking space variation pattern, the point is not voted for. For the second car frame point, the eight neighboring points are voted for in the same manner.

[0062] The number of votes for the first and second vehicle frame points is determined by the above voting method, and the vehicle frame point with the largest number of votes is used as the final vehicle frame point. For each first and second vehicle frame point, the final vehicle frame point is determined by voting, resulting in multiple final vehicle frame points.

[0063] S5: Determine the parking space based on the multiple final parking frame points obtained by voting.

[0064] It is understandable that, based on the finally obtained vehicle frame points, the parking space frame is usually rectangular, and the final parking space frame is determined based on the four vehicle frame points.

[0065] Figure 5 A parking space recognition system structure diagram provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, a parking space recognition system includes a recognition module 51, a first acquisition module 52, a second acquisition module 53, a first determination module 54 and a second determination module 55, wherein:

[0066] The recognition module 51 is used to obtain an image containing a parking space and identify all straight line segments in the image based on edge extraction and line detection;

[0067] A first acquisition module 52 is configured to splice the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and to acquire a plurality of first parking frame points of the parking space frame;

[0068] A second acquisition module 53 is configured to capture a corresponding vehicle frame correction area in the image with any first vehicle frame point as the center, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point detected by corner point detection;

[0069] A first determining module 54 is configured to vote on the first and second vehicle frame points based on image characteristics of the parking space to determine a final vehicle frame point;

[0070] The second determining module 55 is configured to determine a parking space based on the multiple final vehicle frame points obtained through voting.

[0071] It can be understood that the parking space recognition system provided by the present invention corresponds to the parking space recognition method provided by the aforementioned embodiments. The relevant technical features of the parking space recognition system can refer to the relevant technical features of the parking space recognition method, which will not be repeated here.

[0072] See also Figure 6 , Figure 6 Schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. Figure 6 As shown, an embodiment of the present invention provides an electronic device 600, including a memory 610, a processor 620, and a computer program 611 stored in the memory 610 and executable on the processor 620. When the processor 620 executes the computer program 611, the steps of the parking space identification method of the above embodiment are implemented.

[0073] See also Figure 7 , Figure 7 Schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. Figure 7As shown, this embodiment provides a computer-readable storage medium 700 on which a computer program 711 is stored. When the computer program 711 is executed by a processor, the steps of the parking space recognition method of the above embodiment are implemented.

[0074] The embodiments of the present invention provide a parking space recognition method, recognition system, electronic device and storage medium, which intercept a portion of the area of ​​the parking space for straight line detection to perform corner point detection, correct the misidentification of the straight line detection, and identify the parking space by integrating the straight line detection and corner point detection, thereby improving the accuracy of parking space detection and recognition.

[0075] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0076] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0077] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0078] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0080] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0081] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A parking space recognition method, characterized in that: include: Acquire an image containing a parking space, and identify all straight line segments in the image based on edge extraction and line detection; splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining a plurality of first parking frame points of the parking space frame; Taking any first vehicle frame point as the center, screenshot a corresponding vehicle frame correction area in the image, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point for corner point detection; Voting on the first and second parking frame points based on the image characteristics of the parking space to obtain a final parking frame point; Determine the parking space based on the multiple final parking frame points obtained through voting; Voting on the first vehicle frame point and the second vehicle frame point to determine a final vehicle frame point includes: extracting eight neighboring pixel points of the first vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a change pattern of parking spaces; if so, voting for the any neighboring pixel point; otherwise, not voting; obtaining the number of votes of the eight neighboring pixel points corresponding to the first vehicle frame point; extracting eight neighboring pixel points of the second vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a change pattern of parking spaces; if so, voting for the any neighboring pixel point; otherwise, not voting; obtaining the number of votes of the eight neighboring pixel points corresponding to the first vehicle frame point; and selecting the first vehicle frame point or the second vehicle frame point with the largest number of votes as the final vehicle frame point.

2. The parking space recognition method according to claim 1, characterized in that: The method of obtaining an image containing a parking space includes: obtaining an image containing a parking space based on camera shooting, and converting the image from an original perspective to a bird's-eye view; the method of identifying all straight line segments in the image based on edge extraction and line detection includes: processing noise points around the parking space lines in the image from a bird's-eye view based on adaptive binarization and morphological processing, and detecting all straight line segments in the image based on edge extraction and line detection methods.

3. The parking space recognition method according to claim 1 or 2, characterized in that: The step of splicing the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and obtaining a plurality of first parking frame points of the parking space frame, includes grouping and filtering all straight line segments according to the length, width and angle of the standard parking space, and splicing the parking space, wherein the intersection of the long side and the short side of the parking space frame is the first parking frame point.

4. The parking space recognition method according to claim 1, characterized in that: Performing corner point detection on the vehicle frame correction area to obtain a second vehicle frame point detected by the corner point detection includes: graying the vehicle frame correction area and eliminating noise based on a Gaussian filter method; performing Harris corner point detection on the vehicle frame correction area from which the noise is eliminated to obtain a second vehicle frame point.

5. A parking space recognition system, characterized in that: include: A recognition module is used to obtain an image containing a parking space and identify all straight line segments in the image based on edge extraction and line detection; a first acquisition module, configured to splice the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and to acquire a plurality of first parking frame points of the parking space frame; a second acquisition module, configured to capture a corresponding vehicle frame correction area in the image with any first vehicle frame point as the center, perform corner point detection on the vehicle frame correction area, and obtain a second vehicle frame point detected by corner point detection; a first determining module, configured to vote on the first vehicle frame point and the second vehicle frame point according to image characteristics of the parking space to determine a final vehicle frame point; The second determination module is used to determine the parking space based on the multiple final parking frame points obtained by voting; Voting on the first vehicle frame point and the second vehicle frame point to determine a final vehicle frame point includes: extracting eight neighboring pixel points of the first vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a change pattern of parking spaces; if so, voting for the any neighboring pixel point; otherwise, not voting; obtaining the number of votes of the eight neighboring pixel points corresponding to the first vehicle frame point; extracting eight neighboring pixel points of the second vehicle frame point, and sequentially determining whether a pixel difference between any neighboring pixel point and the first vehicle frame point conforms to a change pattern of parking spaces; if so, voting for the any neighboring pixel point; otherwise, not voting; obtaining the number of votes of the eight neighboring pixel points corresponding to the first vehicle frame point; and selecting the first vehicle frame point or the second vehicle frame point with the largest number of votes as the final vehicle frame point.

6. The parking space recognition system according to claim 5, characterized in that: The first acquisition module is used to splice the straight line segments according to the length and angle of the standard parking space to form a parking space frame, and to obtain multiple first frame points of the parking space frame, including: grouping and filtering all straight line segments according to the length, width and angle of the standard parking space, and splicing the parking space, wherein the intersection of the long side and short side of the parking space frame is the first frame point.

7. An electronic device, characterized in that: The computer program comprises a memory and a processor, wherein the processor is configured to implement the steps of the parking space recognition method according to any one of claims 1 to 4 when executing a computer management program stored in the memory.

8. A computer-readable storage medium, characterized in that A computer management program is stored thereon, and when the computer management program is executed by the processor, the steps of the parking space recognition method according to any one of claims 1 to 4 are implemented.

Images