CN116486406A - Parking space marking method, parking method, device, electronic equipment and storage medium - Google Patents
Parking space marking method, parking method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116486406A CN116486406A CN202310446191.3A CN202310446191A CN116486406A CN 116486406 A CN116486406 A CN 116486406A CN 202310446191 A CN202310446191 A CN 202310446191A CN 116486406 A CN116486406 A CN 116486406A
- Authority
- CN
- China
- Prior art keywords
- parking space
- corner
- parking
- labeling
- missing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000002372 labelling Methods 0.000 claims abstract description 115
- 238000001514 detection method Methods 0.000 claims abstract description 29
- 230000001502 supplementing effect Effects 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000012797 qualification Methods 0.000 claims description 18
- 238000004891 communication Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 4
- 235000004522 Pentaglottis sempervirens Nutrition 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000013589 supplement Substances 0.000 description 3
- 229910003460 diamond Inorganic materials 0.000 description 2
- 239000010432 diamond Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/06—Automatic manoeuvring for parking
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
- G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/586—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of parking space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0043—Signal treatments, identification of variables or parameters, parameter estimation or state estimation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Probability & Statistics with Applications (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the application provides a parking space marking method, a parking device, electronic equipment and a storage medium, and a target image comprising a plurality of parking spaces is acquired; inputting a target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shapes, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data; and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, so as to obtain the labeling data of the parking space. According to the method, automatic labeling of the parking image can be achieved, manual labeling is not needed, labor cost is reduced, and labeling efficiency is high.
Description
Technical Field
The embodiment of the application relates to the technical field of automatic driving, in particular to a parking space marking method, a parking device, electronic equipment and a storage medium.
Background
With the development of automatic driving technology, the demand for automatic parking is also increasing. The automatic parking technology needs to detect the parking space of the parking area and then determine the available parking space according to the detection result, so that the intelligent vehicle can be controlled to move to the parking space to realize automatic parking.
For the accuracy of the parking space detection, a large amount of and comprehensive image data of different types of parking areas with parking space annotation data are required to perform training learning of a parking space detection model.
At present, an image of a parking area is marked by a manual marking method, and the manual marking method is high in labor cost and low in efficiency.
Disclosure of Invention
In view of the foregoing, embodiments of the present application provide a parking space labeling method, a parking method, a device, an electronic apparatus, and a storage medium, so as to at least partially solve the foregoing problems.
According to a first aspect of an embodiment of the present application, there is provided a parking space labeling method, including: acquiring a target image containing a plurality of parking spaces; inputting a target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shapes, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data; and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, so as to obtain the labeling data of the parking space.
Optionally, determining the parking space with missing corner points in the pre-labeled image according to the corner point data and the external frame data includes: according to the external frame data, determining corner coordinates belonging to the same parking space; and determining the parking spaces with missing corner points in the pre-marked image according to the number of the corner point coordinates belonging to the same parking space.
Optionally, supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space missing in the corner, including: determining the central position of the corner missing parking space according to the external frame data of the corner missing parking space; according to the obtained corner coordinates, the corresponding corner positions and the center positions in the corner missing parking spaces, determining the corner coordinates missing in the parking spaces, wherein at least two corner coordinates are obtained in the corner missing parking spaces.
Optionally, the method further comprises: judging whether the labeling data of the parking space is qualified or not according to preset labeling qualification judging conditions, wherein the labeling qualification judging conditions at least comprise labeling coverage rate and error rate; if the parking space marking data does not meet any of the marking qualification judging conditions, the parking space marking data is stored as parking space marking data to be corrected; and outputting parking space marking data if the parking space marking data completely meets the marking qualification judging condition.
Optionally, the availability information includes locking information and occupancy information.
According to a second aspect of embodiments of the present application, there is provided a parking method, including: acquiring a parking bit image of a target parking area; according to the method of the aspect, parking space information in a parking space image is determined, wherein the parking space information at least comprises parking space position information and corresponding availability information, and the availability information is used for indicating whether a parking space is available or not; and automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information.
According to a third aspect of the embodiment of the application, a parking space marking device is provided, which comprises an image acquisition unit, a pre-marking unit, a missing corner point determining unit and a missing corner point supplementing unit, wherein the image acquisition unit is used for acquiring target images containing a plurality of parking spaces; the pre-labeling unit is used for inputting the target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shape, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training the target detection model through a parking space sample image set; the missing corner determining unit is used for determining parking spaces with missing corners in the pre-marked image according to the corner coordinates and the external frame data; the missing corner supplementing unit is used for supplementing the missing corner coordinates in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, and obtaining the labeling data of the parking space.
According to a fourth aspect of an embodiment of the present application, there is provided a parking device including: the system comprises an image acquisition unit, a parking space detection unit and an automatic parking unit, wherein the image acquisition unit is used for acquiring a parking image of a target parking area; the parking space detection unit is used for determining parking space information in the parking space image according to the method of the first aspect, wherein the parking space information comprises parking space position information and corresponding availability information, and the availability information is used for indicating whether the parking space is available or not; the automatic parking unit is used for automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information.
According to a fifth aspect of embodiments of the present application, there is provided an electronic device, including: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform operations corresponding to the methods of the embodiments of the present application.
According to a sixth aspect of embodiments of the present application, there is provided a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the method of embodiments of the present application.
According to the parking space marking method, the parking method, the device, the electronic equipment and the storage medium, which are provided by the embodiment of the application, target images containing a plurality of parking spaces are obtained; inputting a target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shapes, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data; and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, so as to obtain the labeling data of the parking space. According to the method, automatic labeling of the parking image can be achieved, manual labeling is not needed, labor cost is reduced, and labeling efficiency is high. The method can supplement the missing parking space angle points, the problem that the detection of the parking space angle points is incomplete due to the fact that parking spaces are blocked and the like is avoided, the parking space annotation data obtained by the method also comprise availability information, parking space shape and other information, the annotation data is rich and comprehensive, automatic parking is assisted by using the annotation data, and the accuracy of automatic parking can be improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly introduce the drawings that are required to be used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present application, and other drawings may also be obtained according to these drawings for a person having ordinary skill in the art.
Fig. 1 is a flowchart of a parking space labeling method according to an exemplary embodiment of the present application;
FIG. 2 is a flow chart of a parking method according to an exemplary embodiment of the present application;
FIG. 3 is a block diagram of a parking space marking device according to an exemplary embodiment of the present application;
fig. 4 is a block diagram of a parking device according to an exemplary embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present application;
fig. 6 is an effect schematic diagram of a parking space labeling method according to an exemplary embodiment of the present application.
Detailed Description
In order to better understand the technical solutions in the embodiments of the present application, the following descriptions will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, but not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the embodiments of the present application shall fall within the scope of protection of the embodiments of the present application.
Embodiments of the present application are further described below with reference to the accompanying drawings of embodiments of the present application.
Fig. 1 is a schematic flow chart of a parking space labeling method according to an exemplary embodiment of the present application, as shown in the drawing, the embodiment mainly includes the following steps:
s101, acquiring a target image containing a plurality of parking spaces.
Illustratively, the method of the present embodiment may be performed at the cloud. The target image containing a plurality of parking spaces can also be a parking space image which is acquired from a cloud database and needs to be marked, and the parking space image can be a bird's eye view of a parking area. The bird's eye view is a perspective view drawn by looking down the ground surface from a point at a high place by high-view perspective according to the perspective principle. In short, a bird's eye view is an image that is seen in a region in plan view in the air. Because the parking space is usually a plurality of parallelogram areas which are divided on the ground by lines and can be used for parking vehicles, the data of the parking space in one parking area can be comprehensively acquired through the aerial view image.
It should be noted that, the standard size of the parking space can be divided into two sizes, namely a large size and a small size, and the large parking space is usually 15.6 meters long and 3.25 meters wide, and is suitable for parking medium and large vehicles; the small parking space is 6 m long and 2.5 m wide, and is suitable for parking small vehicles. The arrangement modes of parking spaces can be divided into three modes: parallel type, inclined type (generally, the inclined angle is divided into 30 degrees, 45 degrees and 60 degrees) and vertical type, and the large parking space is not generally arranged in an inclined type and vertical type. The shape of the parking spaces generally varies according to the arrangement, for example, the shape of the parking spaces arranged in parallel and in vertical is generally rectangular, and the shape of the parking spaces arranged in inclined is generally diamond. The parking space is usually divided by using lines, and the parking space usually comprises 4 corner points, wherein two corner points are entrance corner points, and the lines at the entrance corner points are discontinuous lines, so that the corner point positions of all the corner points in the parking space can be distinguished, whether the corner points are the entrance corner points or not, and the like.
S102, inputting the target image into a preset pre-labeling model to obtain parking space pre-labeling data.
The parking space pre-marking data at least comprise parking space shape, availability information, corner coordinates, corner positions and external frame data of the parking space. The pre-labeling model is obtained by training the target detection model through a parking space sample image set.
Illustratively, the parking space shape may include a diamond shape and a rectangle, the availability information may include that the parking space is available or that the parking space is unavailable, the corner point position refers to a position where the corner point is located in the parking space, for example, the corner point is a corner point at an entrance of the parking space, further, the position of the corner point in the parking space may be marked by a serial number, for example, a position right in front of the entrance of the parking space is marked as a number 1, and then, in a counterclockwise direction, the corner point positions are sequentially ordered as follows: and the positions of two angular points at the entrance of the parking space are the positions 1 and 2, so that the arrangement condition, the entrance direction and the like of the parking space can be determined according to the angular point positions. The external frame data is used for determining corner coordinates, a parking space shape and availability information of parking spaces belonging to the same parking space, specifically, the external frame can refer to a smallest rectangular frame surrounding 4 corner points of one parking space, the external frame data can be composed of center coordinates and size data, and the parking space shape and availability information can be transmitted through classification tag information of the parking spaces in a parking space sample image set.
Specifically, the pre-labeling model is obtained by training a target detection model through a parking space sample image set. The parking space sample image set can be obtained by preprocessing the acquired parking space image, for example, a target detection frame with fixed size can be clustered at the corner points of the parking space image by using a k-means method and the like, and the target detection frame represents the length-width scale of the main distribution of the corner points in the parking space image. Meanwhile, classification label information is given to whether the corner points are visible or not and whether the corner points are entrance corners or not, in addition, the parking space sample image in the parking space sample image set can also comprise other classification label information of parking spaces, such as external frame data of the parking spaces, space shape and availability information, and the like, the external frame data can indicate the overall size of the parking spaces and determine attribution information of corresponding parking spaces of different corner points, the space shape can be determined according to an arrangement mode, and comprises parallel, vertical, inclined and other availability information, and the availability information can comprise whether the information is occupied or not and whether the information is locked or not, so that the inference result of a pre-labeled model obtained through training can be more accurate. The object detection model may be any object detection model commonly used in the art, such as the YOLOv7 object detection model, which is not limited in this embodiment,
in one particular implementation, the availability information may include whether to lock information and whether to occupy information. By adding the availability information to the parking space marking data, in the parking space detection of the automatic parking by using the target detection model at the vehicle end obtained by training the parking space marking data in the embodiment, the use state of the parking space can be obtained so as to assist in selecting the parking space, and safety accidents are caused by collision between the vehicle and the parked vehicle, the parking space locking device or other occupied objects on the parking space due to lack of the availability information when the vehicle is automatically parked.
And S103, determining parking spaces with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data.
Illustratively, the parking space shapes, availability information, corner coordinates, corner positions and circumscribed frame data of a plurality of parking spaces in the target image are output through a pre-labeling model. Because each parking space comprises a plurality of corner coordinates and because the parking spaces are connected with each other, a plurality of common corner points exist, the corner coordinates belonging to the same parking space can be determined according to external frame data, and therefore whether the corner coordinates of each parking space are complete or not can be determined.
In a specific implementation manner, determining a parking space with missing corner points in a pre-labeled image according to corner point data and external frame data comprises: according to the external frame data, determining corner coordinates belonging to the same parking space; and determining the parking spaces with missing corner points in the pre-marked image according to the number of the corner point coordinates belonging to the same parking space.
The corner coordinates belonging to the same parking space are determined according to the external frame data, the number of the corner coordinates belonging to the same parking space is determined, and if the number of the corner coordinates of the parking space is smaller than 4, the parking space belonging to the corner missing parking space is determined. For example, if the number of the corner coordinates of a part of parking spaces is 4, the corner coordinates of the parking space are complete, if some parking spaces are blocked by a building or other reasons, the number of the corner coordinates of the parking space detected by the pre-labeling model is only 2 or 3, and if the parking space belongs to a parking space with a missing corner.
Because the target image has the problems of shooting angles, positions of parking spaces, other shielding objects such as buildings and the like in the shooting process, angular point coordinates which cannot be detected by the pre-labeling model exist in the target image, so that the angular point coordinates of part of parking spaces are incomplete, and labeling data of the parking spaces are omitted.
S104, supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, and obtaining the labeling data of the parking space.
For example, according to the number of corner coordinates belonging to the same parking space, a parking space with a missing corner may be determined, and according to the known corner coordinates of the parking space with the missing corner and the corner position corresponding to the corner coordinates, that is, the position of the parking space where the corner coordinates are located, for example, the corner coordinates may be located at any one of the number 1, the number 2, the number 3 or the number 4 positions in the parking space where the corner coordinates are located, where the number 1 position in the parking space may refer to the position right in front of the entrance of the parking space, and the number 2, the number 3 and the number 4 positions are obtained by performing counterclockwise sequencing based on the number 1 position. And supplementing the corner coordinates missing in the parking space according to the known corner coordinates and the corner positions corresponding to the corner coordinates to obtain the labeling data of the parking space. Referring to fig. 6, the left side of the drawing is an effect diagram of the pre-labeling result, the right side is an effect diagram of the labeling result, and the parking space labeling data includes parking space pre-labeling data and corner coordinates which are complemented, such as corner coordinates of a parking space, a parking space shape, corner positions, availability information and the like.
In a specific implementation manner, supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with missing corner, including: determining the central position of the corner missing parking space according to the external frame data of the corner missing parking space; according to the obtained corner coordinates, the corresponding corner positions and the center positions in the corner missing parking spaces, determining the corner coordinates missing in the parking spaces, wherein at least two corner coordinates are obtained in the corner missing parking spaces.
The central position coordinate of the external frame is determined as the central position coordinate of the parking space with the missing corner according to the external frame data of the parking space with the missing corner. Furthermore, according to the known angular point coordinates of the parking space with the missing angular point and the angular point positions corresponding to the angular point coordinates, the known angular point coordinates are at least two, and the parking space is generally in a parallelogram shape and has symmetry, so that the axisymmetric or centrosymmetric transformation can be performed according to the angular point coordinates, the angular point positions corresponding to the angular point coordinates and the central position coordinates of the parking space, and the missing angular point coordinates can be obtained.
In the implementation mode, the missing corner points of the parking space are complemented in an axisymmetric or centrosymmetric mode, so that the obtained parking space marking data are more complete, and the marking effect is better.
In one specific implementation, the method further comprises: judging whether the labeling data of the parking space is qualified or not according to preset labeling qualification judging conditions, wherein the labeling qualification judging conditions at least comprise labeling coverage rate and error rate; if the parking space marking data does not meet any of the marking qualification judging conditions, the parking space marking data is stored as parking space marking data to be corrected; and outputting parking space marking data if the parking space marking data completely meets the marking qualification judging condition.
The marking coverage rate threshold value and the error rate threshold value can be set according to the requirements of the field on the marking coverage rate and the error rate of the marking data of the parking space, and whether the marking coverage rate and the error rate of the marking data of the parking space meet the marking coverage rate threshold value and the error rate threshold value or not is determined, namely, the marking coverage rate threshold value and the error rate threshold value are larger than or equal to each other, if yes, the marking data of the parking space is qualified, and the marking coverage rate and the error rate threshold value are output and stored; if one of the marking coverage rate threshold and the error rate threshold is not met, the parking space marking data are stored as parking space marking data to be corrected, so that revision or supplementary marking can be carried out later. The labeling qualification judging condition can further comprise a pixel deviation threshold value, wherein the pixel deviation refers to a distance error between a corner coordinate in parking space labeling data and a real corner coordinate in a target image of a parking space, if the distance error is smaller than or equal to the pixel deviation threshold value, the parking space labeling data is qualified, and if the distance error is larger than the pixel deviation threshold value, the parking space labeling data is stored as parking space labeling data to be corrected so as to be revised later.
In the implementation mode, the labeling result of the parking space is checked by setting the labeling qualification judging condition, and unqualified labeling data of the parking space is stored as labeling data of the parking space to be corrected, so that modification and supplementary labeling are carried out at the following time, the accuracy and the data integrity of the output labeling result are ensured, and the quality of the labeling data is improved.
According to the parking space labeling method provided by the embodiment of the application, a target image containing a plurality of parking spaces is acquired; inputting a target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shapes, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data; and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, so as to obtain the labeling data of the parking space. According to the method, automatic labeling of the parking image can be achieved, manual labeling is not needed, labor cost is reduced, and labeling efficiency is high. In addition, the missing parking space angle points can be supplemented, the problem that the detection of the parking space angle points is incomplete due to the fact that parking spaces are blocked and the like is avoided, and the parking space annotation data obtained by the method further comprise availability information, parking space shape and other information, so that the annotation data are rich and comprehensive, and the accuracy of automatic parking can be improved when the automatic parking is assisted by using the parking space annotation data.
Fig. 2 is a flow chart of a parking method according to an exemplary embodiment of the present application. The method can be executed at the vehicle end, and as shown in the figure, the embodiment mainly comprises the following steps:
s201, acquiring a parking image of a target parking area.
S202, according to the parking space labeling method, parking space information in a parking space image is determined, wherein the parking space information at least comprises parking space position information and corresponding availability information, and the availability information is used for indicating whether a parking space is available or not.
S203, automatic parking is carried out according to the position information of the vehicle, the parking position information and the corresponding availability information.
The parking space marking method comprises the steps that a parking space marking method is used for marking a parking space, a parking end obtains a parking space image of a target parking area, the parking space image is processed according to the parking space marking method, parking space information in the parking space image is determined, the parking space information at least comprises parking space position information and corresponding availability information, the parking space position information comprises corner coordinates and corner positions of a parking space, and the availability information is used for indicating whether the parking space is available or not. And determining a navigation route according to the parking space position information and the position information of the vehicle so as to automatically park.
According to the parking method, a parking bit image of a target parking area is acquired; according to the parking space labeling method, parking space information in a parking space image is determined, wherein the parking space information at least comprises parking space position information and corresponding availability information, and the availability information is used for indicating whether a parking space is available or not; and automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information. According to the embodiment, the parking space data obtained by adopting the parking space marking method comprises the information such as the corner coordinates, the corner positions, the availability information and the parking space shape, so that the obtained parking space data is more comprehensive, the accuracy of automatic parking can be improved when the automatic parking is assisted by using the parking space data, and the occurrence of collision of obstacles on a parking space during automatic parking can be avoided through the availability information, so that the safety problem is solved.
Fig. 3 is a block diagram illustrating a structure of a parking space marking apparatus according to an exemplary embodiment of the present application.
As shown in the drawing, the parking space marking apparatus 300 of the present embodiment mainly includes: an image acquisition unit 301, a pre-labeling unit 302, a missing corner determination unit 303, and a missing corner supplementation unit 304.
Wherein, the image acquisition unit 301 is used for acquiring a target image containing a plurality of parking spaces; the pre-labeling unit 302 is configured to input a target image into a preset pre-labeling model to obtain parking space pre-labeling data, where the parking space pre-labeling data at least includes a parking space shape, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; the missing corner determining unit 303 is configured to determine a parking space with missing corners in the pre-labeled image according to the corner coordinates and the circumscribed frame data; the missing corner supplementing unit 304 is configured to supplement the corner coordinates of the missing parking space according to the corner coordinates and the corner positions of the parking space where the corner is missing, and obtain labeling data of the parking space.
According to the parking space marking device, the target images comprising a plurality of parking spaces are obtained; inputting a target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shapes, availability information, corner coordinates, corner positions and external frame data of a parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set; determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data; and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space with the missing corner, so as to obtain the labeling data of the parking space. According to the method, automatic labeling of the parking image can be achieved, manual labeling is not needed, labor cost is reduced, and labeling efficiency is high. The method can supplement the missing parking space angle points, the problem that the detection of the parking space angle points is incomplete due to the fact that parking spaces are blocked and the like is avoided, the parking space annotation data obtained by the method also comprise availability information, parking space shape and other information, the annotation data is rich and comprehensive, automatic parking is assisted by using the annotation data, and the accuracy of automatic parking can be improved.
In a specific implementation manner, the missing corner determining unit 303 is further configured to determine, according to the circumscribed frame data, corner coordinates belonging to the same parking space; and determining the parking spaces with missing corner points in the pre-marked image according to the number of the corner point coordinates belonging to the same parking space.
In a specific implementation manner, the missing corner point supplementing unit 304 is further configured to determine a center position of the corner missing parking space according to the circumscribed frame data of the corner missing parking space; according to the obtained corner coordinates, the corresponding corner positions and the center positions in the corner missing parking spaces, determining the corner coordinates missing in the parking spaces, wherein at least two corner coordinates are obtained in the corner missing parking spaces.
In a specific implementation manner, the device further comprises a result checking unit 305, and the result checking unit 305 is used for judging whether the labeling data of the parking space is qualified according to preset labeling qualification judging conditions, wherein the labeling qualification judging conditions at least comprise labeling coverage rate and error rate; if the parking space marking data does not meet any of the marking qualification judging conditions, the parking space marking data is stored as parking space marking data to be corrected; and outputting parking space marking data if the parking space marking data completely meets the marking qualification judging condition.
In one particular implementation, the availability information includes whether to lock information and whether to occupy information.
In addition, the parking space marking device 300 of the embodiment of the present application may be further used to implement other steps in the foregoing embodiments of the method for marking a parking space, and have the beneficial effects of the corresponding embodiments of the method steps, which are not described herein again.
Fig. 4 is a block diagram of a parking device according to an exemplary embodiment of the present application.
As shown in the drawing, the parking device 400 of the present embodiment mainly includes: an image acquisition unit 401, a parking space detection unit 402, and an automatic parking unit 403.
The image acquisition unit 401 is configured to acquire a parking bit image of a target parking area; the parking space detection unit 402 is configured to determine parking space information in the parking space image according to the method of the first aspect, where the parking space information includes parking space position information and corresponding availability information, and the availability information is used to indicate whether a parking space is available; the automatic parking unit 403 is configured to automatically park according to position information of a vehicle, parking space position information, and corresponding availability information.
According to the parking device, a parking bit image of a target parking area is acquired; according to the parking space labeling method, parking space information in a parking space image is determined, wherein the parking space information at least comprises parking space position information and corresponding availability information, and the availability information is used for indicating whether a parking space is available or not; and automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information. According to the embodiment, the parking space data obtained by adopting the parking space marking method comprises the information such as the corner coordinates, the corner positions, the availability information and the parking space shape, so that the obtained parking space data is more comprehensive, the accuracy of automatic parking can be improved when the automatic parking is assisted by using the parking space data, and the occurrence of collision of obstacles on a parking space during automatic parking can be avoided through the availability information, so that the safety problem is solved.
Referring to fig. 5, a schematic structural diagram of an electronic device 500 according to another embodiment of the present invention is shown, and the specific embodiment of the present invention is not limited to the specific implementation of the electronic device.
As shown in fig. 5, the electronic device 500 may include: a processor (processor) 501, a memory (memory) 503, a communication bus 504, and a communication interface (Communications Interface) 505.
Wherein:
processor 501, memory 503, and communication interface 505 perform communication with each other via communication bus 504.
A communication interface 505 for communicating with other electronic devices or servers.
The processor 501 is configured to execute the program 502, and may specifically perform the steps of the method in any of the foregoing embodiments.
In particular, program 502 may include program code including computer operating instructions.
The processor 501 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present application. The one or more processors comprised by the smart device may be the same type of processor, such as one or more CPUs; but may also be different types of processors such as one or more CPUs and one or more ASICs.
A memory 503 for storing the program 502. The memory 503 may comprise high-speed RAM memory or may further comprise non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 502 is specifically operable to cause the processor 501 to execute to implement the steps of any one of the methods described in the embodiments. The specific implementation of each step in the program 502 may refer to the steps and corresponding descriptions in the units executed by the method in any of the above steps, which are not described herein. It will be apparent to those skilled in the art that for convenience and brevity of description, the specific operation of the apparatus and modules described above may be described with reference to corresponding processes in the foregoing method embodiments.
The exemplary embodiments of the present application also provide a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the methods of the embodiments of the present application.
The above-described methods according to embodiments of the present invention may be implemented in hardware, firmware, or as software or computer code storable in a recording medium such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk, or as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium and to be stored in a local recording medium downloaded through a network, so that the methods described herein may be stored on such software processes on a recording medium using a general purpose computer, special purpose processor, or programmable or special purpose hardware such as an ASIC or FPGA. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code that, when accessed and executed by a computer, processor, or hardware, performs the methods described herein. Furthermore, when a general purpose computer accesses code for implementing the methods illustrated herein, execution of the code converts the general purpose computer into a special purpose computer for performing the methods illustrated herein.
Thus, specific embodiments of the present invention have been described. Other embodiments are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may be advantageous.
It should be noted that all directional indicators (such as up, down, left, right, and rear … …) in the embodiments of the present invention are merely used to explain the relative positional relationship, movement conditions, etc. between the components in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indicator is correspondingly changed.
In the description of the present invention, the terms "first," "second," and the like are used merely for convenience in describing the various components or names, and are not to be construed as indicating or implying a sequential relationship, relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
It should be noted that, although specific embodiments of the present invention have been described in detail with reference to the accompanying drawings, the present invention should not be construed as limiting the scope of the present invention. Various modifications and variations which may be made by those skilled in the art without the creative effort fall within the protection scope of the present invention within the scope described in the claims.
Examples of embodiments of the present invention are intended to briefly illustrate technical features of embodiments of the present invention so that those skilled in the art may intuitively understand the technical features of the embodiments of the present invention, and are not meant to be undue limitations of the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The parking space marking method is characterized by comprising the following steps of:
acquiring a target image containing a plurality of parking spaces;
inputting the target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shape, availability information, corner coordinates, corner positions and external frame data of the parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set;
determining parking positions with missing corner points in the pre-marked image according to the corner point coordinates and the external frame data;
and supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space missing in the corner, so as to obtain the labeling data of the parking space.
2. The method of claim 1, wherein determining the parking space for which the corner point in the pre-labeled image is missing based on the corner point data and the circumscribed frame data comprises:
determining the corner coordinates belonging to the same parking space according to the external frame data;
and determining the parking spaces with missing corner points in the pre-marked image according to the number of the corner point coordinates belonging to the same parking space.
3. The method according to claim 1 or 2, wherein the supplementing of the missing corner coordinates in the parking space according to the corner coordinates and the corner positions of the missing parking space comprises:
determining the central position of the corner-missing parking space according to the external frame data of the corner-missing parking space;
and determining the corner coordinates missing in the parking space according to the obtained corner coordinates, the corresponding corner positions and the center positions in the parking space missing in the corner, wherein at least two corner coordinates are obtained in the parking space missing in the corner.
4. The method according to claim 1, wherein the method further comprises:
judging whether the labeling data of the parking space is qualified or not according to preset labeling qualification judging conditions, wherein the labeling qualification judging conditions at least comprise labeling coverage rate and error rate;
if the parking space marking data does not meet any one of the marking qualification judging conditions, the marking data is stored as parking space marking data to be corrected;
and if the labeling data of the parking space completely meets the labeling qualification judging condition, outputting the labeling data of the parking space.
5. The method of claim 1, wherein the availability information includes locking information and occupancy information.
6. A method of parking, comprising:
acquiring a parking bit image of a target parking area;
the method of any of claims 1-5, determining parking space information in the parking space image, the parking space information including at least parking space position information and corresponding availability information, the availability information being indicative of whether the parking space is available;
and automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information.
7. A parking space marking device, characterized by comprising:
the image acquisition unit is used for acquiring target images containing a plurality of parking spaces;
the pre-labeling unit is used for inputting the target image into a preset pre-labeling model to obtain parking space pre-labeling data, wherein the parking space pre-labeling data at least comprise parking space shape, availability information, corner coordinates, corner positions and external frame data of the parking space, and the pre-labeling model is obtained by training a target detection model through a parking space sample image set;
the missing corner determining unit is used for determining a parking space with missing corner in the pre-marked image according to the corner coordinates and the external frame data;
the missing corner supplementing unit is used for supplementing the corner coordinates missing in the parking space according to the corner coordinates and the corner positions of the parking space missing from the corner, so as to obtain the labeling data of the parking space.
8. A parking apparatus, characterized by comprising:
an image acquisition unit for acquiring a parking bit image of a target parking area;
a parking space detection unit for determining parking space information in the parking space image according to the method of any one of claims 1-5, the parking space information comprising at least parking space position information and corresponding availability information for indicating whether the parking space is available;
and the automatic parking unit is used for automatically parking according to the position information of the vehicle, the parking position information and the corresponding availability information.
9. An electronic device, comprising: the device comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete communication with each other through the communication bus; the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method of any one of claims 1-6.
10. A computer storage medium having stored thereon a computer program, which when executed by a processor performs the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446191.3A CN116486406A (en) | 2023-04-24 | 2023-04-24 | Parking space marking method, parking method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446191.3A CN116486406A (en) | 2023-04-24 | 2023-04-24 | Parking space marking method, parking method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116486406A true CN116486406A (en) | 2023-07-25 |
Family
ID=87211422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310446191.3A Pending CN116486406A (en) | 2023-04-24 | 2023-04-24 | Parking space marking method, parking method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116486406A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117078799A (en) * | 2023-07-31 | 2023-11-17 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
| CN118334627A (en) * | 2024-06-07 | 2024-07-12 | 比亚迪股份有限公司 | Parking space detection method, device, medium, computer program product and vehicle |
| CN118570771A (en) * | 2024-08-01 | 2024-08-30 | 比亚迪股份有限公司 | Parking space marking method, electronic device, vehicle and computer-readable storage medium |
| CN120071274A (en) * | 2025-04-29 | 2025-05-30 | 浙江大华技术股份有限公司 | Parking space identification method, electronic equipment and computer readable storage medium |
-
2023
- 2023-04-24 CN CN202310446191.3A patent/CN116486406A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117078799A (en) * | 2023-07-31 | 2023-11-17 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
| CN117078799B (en) * | 2023-07-31 | 2024-05-03 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
| CN118334627A (en) * | 2024-06-07 | 2024-07-12 | 比亚迪股份有限公司 | Parking space detection method, device, medium, computer program product and vehicle |
| CN118570771A (en) * | 2024-08-01 | 2024-08-30 | 比亚迪股份有限公司 | Parking space marking method, electronic device, vehicle and computer-readable storage medium |
| CN120071274A (en) * | 2025-04-29 | 2025-05-30 | 浙江大华技术股份有限公司 | Parking space identification method, electronic equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116486406A (en) | Parking space marking method, parking method, device, electronic equipment and storage medium | |
| EP3751519B1 (en) | Method, apparatus, device and medium for calibrating pose relationship between vehicle sensor and vehicle | |
| US11506769B2 (en) | Method and device for detecting precision of internal parameter of laser radar | |
| US11226200B2 (en) | Method and apparatus for measuring distance using vehicle-mounted camera, storage medium, and electronic device | |
| US11320833B2 (en) | Data processing method, apparatus and terminal | |
| CN110163930B (en) | Lane line generation method, device, equipment, system and readable storage medium | |
| CN109074668B (en) | Path navigation method, related device and computer readable storage medium | |
| EP3633539A2 (en) | Method for position detection, device, and storage medium | |
| CN109949439B (en) | Driving live-action information labeling method and device, electronic equipment and medium | |
| CN110132291B (en) | Grid map generation method, system, equipment and storage medium for port | |
| CN113657224A (en) | Method, device and equipment for determining object state in vehicle-road cooperation | |
| CN109918977A (en) | Method, device and device for determining vacant parking spaces | |
| CN112782721A (en) | Passable area detection method and device, electronic equipment and storage medium | |
| CN114140608B (en) | Photovoltaic panel marking method and device, electronic equipment and storage medium | |
| CN114047487B (en) | Radar and vehicle body external parameter calibration method and device, electronic equipment and storage medium | |
| CN113945937A (en) | Precision detection method, device and storage medium | |
| CN116386373A (en) | Vehicle positioning method and device, storage medium and electronic equipment | |
| EP4206610A1 (en) | Map matching method and apparatus, and electronic device and storage medium | |
| CN115050005B (en) | Target detection method and detection device for high-level video intelligent parking scene | |
| CN114255274B (en) | Vehicle positioning method, system, device and storage medium based on QR code recognition | |
| CN113581184B (en) | Method, device, equipment and medium for determining maximum passable area | |
| CN114092909B (en) | Lane line extraction method, device, vehicle and storage medium | |
| CN115457506A (en) | Target detection method, device and storage medium | |
| CN116412804A (en) | Laser radar vision fusion map building method, device and medium for traffic sign board | |
| CN117765067A (en) | Vehicle motion index measurement method, device, equipment and automatic driving vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |