WO2020048152A1 - 高精度地图制作中地下车库停车位提取方法及系统 - Google Patents

高精度地图制作中地下车库停车位提取方法及系统 Download PDF

Info

Publication number
WO2020048152A1
WO2020048152A1 PCT/CN2019/086895 CN2019086895W WO2020048152A1 WO 2020048152 A1 WO2020048152 A1 WO 2020048152A1 CN 2019086895 W CN2019086895 W CN 2019086895W WO 2020048152 A1 WO2020048152 A1 WO 2020048152A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
parking
parking space
contrast
eye view
Prior art date
Application number
PCT/CN2019/086895
Other languages
English (en)
French (fr)
Inventor
张程
李叶伟
罗跃军
Original Assignee
武汉中海庭数据技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 武汉中海庭数据技术有限公司 filed Critical 武汉中海庭数据技术有限公司
Priority to US16/618,440 priority Critical patent/US20200152060A1/en
Publication of WO2020048152A1 publication Critical patent/WO2020048152A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
    • G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
    • G09B29/003—Maps
    • G09B29/005—Map projections or methods associated specifically therewith
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804—Creation or updating of map data
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/36—Input/output arrangements for on-board computers
    • G01C21/3679—Retrieval, searching and output of POI information, e.g. hotels, restaurants, shops, filling stations, parking facilities
    • G01C21/3685—Retrieval, searching and output of POI information, e.g. hotels, restaurants, shops, filling stations, parking facilities the POI's being parking facilities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • 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
    • G06V20/60—Type of objects
    • G06V20/64—Three-dimensional objects
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
    • G08G1/141—Traffic control systems for road vehicles indicating individual free spaces in parking areas with means giving the indication of available parking spaces
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
    • G08G1/145—Traffic control systems for road vehicles indicating individual free spaces in parking areas where the indication depends on the parking areas
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
    • G09B29/00—Maps; Plans; Charts; Diagrams, e.g. route diagram
    • G09B29/003—Maps
    • G09B29/006—Representation of non-cartographic information on maps, e.g. population distribution, wind direction, radiation levels, air and sea routes
    • G09B29/007—Representation of non-cartographic information on maps, e.g. population distribution, wind direction, radiation levels, air and sea routes using computer methods
    • 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
    • G06V2201/12—Acquisition of 3D measurements of objects

Definitions

  • the present application relates to the technical field of high-precision map production, and in particular, to a method and system for extracting parking spaces in an underground garage during high-precision map production.
  • High-precision maps are one of the core technologies of unmanned driving. Accurate maps are crucial to the positioning, navigation and control of unmanned vehicles, as well as safety. How to generate high-precision maps is also an urgent problem in the field of unmanned driving.
  • An underground garage parking space means an area that is built underground and can be used for long-term or long-term or temporary parking of motor vehicles.
  • the parking line divides the parking area for each vehicle and each vehicle according to a certain size.
  • the underground parking lot cooperates with different grades of urban roads to meet the parking needs of different scales, and plays a very important role in regulating and controlling the traffic in the central area of the city.
  • High-precision underground parking space data is particularly important as an important part of high-precision maps.
  • the existing parking space extraction method is often based on the original image data extraction method.
  • the edge detection method is used to perform edge detection to obtain the edge point set of the parking line. Then, the edge point set is subjected to Huff transform and straight line extraction. The extraction gets the final parking space.
  • the embodiments of the present application are expected to propose a method and system for extracting parking spaces in an underground garage during high-precision map production.
  • An embodiment of the present application provides a method for extracting parking spaces in an underground garage during high-precision map production.
  • the method includes:
  • the center point of the value image is the circle center, and the binary image is rotated to obtain a rotated image; the number of pixels in each row and column in the rotated image that includes the parking line is counted, and the integral projection of the horizontal and vertical methods is obtained respectively;
  • the integral projection search in the horizontal and vertical directions obtains the coordinates of the four inner corner points corresponding to the parking space; the parking space is extracted by inverse transformation to the point cloud data according to the four inner corner point coordinates.
  • the determination of a contrast estimation index of the two-dimensional bird's-eye view mode image, and the two-dimensional bird's-eye view image according to the contrast estimation index Perform image preprocessing to obtain binary images, including:
  • I represents a two-dimensional bird's-eye view image
  • e represents contrast
  • the two-dimensional bird's-eye view image is sequentially subjected to morphological closing processing, local Laplacian filtering processing, and Gaussian adaptive binarization processing. Get a binary image.
  • the detecting a straight line segment of the binary image and determining a rotation angle of the parking line of the parking space according to the detection result includes:
  • the probabilistic Hough transform detects a set of straight line segments with the directionality of the parking line in the binary image, traversing the set of straight line segments to obtain a straight line segment length greater than a first threshold, and an angle between the straight line segment and a specific direction meets a preset condition straight Segment subset; calculating a straight segment length and an inclination angle in the straight segment subset, and determining a parking line rotation angle of the parking space according to the straight segment length and the inclination angle.
  • the rotating the binary image according to the rotation angle of the parking line and using the center point of the binary image as a circle center includes: The rotation angle of the parking line is used as the rotation angle, and the binary image is rotated with the center point of the binary image as a circle center, and the parking image in the obtained rotated image is parallel or perpendicular to the horizontal direction.
  • the integral projection includes: separately determining the number of pixels in each column and row of the rotated image that includes a parking line, and obtaining one-dimensional vectors representing horizontal and vertical integral projections, respectively.
  • the coordinates of four inner corners of the parking space are obtained according to the horizontal and vertical integral projection search, including: expressing horizontal and At the center index of the vertical integral projection vector, search for the first element whose gray value corresponding to the index is greater than the second threshold in the positive and negative directions, and obtain the elements v v [i], v v [j], and v h [m] And element v h [n]; where i, j, m, and n respectively represent element indexes; and based on the element indexes i, j, m, and n, coordinates of four intersections corresponding to parking spaces in the rotated image are obtained.
  • the step of extracting a parking space through inverse transformation to point cloud data according to the coordinates of the four internal corner points includes: Bit rotation angle, the coordinates of the four internal corner points are reversely rotated with the center point of the rotated image as the center of the circle, and projected into the input point cloud after inverse transformation to extract parking spaces.
  • An embodiment of the present application further provides a parking space extraction system for an underground garage in high-precision map production, and the system includes:
  • a projection unit configured to obtain three-dimensional laser point cloud data including a parking space, and project the three-dimensional laser point cloud data into a two-dimensional bird's-eye view image;
  • a contrast estimation unit configured to determine a ratio estimation index of the two-dimensional bird's-eye view mode image, and performing image preprocessing on the two-dimensional bird's-eye view mode image according to the contrast estimation index to obtain a binary image
  • An angle estimation unit configured to detect a straight line segment of the binary image, and determine a parking line rotation angle of the parking space according to the detection result
  • a rotation unit configured to rotate the binary image according to the rotation angle of the parking line and use the center point of the binary image as a circle center to obtain a rotated image
  • a statistics unit configured to count the number of pixels in each row and each column in the rotated image that includes a parking line, and obtain integral projections of horizontal and vertical methods, respectively;
  • a coordinate search unit configured to obtain coordinates of four inner corner points corresponding to a parking space according to the horizontal and vertical integral projection search
  • the parking space extraction unit is configured to extract a parking space through inverse transformation to point cloud data according to the four internal corner point coordinates.
  • the angle estimation unit is configured to detect, by a probabilistic Hough transform, a set of straight line segments in which the binary image has parking line directionality. Traversing the set of straight segments to obtain a subset of straight segments whose length is greater than a first threshold and the angle between the straight segment and a specific direction satisfies a preset condition; calculating the length and tilt angle of the straight segments in the subset of straight segments, according to the The length of the straight line segment and the inclination angle determine the parking line rotation angle of the parking space.
  • the rotation unit is configured to use the rotation angle of the parking line as a rotation angle, and use the center point of the binary image as a circle center.
  • the binary image is rotated, and the parking line in the obtained rotated image is parallel or vertical to the horizontal direction.
  • the statistical unit is configured to determine the number of pixels in each column and each row in the rotated image that includes parking lines, One-dimensional vectors representing the horizontal and vertical integral projections are obtained.
  • the coordinate search unit is configured to search for the gray corresponding to the index in the positive and negative directions by the horizontal and vertical integral projection vector center indexes respectively.
  • element v v [i] element v v [j]
  • element v h [m] element v h [n]
  • i, j, m, n represents an element index, respectively
  • coordinates of four intersections corresponding to parking spaces in the rotated image are obtained based on the element indexes i, j, m, and n.
  • the parking space extraction unit is configured to perform inverse rotation transformation based on the rotation angle of the parking space and the center point of the rotated image as the center of the circle.
  • the coordinates of the four inner corner points are projected into the input point cloud through an inverse transformation to extract a parking space.
  • An embodiment of the present application further provides a parking space extraction system for an underground garage in high-precision map production, including: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is used to run When the computer program is described, the steps of the method described in the embodiments of the present application are executed.
  • An embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored.
  • a computer program is stored on which a computer program is stored.
  • the method and system for extracting parking spaces in an underground garage during high-precision map production are based on using three-dimensional point cloud data as input.
  • the data is obtained by a laser scanner, which is an active light source and is not affected by light. ;
  • the extraction of parking spaces can effectively ensure the accuracy of the extracted parking space data, and meet the needs for the production of high-precision maps.
  • FIG. 1 is a schematic flowchart of a method for extracting parking spaces in an underground garage during high-precision map production.
  • An embodiment of the present application provides a method for extracting parking spaces in an underground garage during high-precision map production. As shown in FIG. 1, the method includes the following steps:
  • the parking space is extracted by inverse transformation to point cloud data.
  • High-precision maps represent maps based on Lane's topology network elements, which are more accurate than traditional maps.
  • the determination of a contrast estimation index of the two-dimensional bird's-eye view image is performed, and the two-dimensional bird's-eye view image is determined according to the contrast estimation index.
  • Perform image preprocessing to obtain binary images including:
  • the binary image is obtained by Russ filtering and Gaussian adaptive binarization.
  • the binary image satisfies the following expression:
  • I b represents a binary image
  • I represents a two-dimensional bird's-eye view image
  • t e represents a first threshold value
  • e represents contrast
  • medianBlur () represents a median filtering process
  • gB represents a Gaussian adaptive binarization process
  • close Indicates morphological closed processing
  • localLaplacian () indicates local Laplacian filtering processing.
  • the detecting a straight line segment of the binary image and determining a rotation angle of the parking line of the parking space according to the detection result includes:
  • the probabilistic Hough transform detects a set of straight line segments with the directionality of the parking line in the binary image, traversing the set of straight line segments to obtain a straight line segment length greater than a first threshold, and an angle between the straight line segment and a specific direction meets a preset condition straight line Segment subset; calculating a straight segment length and an inclination angle in the straight segment subset, and determining a parking line rotation angle of the parking space according to the straight segment length and the inclination angle.
  • the binary image I b has a set of straight line segments with the directionality of the parking line detected by the probabilistic Hough transform, and the set of straight line segments is traversed to obtain a length of the straight line segment greater than a first threshold (the first threshold is, for example, t (Where t is a positive integer) and the angle between a straight line segment and a specific direction satisfies a preset condition.
  • the first threshold is, for example, t (Where t is a positive integer) and the angle between a straight line segment and a specific direction satisfies a preset condition.
  • the specific direction may be a horizontal direction or a vertical direction.
  • an angle between the straight line segment and the specific direction satisfies a preset condition, including: The included angle is within a preset threshold range, that is, the straight line segments in the obtained straight line segment subset l k are straight line segments with a length greater than t and a small angle difference between the straight line segments; the straight line segment subset l is calculated The length of the straight line segment d k in k and the inclination angle a k , where k represents the index, and the inclination angle a k represents the angle between the straight line segment and a specific direction, for example, the angle between the straight line segment and the horizontal direction; The weight coefficient is determined, and the rotation angle of the parking line is determined according to the weight coefficient.
  • the weight coefficient can be expressed by the following expression:
  • the rotating the binary image according to the rotation angle and using the center point of the binary image as a circle center includes: The rotation angle of the parking line is taken as the rotation angle, and the binary image is rotated by using the center point of the binary image as a circle center. The obtained rotated image is parallel or perpendicular to the horizontal direction.
  • the resulting rotated image satisfies the following expression:
  • I (x ', y') r ((xx c ) cos ( ⁇ )-(yy c ) sin ( ⁇ ) + x c , (xx c ) sin ( ⁇ )
  • I r represents a rotated image
  • I b represents a binary image
  • (x c , y c ) represents a center point of I b
  • ⁇ represents a rotation angle of the parking line.
  • a two-dimensional coordinate system is established by using the length and width of an image (such as a binary image) as a coordinate axis, and the horizontal direction in this embodiment may be a length direction or a width direction of the image.
  • Integral projection including:
  • the number of pixels in each column and row in the rotated image containing parking lines is determined, and a one-dimensional vector representing horizontal and vertical integral projections is obtained.
  • the horizontal integral projection vector representing horizontal can be expressed as v v , which represents vertical
  • the integral projection vector of can be represented by v h .
  • the coordinates of four inner corners of the parking space are obtained according to the horizontal and vertical integral projection search, including: expressing horizontal and At the center index of the vertical integral projection vector, the first element whose gray value corresponding to the index is greater than the second threshold is searched in the positive and negative directions, and the elements v v [i], v v [j], and v h [m] And element v h [n]; where i, j, m, and n respectively represent element indexes; and based on the element indexes i, j, m, and n, coordinates of four intersections corresponding to parking spaces in the rotated image are obtained.
  • the first element as an embodiment, and by the vector v v v h center index corresponding to the search index to the positive and negative directions gray value greater than a second threshold value, respectively, to give the element v v [i], the element v v [j ], Element v h [m] and element v h [n], and the coordinates of the four intersection points (x i , y m ) of the parking line in the rotated image Ir are obtained from the element indexes i, j, m, n, ( x j , y m ), (x j , y n ), (x i , y n ) are the coordinates corresponding to the four inner corner points.
  • the extracting a parking space through inverse transformation to point cloud data according to the coordinates of the four internal corner points includes:
  • the coordinates of the four inner corner points on the inverse rotation transformation are projected into the input point cloud after inverse transformation to extract parking space.
  • An embodiment of the present application further provides a parking space extraction system for an underground garage in high-precision map production, and the system includes:
  • a projection unit configured to obtain three-dimensional laser point cloud data including a parking space, and project the three-dimensional laser point cloud data into a two-dimensional bird's-eye view image;
  • a contrast estimation unit configured to determine a ratio estimation index of the two-dimensional bird's-eye view mode image, and performing image preprocessing on the two-dimensional bird's-eye view mode image according to the contrast estimation index to obtain a binary image
  • An angle estimation unit configured to detect a straight line segment of the binary image, and determine a parking line rotation angle of the parking space according to the detection result
  • a rotation unit configured to rotate the binary image according to the rotation angle of the parking line and use the center point of the binary image as a circle center to obtain a rotated image
  • a statistics unit configured to count the number of pixels in each row and each column in the rotated image that includes a parking line, and obtain integral projections of horizontal and vertical methods, respectively;
  • a coordinate search unit configured to obtain coordinates of four inner corner points corresponding to a parking space according to the horizontal and vertical integral projection search
  • the parking space extraction unit is configured to extract a parking space through inverse transformation to point cloud data according to the four internal corner point coordinates.
  • the morphological closing process, the local Laplacian filtering process, and the Gaussian adaptive binarization process are performed in order to obtain a binary image.
  • the angle estimation unit is configured to detect a set of straight line segments having a parking line directionality in the binary image by a probabilistic Hough transform, and traverse the set of straight line segments to obtain a straight line.
  • the segment length is greater than the first threshold and the angle between the straight line segment and the specific direction satisfies a preset condition.
  • a subset of straight line segments is calculated.
  • the straight segment length and the inclination angle in the straight segment subset are calculated, and determined based on the straight segment length and the inclination angle.
  • the rotation angle of the parking line of the parking space is configured to detect a set of straight line segments having a parking line directionality in the binary image by a probabilistic Hough transform, and traverse the set of straight line segments to obtain a straight line.
  • the segment length is greater than the first threshold and the angle between the straight line segment and the specific direction satisfies a preset condition.
  • a subset of straight line segments is calculated.
  • the rotation unit is configured to use the rotation angle of the parking line as a rotation angle, and rotate the binary image by using the center point of the binary image as the center of the circle.
  • the parking line is parallel or vertical to the horizontal direction.
  • the statistics unit is configured to separately determine the number of pixels in each column and row in the rotated image that includes parking lines, and obtain a horizontal and vertical integral projection, respectively. Dimensional vector.
  • the coordinate search unit is configured to include a first gray value corresponding to a search index in the positive and negative directions at the center index of the horizontal and vertical integral projection vector, respectively, which is greater than the second threshold value.
  • the parking space extraction unit is configured to inversely rotate the coordinates of the four inner corner points based on the rotation angle of the parking space with the center point of the rotated image as the center of the circle, After inverse transformation projection into the input point cloud, parking spaces are extracted.
  • the underground parking space extraction system in the high-precision map production only uses the above-mentioned division of the program modules for illustration when performing underground parking space extraction.
  • the above processing allocation is completed by different program modules, that is, the internal structure of the system is divided into different program modules to complete all or part of the processing described above.
  • the underground parking space extraction system in the high-precision map production provided by the foregoing embodiment and the embodiment of the underground parking space extraction method in the high-precision map production belong to the same concept, and the specific implementation process is described in the method embodiment, and is not repeated here. .
  • An embodiment of the present application further provides a parking space extraction system for an underground garage in high-precision map production, including: a processor and a memory for storing a computer program capable of running on the processor, wherein the processor is used to run When the computer program is described, the steps of the method described in the embodiments of the present application are executed.
  • the memory may be a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memory.
  • the non-volatile memory may be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), or an erasable programmable read-only memory (EPROM, Erasable Programmable Read- Only Memory), Electrically Erasable Programmable Read-Only Memory (EEPROM), Magnetic Random Access Memory (FRAM, ferromagnetic random access memory), Flash Memory (Flash Memory), Magnetic Surface Memory , Compact disc, or read-only compact disc (CD-ROM, Compact Disc-Read-Only Memory); the magnetic surface memory can be a disk memory or a tape memory.
  • the volatile memory may be random access memory (RAM, Random Access Memory), which is used as an external cache.
  • RAM random access memory
  • RAM Random Access Memory
  • many forms of RAM are available, such as Static Random Access Memory (SRAM, Static Random Access Memory), Synchronous Static Random Access Memory (SSRAM, Static Random Access, Memory), Dynamic Random Access DRAM (Dynamic Random Access Memory), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM, Double Data Rate Synchronous Dynamic Random Access Memory), enhanced Type Synchronous Dynamic Random Access Memory (ESDRAM, Enhanced Random Dynamic Access Memory), Synchronous Link Dynamic Random Access Memory (SLDRAM, SyncLink Dynamic Random Access Memory), Direct Memory Bus Random Access Memory (DRRAM, Direct Rambus Random Access Memory) ).
  • SRAM Static Random Access Memory
  • SSRAM Synchronous Static Random Access Memory
  • SDRAM Synchronous Dynamic Random Access Memory
  • DDRSDRAM Double Data Rate Synchronous Dynamic Random Access Memory
  • the method disclosed in the foregoing embodiments of the present invention may be applied to a processor, or implemented by a processor.
  • the processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
  • the foregoing processor may be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like.
  • DSP Digital Signal Processor
  • the processor may implement or execute various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention.
  • a general-purpose processor may be a microprocessor or any conventional processor.
  • the steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a storage medium.
  • the storage medium is located in the memory.
  • the processor reads the information in the memory and completes the steps of the foregoing method in combination with its hardware.
  • An embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored.
  • a computer program is stored on which a computer program is stored.
  • the method and system for extracting parking spaces in underground garages provided in the present invention for implementing high-precision map production have the following beneficial effects: the use of three-dimensional point cloud data as input is obtained by a laser scanner, and the laser scanner is An active light source that is not affected by light; using different image preprocessing methods to improve the robustness of the algorithm based on image quality evaluation; improved probabilistic Huff transform to detect image tilt angles and improve consistency between detection and detection objects; using rotational projection
  • the image method obtains the coordinates of the intersection of the parking lines and extracts the parking space, which can effectively ensure the accuracy of the extracted parking space data and meet the precision requirements for the production of high-precision maps.
  • the units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, which may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the objective of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above integration
  • the unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • the foregoing program may be stored in a computer-readable storage medium.
  • the program is executed, the program is executed.
  • the method includes the steps of the foregoing method embodiment.
  • the foregoing storage medium includes: various types of media that can store program codes, such as a mobile storage device, a ROM, a RAM, a magnetic disk, or an optical disc.
  • the above-mentioned integrated unit of the present application is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
  • the computer software product is stored in a storage medium and includes several instructions for A computer device (which may be a personal computer, a server, or a network device) is caused to perform all or part of the methods described in the embodiments of the present application.
  • the foregoing storage medium includes: various types of media that can store program codes, such as a mobile storage device, a ROM, a RAM, a magnetic disk, or an optical disc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Remote Sensing (AREA)
  • Multimedia (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Automation & Control Theory (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Astronomy & Astrophysics (AREA)
  • Educational Administration (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Educational Technology (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Ecology (AREA)
  • Probability & Statistics with Applications (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Hardware Design (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

一种高精度地图制作中地下车库停车位提取方法及系统，该方法包括：获得包含停车位的三维激光点云数据，将三维激光点云数据投影为二维鸟瞰模式图像（S1）；确定二维鸟瞰模式图像的对比度估计指标，根据对比度估计指标对二维鸟瞰模式图像进行图像预处理，获得二值图像（S2）；检测二值图像的直线段，根据检测结果确定停车位的停车线旋转角度（S3）；根据停车线旋转角度并以二值图像中心点为圆心旋转二值图像，获得旋转图像（S4）；统计旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方向的积分投影（S5）；根据水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标（S6）；根据四个内角点坐标通过逆变换到点云数据，提取出停车位（S7）。

Description

高精度地图制作中地下车库停车位提取方法及系统
相关申请的交叉引用
本申请基于申请号为201811033589.X、申请日为2018年9月5日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此以引入方式并入本申请。
技术领域
本申请涉及高精度地图制作技术领域，特别涉及一种高精度地图制作中地下车库停车位提取方法及系统。
背景技术
高精地图是无人驾驶核心技术之一，精准的地图对无人车定位、导航与控制，以及安全至关重要，如何生成高精度地图也是无人驾驶的领域亟待解决的问题。地下车库停车位表示建筑在地下并可供机动车长期或长期或临时停放的区域，由停车线按一定大小划分了每辆车了每辆车的停车区域。地下停车场与不同等级的城市道路相配合，满足不同规模的停车需要，对城市中心区的交通起到非常重要的调节和控制作用。高精度的地下车库停车位数据，作为高精度地图的重要一部分尤为重要。
现有的停车位提取方法，往往是基于原始图像数据的提取方法，采用边缘检测方法进行边缘检测得到停车线的边缘点集，再对边缘点集进行霍夫变换与直线提取，通过对停车线的提取得到最终的停车位。
但是该方法对光照敏感，在不同光照条件下，停车位在图像中的梯度差异较大，容易导致错提取和漏提取。与此同时，由于实际应用中边缘检测方法得到的边缘点集中存在噪声不完全是停车线的边缘，单一使用霍夫 变换与直线提取容易造成误差，从而导致提取精度不高，无法满足高精度地图的精度需求。
发明内容
本申请实施例期望提出一种高精度地图制作中地下车库停车位提取方法及系统。
本申请实施例提供的一种高精度地图制作中地下车库停车位提取方法，所述方法包括：
获得包含停车位的三维激光点云数据，将所述三维激光点云数据投影为二维鸟瞰模式图像；确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，获得旋转图像；统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像，包括：
使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；
其中，I表示二维鸟瞰模式图像，e表示对比度；
比较所述对比度和给定阈值，获得比较结果；
在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处 理得到二值图像；
在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度，包括：由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线线段长度大于第一阈值且直线段与特定方向之间夹角满足预设条件直线段子集；计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，包括：以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影，包括：分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和垂直积分投影一维向量。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述水平与垂直方向的积分投影搜索得到停车位四个内角点坐标，包括：由分别表示水平和垂直积分投影向量中心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素，分别得到元素v v[i]、 元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位，包括：根据所述停车位旋转角度，以所述旋转图像的中心点为圆心逆旋转变换所述四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
本申请实施例还提供一种高精度地图制作中地下车库停车位提取系统，所述系统包括：
投影单元，配置为获得包含停车位的三维激光点云数据，将所述三维激光点云数据投影为二维鸟瞰模式图像；
对比度估计单元，配置为确定所述二维鸟瞰模式图像的比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；
角度估计单元，配置为检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；
旋转单元，配置为根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，获得旋转图像；
统计单元，配置为统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；
坐标搜索单元，配置为根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；
车位提取单元，配置为对根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述对比度估计单元，配置为使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；其中，I表示二维鸟瞰模式图像，e表示对比度；比较所述对比度和给定阈值，获得比较结果；在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处理得到二值图像；在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述角度估计单元，配置为由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线段长度大于第一阈值且直线线段与特定方向之间夹角满足预设条件直线段子集；计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述旋转单元，配置为以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述统计单元，配置为分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和垂直积分投影一维向量。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述坐标搜索单元，配置为由分别表示水平和垂直积分投影向量中心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素，分 别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
在本申请实施例所述的高精度地图制作中地下车库停车位提取系统中，所述车位提取单元，配置为根据所述停车位旋转角度，以所述旋转图像的中心点为圆心逆旋转变换所述四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
本申请实施例还提供了一种高精度地图制作中地下车库停车位提取系统，包括：处理器和用于存储能够在处理器上运行的计算机程序的存储器，其中，所述处理器用于运行所述计算机程序时，执行本申请实施例所述方法的步骤。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现本申请实施例所述方法的步骤。
本申请实施例提供的高精度地图制作中地下车库停车位提取方法及系统，在于使用三维点云数据作为输入，该数据由激光扫描仪得到，激光扫描仪为一种主动光源，不受光照影响；根据图像质量评价使用不同的图像预处理方法提高了算法鲁棒性；改进概率霍夫变换检测图像倾斜角，提高检测与检测对象的一致性；使用旋转投影图像的方法求取停车线交点坐标进行停车位提取，能够有效保证提取的停车位数据的精度，满足高精度地图的制作精度需求。
附图说明
图1是高精度地图制作中地下车库停车位提取方法的流程示意图。
具体实施方式
下面结合附图及具体实施例对本发明作进一步详细的说明。
本申请实施例提供了一种高精度地图制作中地下车库停车位提取方法，如图1所示，所述方法包括如下步骤：
S1、获得包含停车位的三维(3D)激光点云数据，将所述三维激光点云投影为二维(2D)鸟瞰模式图像；
S2、确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；
S3、检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；
S4、根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，获得旋转图像；
S5、统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；
S6、根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；
S7、根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位。
高精度地图表示基于Lane的拓扑网元素组成的地图，相较于传统地图地理信息更精确。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像，包括：
使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；其中，I表示二维鸟瞰模式图像，e表示对比度，std()表示图像标准差估计处理；比较所述对比度和给定阈值，获得比较结果；在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处理得到二值图 像；在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
作为一种示例，所述二值图像满足以下表达式：
Figure PCTCN2019086895-appb-000001
其中，I b表示二值图像，I表示二维鸟瞰模式图像；t e表示第一阈值；e表示对比度；medianBlur()表示中值滤波处理，gB()表示高斯自适应二值化处理，close()表示形态学闭处理，localLaplacian()表示局部拉普拉斯滤波处理。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度，包括：由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线线段长度大于第一阈值且直线线段与特定方向之间夹角满足预设条件直线段子集；计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
作为一种实施方式，由概率霍夫变换检测出二值图像I b具有停车线方向性的直线段集合，遍历直线段集合，获得直线段长度大于第一阈值(所述第一阈值例如通过t表示，t为正整数)且直线段与特定方向之间夹角满足预设条件直线段子集。作为一种示例，所述特定方向可以是水平方向或垂直方向，以所述水平方向为例，则所述直线段与特定方向之间夹角满足预设条件，包括：直线段与水平方向之间的夹角在预设阈值范围内，也即获得的直线段子集l k中的直线段均为长度大于t、且各直线段之间的角度相差较小的直线段；计算直线段子集l k中的直线段长度d k和倾斜角a k，其中，k表示索引，所述倾斜角a k表示直线段与特定方向的夹角，例如表示直线 段与水平方向的夹角；再根据所述确定权重系数，根据所述权重系数确定停车线旋转角度。作为一种示例，所述权重系数可通过以下表达式表示：
Figure PCTCN2019086895-appb-000002
则所述停车位旋转角度可表示为：θ＝w ka k。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，包括：以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
作为一种示例，得到的旋转图像满足以下表达式：
I(x',y') r＝((x-x c)cos(θ)-(y-y c)sin(θ)+x c,(x-x c)sin(θ)
-(y-y c)cos(θ)+y c)         (2)
其中，I r表示旋转图像；I b表示二值图像，(x c，y c)表示I b的中心点；θ表示停车线旋转角度。本实施例中以图像(如二值图像)的长度和宽度为坐标轴建立二维坐标系，本实施例中的水平方向可以为图像的长度方向或宽度方向。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影，包括：
分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和垂直积分投影一维向量；其中，表示水平的积分投影向量可表示为v v，表示垂直的积分投影向量可表示v h。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述水平与垂直方向的积分投影搜索得到停车位四个内角点坐标，包括：由分别表示水平和垂直积分投影向量中心索引处向正负方向 搜索索引对应的灰度值大于第二阈值的第一个元素，分别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
作为一种实施方式，由向量v v和v h中心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素分别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]，并由元素索引i、j、m、n的得到停车线在旋转图像Ir中的四个交点坐标(x i，y m)、(x j，y m)、(x j，y n)、(x i，y n)，即为四个内角点对应的坐标。
在本申请实施例所述的高精度地图制作中地下车库停车位提取方法中，所述根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位，包括：
根据所述停车位旋转角度，以所述旋转图像的中心点(x c’,y c’)为圆心逆旋转变换上四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
本申请实施例还提供一种高精度地图制作中地下车库停车位提取系统，所述系统包括：
投影单元，配置为获得包含停车位的三维激光点云数据，将所述三维激光点云数据投影为二维鸟瞰模式图像；
对比度估计单元，配置为确定所述二维鸟瞰模式图像的比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；
角度估计单元，配置为检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；
旋转单元，配置为根据所述停车线旋转角度并以所述二值图像中心点 为圆心旋转所述二值图像，获得旋转图像；
统计单元，配置为统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；
坐标搜索单元，配置为根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；
车位提取单元，配置为对根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位。
在本申请的一种可选实施例中，所述对比度估计单元，配置为使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；其中，I表示二维鸟瞰模式图像，e表示对比度；比较所述对比度和给定阈值，获得比较结果；在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处理得到二值图像；在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
在本申请的一种可选实施例中，所述角度估计单元，配置为由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线段长度大于第一阈值且直线线段与特定方向之间夹角满足预设条件直线段子集；计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
在本申请的一种可选实施例中，所述旋转单元，配置为以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
在本申请的一种可选实施例中，所述统计单元，配置为分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和 垂直积分投影一维向量。
在本申请的一种可选实施例中，所述坐标搜索单元，配置为由分别表示水平和垂直积分投影向量中心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素，分别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
在本申请的一种可选实施例中，所述车位提取单元，配置为根据所述停车位旋转角度，以所述旋转图像的中心点为圆心逆旋转变换所述四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
需要说明的是：上述实施例提供的高精度地图制作中地下车库停车位提取系统在进行地下车库停车位提取时，仅以上述各程序模块的划分进行举例说明，实际应用中，可以根据需要而将上述处理分配由不同的程序模块完成，即将系统的内部结构划分成不同的程序模块，以完成以上描述的全部或者部分处理。另外，上述实施例提供的高精度地图制作中地下车库停车位提取系统与高精度地图制作中地下车库停车位提取方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
本申请实施例还提供了一种高精度地图制作中地下车库停车位提取系统，包括：处理器和用于存储能够在处理器上运行的计算机程序的存储器，其中，所述处理器用于运行所述计算机程序时，执行本申请实施例所述方法的步骤。
可以理解，存储器可以是易失性存储器或非易失性存储器，也可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(ROM，Read Only Memory)、可编程只读存储器(PROM，Programmable Read-Only Memory)、可擦除可编程只读存储器(EPROM，Erasable Programmable Read-Only Memory)、电可擦除可编程只读存储器(EEPROM， Electrically Erasable Programmable Read-Only Memory)、磁性随机存取存储器(FRAM，ferromagnetic random access memory)、快闪存储器(Flash Memory)、磁表面存储器、光盘、或只读光盘(CD-ROM，Compact Disc Read-Only Memory)；磁表面存储器可以是磁盘存储器或磁带存储器。易失性存储器可以是随机存取存储器(RAM，Random Access Memory)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(SRAM，Static Random Access Memory)、同步静态随机存取存储器(SSRAM，Synchronous Static Random Access Memory)、动态随机存取存储器(DRAM，Dynamic Random Access Memory)、同步动态随机存取存储器(SDRAM，Synchronous Dynamic Random Access Memory)、双倍数据速率同步动态随机存取存储器(DDRSDRAM，Double Data Rate Synchronous Dynamic Random Access Memory)、增强型同步动态随机存取存储器(ESDRAM，Enhanced Synchronous Dynamic Random Access Memory)、同步连接动态随机存取存储器(SLDRAM，SyncLink Dynamic Random Access Memory)、直接内存总线随机存取存储器(DRRAM，Direct Rambus Random Access Memory)。本发明实施例描述的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
上述本发明实施例揭示的方法可以应用于处理器中，或者由处理器实现。处理器可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器、数字信号处理器(DSP，Digital Signal Processor)，或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。处理器可以实现或者执行本发明实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本发明实施例所公开的方法的步骤，可以直接体现为硬 件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于存储介质中，该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成前述方法的步骤。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现本申请实施例所述方法的步骤。
实施本发明提供的高精度地图制作中地下车库停车位提取方法及系统与现有技术相比具有以下有益效果：在于使用三维点云数据作为输入，该数据由激光扫描仪得到，激光扫描仪为一种主动光源，不受光照影响；根据图像质量评价使用不同的图像预处理方法提高了算法鲁棒性；改进概率霍夫变换检测图像倾斜角，提高检测与检测对象的一致性；使用旋转投影图像的方法求取停车线交点坐标进行停车位提取，能够有效保证提取的停车位数据的精度，满足高精度地图的制作精度需求。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统和方法，可以通过其它的方式实现。以上所描述的系统实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，如：多个单元或组件可以结合，或可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的各组成部分相互之间的耦合、或直接耦合、或通信连接可以是通过一些接口，设备或单元的间接耦合或通信连接，可以是电性的、机械的或其它形式的。
上述作为分离部件说明的单元可以是、或也可以不是物理上分开的，作为单元显示的部件可以是、或也可以不是物理单元，即可以位于一个地方，也可以分布到多个网络单元上；可以根据实际的需要选择其中的部分或全部单元来实现本实施例方案的目的。
另外，在本申请各实施例中的各功能单元可以全部集成在一个处理单元中，也可以是各单元分别单独作为一个单元，也可以两个或两个以上单 元集成在一个单元中；上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：移动存储设备、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
或者，本申请上述集成的单元如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实施例的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、或者网络设备等)执行本申请各个实施例所述方法的全部或部分。而前述的存储介质包括：移动存储设备、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (16)

  1. 一种高精度地图制作中地下车库停车位提取方法，所述方法包括：
    获得包含停车位的三维激光点云数据，将所述三维激光点云数据投影为二维鸟瞰模式图像；
    确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；
    检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；
    根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，获得旋转图像；
    统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；
    根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；
    根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位。
  2. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述确定所述二维鸟瞰模式图像的对比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像，包括：
    使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；
    其中，I表示二维鸟瞰模式图像，e表示对比度；
    比较所述对比度和给定阈值，获得比较结果；
    在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处 理得到二值图像；
    在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
  3. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度，包括：
    由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线线段长度大于第一阈值且直线段与特定方向之间夹角满足预设条件直线段子集；
    计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
  4. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，包括：
    以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
  5. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影，包括：
    分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和垂直积分投影一维向量。
  6. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述根据所述水平与垂直方向的积分投影搜索得到停车位四个内角点坐标，包括：
    由分别表示水平和垂直积分投影向量中心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素，分别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；
    基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
  7. 如权利要求1所述的高精度地图制作中地下车库停车位提取方法，其中，所述根据所述四个内角点坐标通过逆变换到点云数据，提取出停车位，包括：
    根据所述停车位旋转角度，以所述旋转图像的中心点为圆心逆旋转变换所述四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
  8. 一种高精度地图制作中地下车库停车位提取系统，所述系统包括：
    投影单元，配置为获得包含停车位的三维激光点云数据，将所述三维激光点云数据投影为二维鸟瞰模式图像；
    对比度估计单元，配置为确定所述二维鸟瞰模式图像的比度估计指标，根据所述对比度估计指标对所述二维鸟瞰模式图像进行图像预处理，获得二值图像；
    角度估计单元，配置为检测所述二值图像的直线段，根据检测结果确定所述停车位的停车线旋转角度；
    旋转单元，配置为根据所述停车线旋转角度并以所述二值图像中心点为圆心旋转所述二值图像，获得旋转图像；
    统计单元，配置为统计所述旋转图像中每行、每列包含停车线的像素点个数，分别得到水平与垂直方法的积分投影；
    坐标搜索单元，配置为根据所述水平与垂直方向的积分投影搜索得到对应于停车位的四个内角点坐标；
    车位提取单元，配置为对根据所述四个内角点坐标通过逆变换到点云 数据，提取出停车位。
  9. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述对比度估计单元，配置为使用图像标准差估计所述二维鸟瞰模式图像的对比度，所述对比度满足e＝std(I)；其中，I表示二维鸟瞰模式图像，e表示对比度；比较所述对比度和给定阈值，获得比较结果；在所述比较结果为所述对比度小于第一阈值的情况下，对所述二维鸟瞰模式图像依次进行中值滤波处理、高斯自适应二值化处理和形态学闭处理得到二值图像；在所述比较结果为所述对比度大于等于所述第一阈值的情况下，对所述二维鸟瞰模式图像依次进行形态学闭处理、局部拉普拉斯滤波处理、高斯自适应二值化处理得到二值图像。
  10. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述角度估计单元，配置为由概率霍夫变换检测出所述二值图像具有停车线方向性的直线段集合，遍历所述直线段集合获得直线段长度大于第一阈值且直线线段与特定方向之间夹角满足预设条件直线段子集；计算所述直线段子集中的直线段长度和倾斜角，根据所述直线段长度和所述倾斜角确定所述停车位的停车线旋转角度。
  11. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述旋转单元，配置为以所述停车线旋转角度作为旋转角，并以所述二值图像中心点为圆心旋转所述二值图像，得到的旋转图像中停车线与水平方向平行或垂直。
  12. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述统计单元，配置为分别确定所述旋转图像中每列、每行包含停车线的像素点的个数，得到分别表示水平和垂直积分投影一维向量。
  13. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述坐标搜索单元，配置为由分别表示水平和垂直积分投影向量中 心索引处向正负方向搜索索引对应的灰度值大于第二阈值的第一个元素，分别得到元素v v[i]、元素v v[j]、元素v h[m]和元素v h[n]；其中，i、j、m、n分别表示元素索引；基于所述元素索引i、j、m、n获得所述旋转图像中对应于停车位的四个交点坐标。
  14. 如权利要求8所述的高精度地图制作中地下车库停车位提取系统，其中，所述车位提取单元，配置为根据所述停车位旋转角度，以所述旋转图像的中心点为圆心逆旋转变换所述四个内角点的坐标，经过逆变换投影到输入点云中，提取出停车位。
  15. 一种高精度地图制作中地下车库停车位提取系统，包括：处理器和用于存储能够在处理器上运行的计算机程序的存储器，其中，所述处理器用于运行所述计算机程序时，执行权利要求1至7任一项所述方法的步骤。
  16. 一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现权利要求1至7任一项所述方法的步骤。
PCT/CN2019/086895 2018-09-05 2019-05-14 高精度地图制作中地下车库停车位提取方法及系统 WO2020048152A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/618,440 US20200152060A1 (en) 2018-09-05 2019-05-14 Underground garage parking space extraction method and system for high-definition map making

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811033589.XA CN109243289B (zh) 2018-09-05 2018-09-05 高精度地图制作中地下车库停车位提取方法及系统
CN201811033589.X 2018-09-05

Publications (1)

Publication Number Publication Date
WO2020048152A1 true WO2020048152A1 (zh) 2020-03-12

Family

ID=65067242

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/086895 WO2020048152A1 (zh) 2018-09-05 2019-05-14 高精度地图制作中地下车库停车位提取方法及系统

Country Status (3)

Country Link
US (1) US20200152060A1 (zh)
CN (1) CN109243289B (zh)
WO (1) WO2020048152A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111428616A (zh) * 2020-03-20 2020-07-17 东软睿驰汽车技术（沈阳）有限公司 一种车位检测方法、装置、设备及存储介质
CN111813882A (zh) * 2020-06-18 2020-10-23 浙江大华技术股份有限公司 一种机器人地图构建方法、设备及存储介质
CN115564673A (zh) * 2022-09-26 2023-01-03 浙江省测绘科学技术研究院 三维点云地下车库柱状物提取与矢量自动生成方法及系统
CN115661098A (zh) * 2022-10-31 2023-01-31 河海大学 一种海底管线二维冲刷剖面图像识别与数据提取方法

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109243289B (zh) * 2018-09-05 2021-02-05 武汉中海庭数据技术有限公司 高精度地图制作中地下车库停车位提取方法及系统
CN110132278B (zh) * 2019-05-14 2021-07-02 驭势科技（北京）有限公司 一种即时定位与建图的方法及装置
CN110232835B (zh) * 2019-06-27 2020-11-13 浙江工业大学 一种基于图像处理的地下车库停车位检测方法
CN110390306B (zh) * 2019-07-25 2021-08-10 湖州宏威新能源汽车有限公司 直角车位的检测方法、车辆和计算机可读存储介质
CN111159811B (zh) * 2020-01-02 2023-06-09 广东博智林机器人有限公司 地下车库布局方法、装置、设备和存储介质
CN111754468A (zh) * 2020-06-10 2020-10-09 深圳南方德尔汽车电子有限公司 基于投影的车位检测方法、装置、计算机设备及存储介质
CN112084943A (zh) * 2020-09-09 2020-12-15 中国航空工业集团公司雷华电子技术研究所 深度语义分割结合霍夫变换的高压线检测方法
CN112991808B (zh) * 2020-12-29 2022-09-30 杭州海康威视数字技术股份有限公司 一种停车区域的车位显示方法、装置及电子设备
CN112767425A (zh) * 2020-12-30 2021-05-07 智车优行科技（北京）有限公司 一种基于视觉的车位检测方法及装置
CN113192356B (zh) * 2021-04-01 2023-01-03 上海欧菲智能车联科技有限公司 一种多传感器融合的车位检测方法、装置及车辆
CN113240734B (zh) * 2021-06-01 2024-05-17 深圳市捷顺科技实业股份有限公司 一种基于鸟瞰图的车辆跨位判断方法、装置、设备及介质
GB2607598A (en) * 2021-06-07 2022-12-14 Correvate Ltd Aligning 3D datasets

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010042466A1 (en) * 2008-10-06 2010-04-15 Kevin Scott Williams Apparatus and method for classifying point cloud data based on principal axes
CN101807352A (zh) * 2010-03-12 2010-08-18 北京工业大学 一种模糊模式识别的车位检测方法
US8537338B1 (en) * 2011-08-24 2013-09-17 Hrl Laboratories, Llc Street curb and median detection using LIDAR data
CN105355083A (zh) * 2015-12-14 2016-02-24 宁波裕兰信息科技有限公司 一种基于视觉的360度泊车辅助智能引导系统
CN108281041A (zh) * 2018-03-05 2018-07-13 东南大学 一种基于超声波和视觉传感器相融合的泊车车位检测方法
CN109243289A (zh) * 2018-09-05 2019-01-18 武汉中海庭数据技术有限公司 高精度地图制作中地下车库停车位提取方法及系统

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH1131296A (ja) * 1997-07-10 1999-02-02 Hitachi Cable Ltd 駐車場管理方法及びその装置
CN101656023B (zh) * 2009-08-26 2011-02-02 西安理工大学 一种视频监视模式下的室内停车场管理方法
CN104170369B (zh) * 2012-03-15 2017-12-19 诺基亚技术有限公司 用于捕获图像的方法、设备和计算机可读存储介质
CN103473950B (zh) * 2012-06-06 2017-04-12 刘鉵 一种停车场车位监控方法
DE102014224104A1 (de) * 2014-11-26 2016-06-02 Robert Bosch Gmbh Verfahren und Vorrichtung zum Betreiben eines Fahrzeugs respektive eines Parkplatzes
CN106897655A (zh) * 2015-12-18 2017-06-27 富士通株式会社 停车位的检测装置、方法以及图像处理设备
JP6564713B2 (ja) * 2016-02-01 2019-08-21 三菱重工業株式会社 自動運転制御装置、車両及び自動運転制御方法
CN107463918B (zh) * 2017-08-17 2020-04-24 武汉大学 基于激光点云与影像数据融合的车道线提取方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010042466A1 (en) * 2008-10-06 2010-04-15 Kevin Scott Williams Apparatus and method for classifying point cloud data based on principal axes
CN101807352A (zh) * 2010-03-12 2010-08-18 北京工业大学 一种模糊模式识别的车位检测方法
US8537338B1 (en) * 2011-08-24 2013-09-17 Hrl Laboratories, Llc Street curb and median detection using LIDAR data
CN105355083A (zh) * 2015-12-14 2016-02-24 宁波裕兰信息科技有限公司 一种基于视觉的360度泊车辅助智能引导系统
CN108281041A (zh) * 2018-03-05 2018-07-13 东南大学 一种基于超声波和视觉传感器相融合的泊车车位检测方法
CN109243289A (zh) * 2018-09-05 2019-01-18 武汉中海庭数据技术有限公司 高精度地图制作中地下车库停车位提取方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WANG. SONG ET AL: "Comparison and Discussion of Measurement Method About the Underground Parking Spaces", URBAN GEOTECHNICAL INVESTIGATION & SURVEYING, no. 6, 31 December 2015 (2015-12-31), pages 109 - 112, ISSN: 1672-8262 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111428616A (zh) * 2020-03-20 2020-07-17 东软睿驰汽车技术（沈阳）有限公司 一种车位检测方法、装置、设备及存储介质
CN111428616B (zh) * 2020-03-20 2023-05-23 东软睿驰汽车技术（沈阳）有限公司 一种车位检测方法、装置、设备及存储介质
CN111813882A (zh) * 2020-06-18 2020-10-23 浙江大华技术股份有限公司 一种机器人地图构建方法、设备及存储介质
CN111813882B (zh) * 2020-06-18 2024-05-14 浙江华睿科技股份有限公司 一种机器人地图构建方法、设备及存储介质
CN115564673A (zh) * 2022-09-26 2023-01-03 浙江省测绘科学技术研究院 三维点云地下车库柱状物提取与矢量自动生成方法及系统
CN115564673B (zh) * 2022-09-26 2024-03-15 浙江省测绘科学技术研究院 三维点云地下车库柱状物提取与矢量自动生成方法及系统
CN115661098A (zh) * 2022-10-31 2023-01-31 河海大学 一种海底管线二维冲刷剖面图像识别与数据提取方法
CN115661098B (zh) * 2022-10-31 2024-02-06 河海大学 一种海底管线二维冲刷剖面图像识别与数据提取方法

Also Published As

Publication number Publication date
CN109243289A (zh) 2019-01-18
US20200152060A1 (en) 2020-05-14
CN109243289B (zh) 2021-02-05

Similar Documents

Publication Publication Date Title
WO2020048152A1 (zh) 高精度地图制作中地下车库停车位提取方法及系统
WO2018068653A1 (zh) 点云数据处理方法、装置及存储介质
US9443309B2 (en) System and method for image based mapping, localization, and pose correction of a vehicle with landmark transform estimation
WO2021073656A1 (zh) 图像数据自动标注方法及装置
WO2021217625A1 (zh) 停车检测方法、系统、处理设备和存储介质
US11625851B2 (en) Geographic object detection apparatus and geographic object detection method
WO2020043081A1 (zh) 定位技术
EP3904831A1 (en) Visual localization using a three-dimensional model and image segmentation
CN114120149B (zh) 一种倾斜摄影测量建筑物特征点提取方法、装置、电子设备及介质
WO2020156923A2 (en) Map and method for creating a map
Wang et al. Multi-cue road boundary detection using stereo vision
CN115205803A (zh) 自动驾驶环境感知方法、介质及车辆
CN116452852A (zh) 一种高精度矢量地图的自动生成方法
Bernardi et al. High integrity lane-level occupancy estimation of road obstacles through LiDAR and HD map data fusion
EP4287137A1 (en) Method, device, equipment, storage media and system for detecting drivable space of road
CN114969221A (zh) 一种更新地图的方法及相关设备
CN116740680A (zh) 一种车辆定位方法、装置及电子设备
JP2023102768A (ja) 道路地図生成システム及び使用方法
US20220366604A1 (en) Method and apparatus with calibration
CN115497061A (zh) 一种基于双目视觉的道路可行驶区域识别方法及装置
CN112183378A (zh) 一种基于颜色和深度图像的道路坡度估计方法及装置
CN115797896B (zh) 车道线聚类方法、设备和计算机可读存储介质
CN112801077B (zh) 用于自动驾驶车辆的slam初始化的方法及相关装置
CN116645483B (zh) 车道线级别道路三维模型的构建方法、装置及电子设备
Ding et al. Stereovision based generic obstacle detection and motion estimation using v-stxiel algorithm

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19857169

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19857169

Country of ref document: EP

Kind code of ref document: A1