CN113869129A - Road shoulder detection method and device and storage medium - Google Patents
Road shoulder detection method and device and storage medium Download PDFInfo
- Publication number
- CN113869129A CN113869129A CN202111016619.8A CN202111016619A CN113869129A CN 113869129 A CN113869129 A CN 113869129A CN 202111016619 A CN202111016619 A CN 202111016619A CN 113869129 A CN113869129 A CN 113869129A
- Authority
- CN
- China
- Prior art keywords
- road shoulder
- data
- dimensional
- filled
- plane
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 42
- 238000003860 storage Methods 0.000 title claims abstract description 21
- 238000011049 filling Methods 0.000 claims abstract description 67
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000004590 computer program Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 11
- 238000001914 filtration Methods 0.000 claims description 11
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000009434 installation Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002955 isolation Methods 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 230000005603 centrifugal distortion Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a road shoulder detection method, a road shoulder detection device and a storage medium, wherein the method comprises the following steps: acquiring three-dimensional point cloud data and image data of a target area; extracting road shoulder features in the image data; projecting the three-dimensional point cloud data to the road shoulder features to obtain three-dimensional road shoulder data; filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane; and projecting the road shoulder plane to the ground to obtain a road shoulder curve. According to the embodiment of the invention, the road shoulder characteristics in the image data are extracted by acquiring the three-dimensional point cloud data and the image data of the target area, and the three-dimensional road shoulder data are obtained by projecting the three-dimensional point cloud data on the road shoulder characteristics, so that the accuracy of road shoulder detection is improved; and the road shoulder characteristic surface is filled based on the three-dimensional road shoulder data to obtain a road shoulder plane, and the road shoulder plane is projected to the ground, so that a road shoulder curve can be accurately obtained, the vehicle is prevented from colliding with the road shoulder, and the control precision of automatic driving can be effectively improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a road shoulder detection method, apparatus, and storage medium.
Background
The road shoulder detection technology can be applied to the fields of semantic map drawing, automatic driving and the like. The existing road shoulder detection method mainly classifies the point cloud data collected in real time in a grouping way, and road shoulders and other objects are divided into a group in the grouping and classifying process so as to form an obstacle area which cannot be driven. The existing road shoulder detection method is only to simply integrate the road shoulder and the nearby obstacles to form an obstacle area, so that the road shoulder is difficult to accurately detect, and the accuracy of the road shoulder detection is low.
Disclosure of Invention
The invention provides a road shoulder detection method, a road shoulder detection device and a storage medium, and aims to solve the problem that the road shoulder detection precision is low due to the fact that the road shoulder and nearby obstacles are integrated to form an obstacle area in the existing road shoulder detection method, and the road shoulder is difficult to accurately detect.
One embodiment of the present invention provides a road shoulder detection method, including:
acquiring three-dimensional point cloud data and image data of a target area;
extracting road shoulder features in the image data;
projecting the three-dimensional point cloud data to the road shoulder features to obtain three-dimensional road shoulder data;
filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
and projecting the road shoulder plane to the ground to obtain a road shoulder curve.
Further, the acquiring of the three-dimensional point cloud data of the target area specifically includes:
and acquiring the three-dimensional point cloud data of the target area by adopting a radar device and combining the calibration information of the radar device relative to the position of the vehicle and the positioning information of the vehicle.
Further, the extracting of the road shoulder features in the image data specifically includes:
correcting the image data according to parameters of a camera for shooting the image data to obtain corrected image data;
and acquiring road shoulder features marked in the corrected image data.
Further, a road shoulder extraction model is established by adopting a deep learning method, and road shoulder features in the image data are extracted according to the road shoulder extraction model.
Further, the three-dimensional point cloud data is projected onto the road shoulder features to obtain three-dimensional road shoulder data, and the method specifically comprises the following steps:
and acquiring a space relative relationship between the image data and the three-dimensional point cloud data according to the parameters of the camera, and projecting the three-dimensional point cloud data onto the road shoulder features according to the space relative relationship to obtain three-dimensional road shoulder data.
Further, the step of filling the road shoulder feature plane in the three-dimensional road shoulder data to obtain a road shoulder plane specifically includes:
determining a ground line in the three-dimensional road shoulder data according to the position relation between the point cloud in the three-dimensional road shoulder data and the ground point cloud;
and forming an outer side surface to be filled and an upper plane to be filled according to the ground line, the outer side surface, the corner line and the upper plane, and filling the upper plane to be filled and the outer side surface to be filled to obtain the road shoulder plane.
Further, the determining a ground line in the three-dimensional road shoulder data according to the position relationship between the point cloud in the three-dimensional road shoulder data and the ground point cloud specifically includes:
filtering the three-dimensional road shoulder data corresponding to the road shoulder which is not in the preset height range in the three-dimensional road shoulder data;
dividing the filtered three-dimensional road shoulder data according to a preset distance interval to obtain a plurality of point clouds corresponding to the three-dimensional road shoulder data;
and if the ground line point clouds are positioned at the outer sides of the outmost point clouds in the point clouds, and the number ratio of the ground line point clouds exceeding the outmost point clouds exceeds a preset threshold value, taking the projection of the point clouds on the ground as a ground line.
Further, the outer side face to be filled and the upper plane to be filled are formed according to the ground line, the outer side face, the corner line and the upper plane, and the upper plane to be filled and the outer side face to be filled are filled to obtain the road shoulder plane, which specifically comprises the following steps:
taking the average position points of the densely distributed point clouds as upper plane surface points, taking the corner lines as upper plane surface lines, and forming an upper plane to be filled by using the upper plane surface points and the upper plane surface lines;
projecting the corner lines and the ground lines on the ground at the same time, taking lines formed by outer side points in a projection image as outer side surface lines, taking outer side points of densely distributed point clouds as outer side surface points, and forming outer sides to be filled by the outer side surface points and the outer side surface lines;
and carrying out road shoulder filling on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane.
Further, the road shoulder filling is performed on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane, and the method specifically comprises the following steps:
taking the highest point of the plane of the upper plane to be filled as a cut-off point, and filling the road shoulder upwards in a direction parallel to the ground to obtain a filled upper plane;
carrying out road shoulder filling on the outer side surface to be filled in a direction perpendicular to the normal direction of the ground and perpendicular to the connecting line of the two outermost points, wherein the filling amount does not exceed a preset threshold value, and obtaining a filling outer side surface; or taking the outermost point as a cut-off point, and filling the road shoulder to the outside in the direction vertical to the normal direction of the ground and the direction vertical to the lane line to obtain a filled outer side surface;
and forming a road shoulder plane according to the filling upper plane and the filling outer side surface.
One embodiment of the present invention provides a road shoulder detecting apparatus, including:
the data acquisition module is used for acquiring three-dimensional point cloud data and image data of a target area;
the feature extraction module is used for extracting road shoulder features in the image data;
the first projection module is used for projecting the three-dimensional point cloud data onto the road shoulder features to obtain three-dimensional road shoulder data;
the data filling module is used for filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
and the second projection module is used for projecting the road shoulder plane to the ground to obtain a road shoulder curve.
An embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute the road shoulder detecting method as described above.
According to the embodiment of the invention, the road shoulder characteristics in the image data are extracted by acquiring the three-dimensional point cloud data and the image data of the target area, and the three-dimensional road shoulder data are obtained by projecting the three-dimensional point cloud data onto the road shoulder characteristics, and the three-dimensional road shoulder data have three-dimensional depth information, so that the accuracy of road shoulder detection is improved; and the road shoulder characteristic surface is filled based on the three-dimensional road shoulder data to obtain a road shoulder plane, and the road shoulder plane is projected to the ground, so that a road shoulder curve can be accurately obtained, the vehicle is prevented from colliding with the road shoulder, and the control precision of automatic driving can be effectively improved.
Drawings
Fig. 1 is a schematic flow chart of a road shoulder detection method according to an embodiment of the present invention;
FIG. 2 is a schematic plan view of a road shoulder provided by an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a road shoulder detecting device according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it is to be understood that the terms "first", "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless otherwise specified.
In the description of the present application, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
The road shoulder detection method and the road shoulder detection device can be applied to any vehicle, robot or carrying tool which runs on a road, wherein the vehicle can be a common car, a truck, a logistics vehicle, an unmanned vehicle or any other type of vehicle. The road shoulder detection method provided by the embodiment of the invention can accurately detect the road shoulder curve in the road so as to realize auxiliary driving and automatic driving.
Referring to fig. 1-2, in a first embodiment of the present invention, a method for detecting a road shoulder shown in fig. 1 is provided, which includes:
s1, acquiring three-dimensional point cloud data and image data of the target area;
the target area is the surrounding environment of the vehicle, the three-dimensional point cloud data of the target area is obtained through the laser radar, namely the three-dimensional point cloud data is obtained by combining calibration information of the laser radar relative to the position of the vehicle, and the three-dimensional point cloud data comprises three-dimensional space coordinate information of each laser reflection point around the vehicle.
In the embodiment of the invention, the image data of the target area is acquired by the camera installed on the vehicle, the image data can be acquired by a single camera, or after a plurality of pieces of image data are acquired by a plurality of cameras, the images for extracting the road shoulder characteristics are obtained by splicing the plurality of pieces of image data. The camera in the embodiment of the invention can be a forward-looking short-focus camera or a forward-looking long-focus camera, and for example, the forward-looking long-focus camera is adopted to collect image data so as to increase the image shooting distance and improve the comprehensiveness of road images. In addition, the specific area of the target area can be adjusted according to requirements through the installation positions and parameters of the camera and the laser radar.
The device for acquiring the target three-dimensional point cloud data comprises but is not limited to a multi-line laser radar and a single-line laser radar, and any other type of device capable of acquiring the point cloud data on the road surface.
It should be noted that, similar to the radar operating principle, the lidar determines the distance by measuring the time difference and the phase difference of the laser signals, but has the greatest advantage that a sharp 3D image of a target can be created by using the doppler imaging technology. The laser radar analyzes the turn-back time of the laser after encountering a target object by transmitting and receiving laser beams, calculates the relative distance to the target object, quickly obtains a three-dimensional model of the target object and various related data such as lines, surfaces and bodies by utilizing the information such as three-dimensional coordinates, reflectivity, texture and the like of a large number of dense points on the surface of the target object collected in the process, establishes a three-dimensional point cloud picture, and draws an environment map so as to achieve the purpose of environment perception. Since the speed of light is very fast, the flight time can be very short, thus requiring very high accuracy of the measurement device. In effect, the more the lidar dimensions (beam) are, the higher the measurement accuracy is and the higher the safety is.
For example, in the case of using a multiline lidar, the present step S1 may directly generate a plurality of laser scanning lines at different positions on the road surface. For example, in the case of employing a single line laser radar, three-dimensional point cloud data of a plurality of laser scanning lines at a plurality of positions may be acquired by recording single laser scanning line data acquired at different positions during the travel of the vehicle. For another example, in the case of using a camera, a plurality of scan lines may be indirectly generated on image data acquired by the camera, and scan point cloud data on each scan line may be obtained. When the vehicle surrounding environment is scanned through the laser radar, ultrasonic errors possibly caused by interference of external factors, such as errors caused by influence of surrounding light sources on detectors of the laser radar, errors caused by mechanical vibration during scanning of the laser radar and the like, can cause three-dimensional point cloud data obtained by the laser radar to generally comprise hash points and isolated points. The point cloud filtering method of the embodiment of the invention comprises but is not limited to one of bilateral filtering, Gaussian filtering, conditional filtering, straight-through filtering, random sampling consistent filtering and Voxe l Gr id filtering.
S2, extracting road shoulder features in the image data;
the image data contains road shoulder features, the road shoulder features in the image data can be extracted in a mode of directly labeling the road shoulder features, and the road shoulder features in the image data can also be identified in a deep learning mode, so that the road shoulder features are extracted. The road shoulder is a strip-shaped structure part with a certain width, which is positioned from the outer edge of a traffic lane to the edge of a roadbed at two sides of a road, and comprises a road shoulder and a hard road shoulder. The road shoulder features include upper plane, outer plane, intersection line with ground and two plane corner lines, and for road shoulder in special shape, the marked contents are the main feature plane of the road shoulder, the intersection line with ground and the corner lines.
S3, projecting the three-dimensional point cloud data to the road shoulder features to obtain three-dimensional road shoulder data;
in the embodiment of the invention, the road shoulder features are obtained by extracting from the image data, and in order to accurately project the three-dimensional point cloud data on the road shoulder features, the spatial relative relationship between the three-dimensional point cloud data and the image data needs to be obtained first. According to parameters of a camera used for shooting image data, including focal length, the size of each unit, internal parameters of the position of an origin in an image coordinate system, external parameters of rotation angles and translation amounts in three dimensions, the space relative relationship between the image data and the three-dimensional point cloud data is obtained, and the three-dimensional point cloud data is projected onto the road shoulder features obtained through recognition according to the space relative relationship so as to obtain the three-dimensional road shoulder data with three-dimensional depth information.
S4, filling the road shoulder feature surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
in the embodiment of the invention, the road shoulder heights in different three-dimensional road shoulder data are different, the height of each road shoulder is detected, the road shoulder height is compared with a preset height range, and the road shoulders outside the preset height range are filtered out, so that the filtered three-dimensional road shoulder data are obtained. For example, the height range of the road shoulder in the three-dimensional road shoulder data is 0-30 cm, the preset height range is set to be greater than or equal to 10cm, the road shoulder with the height of 0-10 cm in the three-dimensional data is filtered, and the road shoulder with the height greater than or equal to 10cm in the three-dimensional road shoulder data is reserved; the height of the road shoulder in the three-dimensional road shoulder data is 0-40 cm, the preset height range is larger than or equal to 15 cm, the road shoulder with the height of 0-15 cm in the three-dimensional data is filtered, and the road shoulder with the height larger than or equal to 15 cm in the three-dimensional road shoulder data is reserved.
The road shoulder characteristic surface comprises a ground line, an outer side surface and an upper plane, the upper plane to be filled and the outer side surface to be filled are obtained in a modeling mode, and before the modeling process, the three-dimensional road shoulder data are divided according to a certain interval, for example, the three-dimensional road shoulder data are divided according to a distance interval of 3 centimeters, or divided according to a distance interval of 5 centimeters, or divided according to a distance interval of 10 centimeters. And determining whether the ground line needs to be updated according to the position relation between the point cloud obtained after the division and the ground line point cloud, and if the ground line point cloud is closer to the outer side than the point cloud obtained after the division and the ratio of the number of the ground line point clouds to the number of the ground line point clouds compared with the point cloud obtained after the division exceeds a certain threshold, for example, exceeds 10% of the total ground line point cloud number, or exceeds 20% of the total ground line point cloud number, or exceeds 30% of the total ground line point cloud number, taking the filtered point cloud as the ground line in ground projection to update the original ground line.
And during modeling, forming an outer side surface to be filled and an upper plane to be filled according to the ground line, the outer side surface, the corner line and the upper plane, and filling the upper plane to be filled and the outer side surface to be filled to obtain the road shoulder plane.
And S5, projecting the road shoulder plane to the ground to obtain a road shoulder curve.
In the embodiment of the invention, the road shoulder plane is projected on the ground, and the road shoulder curve close to one side of the vehicle is selected in the road shoulder plane by combining the current position of the vehicle, and the road shoulder curve can be applied to driving scenes such as parking at the side of the vehicle, driving at the side close to the vehicle and the like, so that the accuracy of automatic driving control can be effectively improved, the safety problem caused by collision between the vehicle and the road shoulder in the automatic driving process can be avoided, and the safety of automatic driving of the vehicle can be effectively improved.
In one embodiment, the obtaining of the three-dimensional point cloud data of the target area specifically includes:
and acquiring three-dimensional point cloud data of a target area by adopting a radar device and combining calibration information of the radar device relative to the position of the vehicle and positioning information of the vehicle.
It should be noted that, when the vehicle is automatically driven, the default road surface is a plane or a combination of planes (where the slope, the potholes on the road surface, and the ramps on the two vertical road surfaces all belong to different planes). Therefore, the point cloud data of the vehicle travelling direction and two sides of the travelling direction can be obtained by scanning the planes through the laser radar. Wherein, laser radar passes through laser radar fixed bolster and sets up on motor vehicle roof, and in order to obtain better original three-dimensional point cloud data, can install laser radar with the horizontal position of vehicle direction of travel.
In the embodiment of the invention, the laser radar is adopted to obtain the three-dimensional point cloud data of the target area, the laser radar comprises a mechanical laser radar and a solid laser radar, the mechanical laser radar controls the emission angle of laser through a rotating part, and the solid laser radar controls the emission angle of laser through an electronic part. The laser radar and the vehicle are in rigid connection, the relative posture and displacement between the laser radar and the vehicle are fixed, in order to establish the relative coordinate relation between the laser radars, the installation of the laser radars needs to be calibrated, and the data of the laser radars are converted to the vehicle body coordinate from the laser radar coordinate in a unified mode. According to the embodiment of the invention, a vehicle mass center coordinate system, a radar reference coordinate system and a vehicle-mounted laser radar coordinate system are established, the data of the laser radar are converted into reference coordinates, and then the reference coordinates are uniformly converted into a vehicle coordinate system, so that calibration information relative to the position of a vehicle and positioning information of the vehicle are obtained. The calibration of the external installation parameters of the laser radar is usually completed by adopting an isosceles right triangle calibration plate and a square calibration plate, and the installation parameters of the laser radar to be calibrated comprise a pitch angle, a yaw angle and a roll angle of the laser radar.
In one embodiment, the laser radar has external mounting parameters of a pitch angle of 180 degrees, a yaw angle of 0 degrees, a roll angle of 0 degrees, and a mounting height which can be adjusted according to the height of an obstacle in front of a target area, wherein the transverse mounting position is on the positive central axis of the vehicle.
In one embodiment, the method for extracting road shoulder features in image data specifically includes:
correcting the image data according to the parameters of a camera for shooting the image data to obtain corrected image data;
the parameters of the camera include, but are not limited to, pitch angle, yaw angle, and distortion coefficients including, but not limited to, radial distortion coefficients, centrifugal distortion coefficients, and thin-edged radius distortion coefficients. Because the camera can be influenced by self parameters when shooting images, such as unclear image data caused by the change of the vehicle yaw angle, the image data is corrected through the parameters of the camera, and accurate road shoulder features in the corrected images can be obtained in a road shoulder feature labeling mode.
In one embodiment, when the apparatus for capturing image data of a target area is a general camera, the image data is subjected to rectification processing using the following formula:
wherein k1 is the radial distortion coefficient, p1 and p2 are both the centrifugal distortion coefficients, and s1 and s2 are both the thin edge diameter distortion coefficients.
When the apparatus for capturing image data of a target area is a fisheye camera, correction processing is performed on the image data using the following formula:
θd＝θ(1+k1θ2+k2θ4+k3θ6+k4θ8)
wherein k is1、k2、k3And k4All are fish-eye camera distortion coefficients, theta is the post-distortion position, thetadIs the original position.
And acquiring road shoulder features marked in the corrected image data.
Compared with the traditional manual marking mode, the road shoulder feature marking method and device can effectively reduce errors caused by manual marking, thereby improving the marking accuracy of the road shoulder feature and being beneficial to improving the road shoulder detection accuracy. Illustratively, the shoulder features include a specially shaped shoulder and a conventionally shaped shoulder.
Optionally, after the corrected image data is obtained, the embodiment of the present invention may further obtain a road shoulder recognition model by performing deep learning on the road shoulder features, and obtain the road shoulder features in the corrected image data by road shoulder recognition model recognition.
In the embodiment of the invention, the point cloud of the acquired road shoulder features may contain more noise, and in order to further improve the accuracy of road shoulder detection, denoising processing is performed after the road shoulder features are acquired.
In one embodiment, the three-dimensional point cloud data is projected onto the road shoulder features to obtain three-dimensional road shoulder data, specifically:
and acquiring a spatial relative relationship between the image data and the three-dimensional point cloud data according to the parameters of the camera, and projecting the three-dimensional point cloud data onto the road shoulder features according to the spatial relative relationship to obtain the three-dimensional road shoulder data.
In the embodiment of the invention, the parameters of the camera comprise internal parameters of a focal length, the size of each unit of the camera, the position of the origin of the camera in an image coordinate system, and external parameters of a rotation angle and a translation amount of the camera in three dimensions, and the following formulas are adopted for coordinate conversion according to the parameters of the camera:
and zc is a coordinate in the z direction of a camera coordinate system, u and v are pixel coordinates, dX and dY are single pixel lengths, u0 and v0 are pixel coordinates of an image origin, f is a focal length, R is a rotation matrix, t is a translation vector, and x, y and z are world coordinates.
In one embodiment, the step of filling the road shoulder feature plane in the three-dimensional road shoulder data to obtain a road shoulder plane specifically includes:
determining a ground line in the three-dimensional road shoulder data according to the position relationship between the point cloud in the three-dimensional road shoulder data and the ground point cloud;
in the embodiment of the invention, the three-dimensional shoulder data has three-dimensional depth information of the shoulder, wherein the three-dimensional shoulder data comprises height information of the shoulder, the height information of all shoulders in the target area can be obtained from the three-dimensional shoulder data, and the height of the shoulder refers to the height of the shoulder above the ground. Laser beams emitted by laser radars installed on vehicles mostly irradiate the ground, that is, most of point clouds collected by the laser radars belong to the ground. Therefore, the point cloud of the three-dimensional road shoulder data is directly used as the point cloud for estimating the ground line, the point cloud with the road shoulder height lower than the preset value in the three-dimensional road shoulder data is filtered, and the ground line is determined subsequently.
And forming an outer side surface to be filled and an upper plane to be filled according to the ground line, the outer side surface, the corner line and the upper plane, and filling the upper plane to be filled and the outer side surface to be filled to obtain the road shoulder plane.
In the embodiment of the invention, after point clouds in the three-dimensional road shoulder data are partitioned according to preset intervals, average position points of the densely distributed point clouds are used as upper plane surface points, corner lines are used as upper plane surface lines, and upper planes to be filled are formed by the upper plane surface points and the upper plane surface lines; and simultaneously projecting the corner lines and the ground lines on the ground, taking lines formed by outer side points in a projection image as outer side surface lines, taking outer side points of densely distributed point clouds as outer side surface points, and forming outer sides to be filled by the outer side surface points and the outer side surface lines.
In one embodiment, the determining the ground line in the three-dimensional road shoulder data according to the position relationship between the point cloud in the three-dimensional road shoulder data and the ground point cloud specifically includes:
filtering the three-dimensional road shoulder data corresponding to the road shoulder which is not in the preset height range in the three-dimensional road shoulder data;
it should be noted that, the common road shoulder has road shoulder stones, green belts, isolation fences, etc. in addition, in the road environment, the common obstacles include street lamps, road trees, fire hydrants, garbage cans, etc. Therefore, when detecting and identifying road shoulders, people need to find objects such as sidewalks, green belts and isolation barriers. The height of objects in the road environment can be roughly divided into three levels, objects such as street lamps, street trees and the like are classified into high height, objects such as green belts, isolation fences, fire hydrants and the like are classified into medium height, and the height of a curb beside a sidewalk is classified into low height. The height of the road shoulder is thus taken as a characteristic of the screening of the road shoulder.
In an embodiment of the present invention, the preset height range includes a minimum height threshold and a maximum height threshold, wherein the minimum height threshold is greater than or equal to 0. For example, the height range of the road shoulder in the three-dimensional road shoulder data is 0-30 cm, the preset height range is set to be greater than or equal to 10cm, the road shoulder with the height of 0-10 cm in the three-dimensional data is filtered, and the road shoulder with the height greater than or equal to 10cm in the three-dimensional road shoulder data is reserved; the height of the road shoulder in the three-dimensional road shoulder data is 0-40 cm, the preset height range is larger than or equal to 15 cm, the road shoulder with the height of 0-15 cm in the three-dimensional data is filtered, and the road shoulder with the height larger than or equal to 15 cm in the three-dimensional road shoulder data is reserved.
Dividing the filtered three-dimensional road shoulder data according to a preset distance interval to obtain a plurality of point clouds corresponding to the three-dimensional road shoulder data;
and if the ground line point clouds are positioned at the outer sides of the outmost point clouds in the point clouds, and the number ratio of the ground line point clouds exceeding the outmost point clouds exceeds a preset threshold value, taking the projections of the point clouds on the ground as ground lines.
In the embodiment of the invention, the point cloud position corresponding to the three-dimensional road shoulder data is compared with the point cloud position of the ground line, and the projection of the point cloud corresponding to the three-dimensional road shoulder data on the ground is taken as a new ground line under the condition of meeting the preset condition.
In one embodiment, an outer side surface to be filled and an upper plane to be filled are formed according to a ground line, an outer side surface, a corner line and the upper plane, and the upper plane to be filled and the outer side surface to be filled are filled to obtain a road shoulder plane, specifically:
taking the average position points of the densely distributed point clouds as upper plane surface points, taking the corner lines as upper plane surface lines, and forming an upper plane to be filled by the upper plane surface points and the upper plane surface lines;
in a preferred mode of the embodiment of the present invention, the degree of density of point cloud distribution is first determined, specifically, the number of point clouds in each block is counted, blocks in which the number of point clouds exceeds a preset number are taken as dense blocks, and an average position of the point clouds is obtained in the dense blocks as an upper plane point. For example, histogram statistics are used to count the density of the point cloud distribution. For another example, the point cloud position closest to the center point of the dense block is taken as the average position. For another example, a minimum circle that can contain all point clouds is drawn in the dense partition, and the point cloud position with the nearest center is taken as the average position.
Simultaneously projecting corner lines and ground lines on the ground, taking lines formed by outer side points in a projection image as outer side surface lines, taking outer side points of densely distributed point clouds as outer side surface points, and forming outer sides to be filled by the outer side surface points and the outer side surface lines;
and carrying out road shoulder filling on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane.
In the embodiment of the invention, in order to provide the accuracy of road shoulder detection, modeling is carried out through a ground line, an outer side face and a corner line so as to construct an upper plane to be filled and an outer side face to be filled.
Fig. 2 is a schematic plan view of a road shoulder according to an embodiment of the present invention. The side of the shoulder plane close to the road surface is the outer side, that is, the side close to the vehicle is the outer side of the shoulder when the vehicle runs on the road surface.
In one embodiment, the shoulder filling is performed on the upper plane to be filled and the outer side surface to be filled to obtain a shoulder plane, specifically:
taking the highest point of the plane of the upper plane to be filled as a cut-off point, and filling the road shoulder upwards in a direction parallel to the ground to obtain a filled upper plane;
in the embodiment of the present invention, other points of the upper plane to be filled may also be selected as cut-off points, for example, a position 1 cm lower than the highest point of the upper plane to be filled is used as a cut-off point, and shoulder filling is performed upward in a direction parallel to the ground; for example, taking the position 3 cm lower than the highest point of the plane of the upper plane to be filled as a cut-off point, and filling the road shoulder upwards in the direction parallel to the ground; for example, shoulder filling is performed upward in a direction parallel to the ground, with a position 5 cm below the highest point of the plane of the upper plane to be filled as a cut-off point.
Carrying out road shoulder filling on the outer side surface to be filled in a direction perpendicular to the normal direction of the ground and perpendicular to the connecting line direction of the two last points, wherein the filling amount does not exceed a preset threshold value, and obtaining a filling outer side surface; or taking the outermost point as a cut-off point, and filling the road shoulder to the outside in the direction vertical to the normal direction of the ground and the direction vertical to the lane line to obtain a filled outer side surface;
in the embodiment of the invention, the filling preset threshold value can be set to be 0-10 cm, and the difference value between the outermost point and the innermost point can also be used as the filling threshold value;
in the embodiment of the present invention, other points of the outer side surface to be filled may also be selected as the cut-off point for filling, for example, the position of 1 cm from the outermost point of the outer side surface to be filled is taken as the cut-off point for filling, and shoulder filling is performed in the direction perpendicular to the ground in the direction perpendicular to the lane line of the road; for example, taking the position of 3 cm at the outermost point of the outer side surface to be filled as a cut-off point of filling, and filling the shoulder in the direction vertical to the ground to the direction vertical to the road lane line; for example, shoulder filling is performed in a direction perpendicular to the ground surface and in a direction perpendicular to the road lane line, with a position 5 cm from the outermost point of the outer side surface to be filled as a cut-off point of filling.
In the embodiment of the invention, the lane line data can be directly obtained by adopting a map, can also be obtained by identifying the image data acquired by the camera, and can be used for detecting the lane line by using a laser radar. For example, the lane lines are detected based on the density of radar scanning points, specifically, the coordinates of the radar scanning points are obtained and converted into a grid map, the grid map is mapped by using original data, the grid map can be a direct coordinate grid map or a polar coordinate grid map, the selection is performed according to the requirement of post-processing, the polar coordinate grid map is directly used for lane line identification, namely, grids mapped by a plurality of points are considered as the lane line points. The method extracts the lane lines by using the density of the points in the grid map, and the point density can be calculated by adopting a histogram statistical method, so that the point density can be quickly and intuitively calculated by the histogram statistical method. For example, the laser radar is used to acquire different characteristics of shoulder height information and physical reflection information, detect the shoulder information, and calculate the lane line position from the distance between the shoulder and the lane line in combination with the known road width.
And forming a road shoulder plane according to the filling upper plane and the filling outer side surface.
In the embodiment of the invention, the upper filling plane and the outer filling side surface are spliced in a plane splicing manner to form the road shoulder plane.
In the embodiment, the road shoulder features in the image data are extracted by acquiring the three-dimensional point cloud data and the image data of the target area, and the three-dimensional road shoulder data are obtained by projecting the three-dimensional point cloud data onto the road shoulder features, wherein the three-dimensional road shoulder data have three-dimensional depth information, so that the accuracy of road shoulder detection is improved; and the road shoulder characteristic surface is filled based on the three-dimensional road shoulder data to obtain a road shoulder plane, and the road shoulder plane is projected to the ground, so that a road shoulder curve can be accurately obtained, the vehicle is prevented from colliding with the road shoulder, and the control precision of automatic driving can be effectively improved.
Based on the same inventive concept, an embodiment of the present invention further provides a road shoulder detecting device shown in fig. 3, including:
the data acquisition module 10 is used for acquiring three-dimensional point cloud data and image data of a target area;
the feature extraction module 20 is configured to extract road shoulder features in the image data;
the first projection module 30 is used for projecting the three-dimensional point cloud data onto the road shoulder features to obtain three-dimensional road shoulder data;
the data filling module 40 is used for filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
and the second projection module 50 is used for projecting the road shoulder plane to the ground to obtain a road shoulder curve.
In one embodiment, the obtaining of the three-dimensional point cloud data of the target area specifically includes:
and acquiring three-dimensional point cloud data of a target area by adopting a radar device and combining calibration information of the radar device relative to the position of the vehicle and positioning information of the vehicle.
In one embodiment, feature extraction module 20 is configured to:
correcting the image data according to the parameters of a camera for shooting the image data to obtain corrected image data;
and acquiring road shoulder features marked in the corrected image data.
In one embodiment, the feature extraction module 20 is further configured to:
and constructing a road shoulder extraction model by adopting a deep learning method, and extracting road shoulder features in the image data according to the road shoulder extraction model.
In one embodiment, a first projection module 30 for:
and acquiring a spatial relative relationship between the image data and the three-dimensional point cloud data according to the parameters of the camera, and projecting the three-dimensional point cloud data onto the road shoulder features according to the spatial relative relationship to obtain the three-dimensional road shoulder data.
In one embodiment, data population module 40 is configured to:
determining a ground line in the three-dimensional road shoulder data according to the position relationship between the point cloud in the three-dimensional road shoulder data and the ground point cloud;
and forming an outer side surface to be filled and an upper plane to be filled according to the ground line, the outer side surface, the corner line and the upper plane, and filling the upper plane to be filled and the outer side surface to be filled to obtain the road shoulder plane.
In one embodiment, the data population module 40 is further configured to:
filtering the three-dimensional road shoulder data corresponding to the road shoulder which is not in the preset height range in the three-dimensional road shoulder data;
dividing the filtered three-dimensional road shoulder data according to a preset distance interval to obtain a plurality of point clouds corresponding to the three-dimensional road shoulder data;
and if the ground line point clouds are positioned at the outer sides of the outmost point clouds in the point clouds, and the number ratio of the ground line point clouds exceeding the outmost point clouds exceeds a preset threshold value, taking the projections of the point clouds on the ground as ground lines.
In one embodiment, the data population module 40 is further configured to:
taking the average position points of the densely distributed point clouds as upper plane surface points, taking the corner lines as upper plane surface lines, and forming an upper plane to be filled by the upper plane surface points and the upper plane surface lines;
simultaneously projecting corner lines and ground lines on the ground, taking lines formed by outer side points in a projection image as outer side surface lines, taking outer side points of densely distributed point clouds as outer side surface points, and forming outer sides to be filled by the outer side surface points and the outer side surface lines;
and carrying out road shoulder filling on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane.
In one embodiment, the data population module 40 is further configured to:
taking the highest point of the plane of the upper plane to be filled as a cut-off point, and filling the road shoulder upwards in a direction parallel to the ground to obtain a filled upper plane;
carrying out road shoulder filling on the outer side surface to be filled in a direction perpendicular to the normal direction of the ground and perpendicular to the connecting line of the two outermost points, wherein the filling amount does not exceed a preset threshold value, and obtaining a filling outer side surface; or taking the outermost point as a cut-off point, and filling the road shoulder to the outside in the direction vertical to the normal direction of the ground and the direction vertical to the lane line to obtain a filled outer side surface;
and forming a road shoulder plane according to the filling upper plane and the filling outer side surface.
In one embodiment, a computer device is provided, as shown in fig. 4, comprising a processor, a memory, a network interface, and a sensor connected by a communication line. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program which, when executed by the processor, causes the processor to implement the shoulder detection method. The internal memory may also have stored therein a computer program that, when executed by the processor, causes the processor to perform a shoulder detection method. Those skilled in the art will appreciate that the configuration shown in fig. 4 is a block diagram of only a portion of the configuration associated with the present application and does not constitute a limitation on the computing device to which the present application may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, the road shoulder detecting apparatus provided by the present application may be implemented in the form of a computer program, which is executable on a computer device as shown in fig. 4. The memory of the computer device may store various program modules constituting the shoulder detecting device. The respective program modules constitute computer programs that cause the processor to execute the steps in the road shoulder detection method of the respective embodiments of the present application described in the present specification.
In some possible embodiments, aspects of a road shoulder detection method provided by the present application may also be implemented in the form of a program product including program code for causing a terminal device to perform the steps of a road shoulder detection method according to various exemplary embodiments of the present application described above in this specification when the program product is run on the terminal device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
A shoulder detection program product of an embodiment of the present application may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be executable on a computing device. However, the program product of the present application is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including a physical programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device over any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., over the internet using an internet service provider).
It should be noted that although several units or sub-units of the apparatus are mentioned in the above detailed description, such division is merely exemplary and not mandatory. Indeed, the features and functions of two or more units described above may be embodied in one unit, according to embodiments of the application. Conversely, the features and functions of one unit described above may be further divided into embodiments by a plurality of units.
Further, while the operations of the methods of the present application are depicted in the drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable shoulder detection apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable shoulder detection apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable shoulder detection apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable shoulder detection apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should be noted that the above-described device embodiments are merely illustrative, and units illustrated as separate components may or may not be physically separate, and components illustrated as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
The foregoing is a preferred embodiment of the present invention, and it should be noted that it would be apparent to those skilled in the art that various modifications and enhancements can be made without departing from the principles of the invention, and such modifications and enhancements are also considered to be within the scope of the invention.
Claims (11)
1. A road shoulder detection method is characterized by comprising the following steps:
acquiring three-dimensional point cloud data and image data of a target area;
extracting road shoulder features in the image data;
projecting the three-dimensional point cloud data to the road shoulder features to obtain three-dimensional road shoulder data;
filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
and projecting the road shoulder plane to the ground to obtain a road shoulder curve.
2. The road shoulder detection method according to claim 1, wherein the acquiring three-dimensional point cloud data of the target area specifically comprises:
and acquiring the three-dimensional point cloud data of the target area by adopting a radar device and combining the calibration information of the radar device relative to the position of the vehicle and the positioning information of the vehicle.
3. The road shoulder detection method according to claim 1, wherein the extracting of the road shoulder features in the image data specifically includes:
correcting the image data according to parameters of a camera for shooting the image data to obtain corrected image data;
and acquiring road shoulder features marked in the corrected image data.
4. The road shoulder detection method according to claim 1, wherein the extracting of the road shoulder features in the image data specifically includes:
and constructing a road shoulder extraction model by adopting a deep learning method, and extracting road shoulder features in the image data according to the road shoulder extraction model.
5. The road shoulder detection method according to claim 1, wherein the three-dimensional point cloud data is projected onto the road shoulder feature to obtain three-dimensional road shoulder data, specifically:
and acquiring a spatial relative relationship between the image data and the three-dimensional point cloud data according to parameters of a camera, and projecting the three-dimensional point cloud data onto the road shoulder features according to the spatial relative relationship to obtain three-dimensional road shoulder data.
6. The road shoulder detection method according to claim 1, wherein the filling processing is performed on the road shoulder feature plane in the three-dimensional road shoulder data to obtain a road shoulder plane, specifically:
determining a ground line in the three-dimensional road shoulder data according to the position relation between the point cloud in the three-dimensional road shoulder data and the ground point cloud;
and forming an outer side surface to be filled and an upper plane to be filled according to the ground line, the outer side surface, the corner line and the upper plane, and filling the upper plane to be filled and the outer side surface to be filled to obtain the road shoulder plane.
7. The method according to claim 1, wherein the determining a ground line in the three-dimensional road shoulder data according to a position relationship between a point cloud in the three-dimensional road shoulder data and a ground point cloud comprises:
filtering the three-dimensional road shoulder data corresponding to the road shoulder which is not in the preset height range in the three-dimensional road shoulder data;
dividing the filtered three-dimensional road shoulder data according to a preset distance interval to obtain a plurality of point clouds corresponding to the three-dimensional road shoulder data;
and if the ground line point clouds are positioned at the outer sides of the outmost point clouds in the point clouds, and the number ratio of the ground line point clouds exceeding the outmost point clouds exceeds a preset threshold value, taking the projection of the point clouds on the ground as a ground line.
8. The method for detecting a road shoulder according to claim 6, wherein the outside surface to be filled and the top plane to be filled are formed according to the ground line, the outside surface, the corner line and the top plane, and the top plane to be filled and the outside surface to be filled are filled to obtain a road shoulder plane, specifically:
taking the average position points of the densely distributed point clouds as upper plane surface points, taking the corner lines as upper plane surface lines, and forming an upper plane to be filled by using the upper plane surface points and the upper plane surface lines;
projecting the corner lines and the ground lines on the ground at the same time, taking lines formed by outer side points in a projection image as outer side surface lines, taking outer side points of densely distributed point clouds as outer side surface points, and forming outer sides to be filled by the outer side surface points and the outer side surface lines;
and carrying out road shoulder filling on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane.
9. The road shoulder detection method according to claim 8, wherein the road shoulder filling is performed on the upper plane to be filled and the outer side surface to be filled to obtain a road shoulder plane, specifically:
taking the highest point of the plane of the upper plane to be filled as a cut-off point, and filling the road shoulder upwards in a direction parallel to the ground to obtain a filled upper plane;
carrying out road shoulder filling on the outer side surface to be filled in a direction perpendicular to the normal direction of the ground and perpendicular to the connecting line of the two outermost points, wherein the filling amount does not exceed a preset threshold value, and obtaining a filling outer side surface; or taking the outermost point as a cut-off point, and filling the road shoulder to the outside in the direction vertical to the normal direction of the ground and the direction vertical to the lane line to obtain a filled outer side surface;
and forming a road shoulder plane according to the filling upper plane and the filling outer side surface.
10. A road shoulder detection device, characterized by comprising:
the data acquisition module is used for acquiring three-dimensional point cloud data and image data of a target area;
the feature extraction module is used for extracting road shoulder features in the image data;
the first projection module is used for projecting the three-dimensional point cloud data onto the road shoulder features to obtain three-dimensional road shoulder data;
the data filling module is used for filling the road shoulder characteristic surface in the three-dimensional road shoulder data to obtain a road shoulder plane;
and the second projection module is used for projecting the road shoulder plane to the ground to obtain a road shoulder curve.
11. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the road shoulder detection method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111016619.8A CN113869129A (en) | 2021-08-31 | 2021-08-31 | Road shoulder detection method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111016619.8A CN113869129A (en) | 2021-08-31 | 2021-08-31 | Road shoulder detection method and device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113869129A true CN113869129A (en) | 2021-12-31 |
Family
ID=78989004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111016619.8A Pending CN113869129A (en) | 2021-08-31 | 2021-08-31 | Road shoulder detection method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113869129A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110063097A1 (en) * | 2008-09-17 | 2011-03-17 | Hitachi Automotive Systems, Ltd | Device for Detecting/Judging Road Boundary |
| CN107066937A (en) * | 2016-01-28 | 2017-08-18 | 罗伯特·博世有限公司 | The apparatus and method of the curb stone in surrounding environment for detecting vehicle and for vehicle curb stone control system |
| CN111191600A (en) * | 2019-12-30 | 2020-05-22 | 深圳元戎启行科技有限公司 | Obstacle detection method, obstacle detection device, computer device, and storage medium |
-
2021
- 2021-08-31 CN CN202111016619.8A patent/CN113869129A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110063097A1 (en) * | 2008-09-17 | 2011-03-17 | Hitachi Automotive Systems, Ltd | Device for Detecting/Judging Road Boundary |
| CN107066937A (en) * | 2016-01-28 | 2017-08-18 | 罗伯特·博世有限公司 | The apparatus and method of the curb stone in surrounding environment for detecting vehicle and for vehicle curb stone control system |
| CN111191600A (en) * | 2019-12-30 | 2020-05-22 | 深圳元戎启行科技有限公司 | Obstacle detection method, obstacle detection device, computer device, and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| SELMA EVELYN CATALINA GOGA ET AL.: "Fusing semantic labeled camera images and 3D LiDAR data for the detection of urban curbs", 《2018 IEEE 14TH INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTER COMMUNICATION AND PROCESSING (ICCP)》, 1 November 2018 (2018-11-01), pages 301 - 308 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021226776A1 (en) | Vehicle drivable area detection method, system, and automatic driving vehicle using system | |
| CN111352112B (en) | Target detection method based on vision, laser radar and millimeter wave radar | |
| CN110147706B (en) | Obstacle recognition method and device, storage medium, and electronic device | |
| CN112347999B (en) | Obstacle recognition model training method, obstacle recognition method, device and system | |
| CN112017251A (en) | Calibration method and device, road side equipment and computer readable storage medium | |
| CN112740225B (en) | Method and device for determining road surface elements | |
| EP4160346A1 (en) | Method and device for identifying vehicle lane changing tendency | |
| CN114677446B (en) | Vehicle detection method, device and medium based on road side multi-sensor fusion | |
| US11798289B2 (en) | Streaming object detection and segmentation with polar pillars | |
| CN113743171A (en) | Target detection method and device | |
| WO2022047744A1 (en) | Road surface extraction method and device for map | |
| Lim et al. | Implementation of semantic segmentation for road and lane detection on an autonomous ground vehicle with LIDAR | |
| CN114550142A (en) | Parking space detection method based on fusion of 4D millimeter wave radar and image recognition | |
| CN115424263A (en) | Traffic light labeling method, vehicle-mounted device and storage medium | |
| CN114913399A (en) | Vehicle track optimization method and intelligent traffic system | |
| US20230177724A1 (en) | Vehicle to infrastructure extrinsic calibration system and method | |
| CN114612869A (en) | Information fusion method for road side laser radar and vehicle-mounted laser radar | |
| CN113076824B (en) | Parking space acquisition method and device, vehicle-mounted terminal and storage medium | |
| CN113988197A (en) | Multi-camera and multi-laser radar based combined calibration and target fusion detection method | |
| US20220371606A1 (en) | Streaming object detection and segmentation with polar pillars | |
| CN113869129A (en) | Road shoulder detection method and device and storage medium | |
| CN115359332A (en) | Data fusion method and device based on vehicle-road cooperation, electronic equipment and system | |
| Oniga et al. | A fast ransac based approach for computing the orientation of obstacles in traffic scenes | |
| US11288520B2 (en) | Systems and methods to aggregate and distribute dynamic information of crowdsourcing vehicles for edge-assisted live map service | |
| US20220017095A1 (en) | Vehicle-based data acquisition |
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 |