CN106228110B - A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera - Google Patents
A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera Download PDFInfo
- Publication number
- CN106228110B CN106228110B CN201610538582.8A CN201610538582A CN106228110B CN 106228110 B CN106228110 B CN 106228110B CN 201610538582 A CN201610538582 A CN 201610538582A CN 106228110 B CN106228110 B CN 106228110B
- Authority
- CN
- China
- Prior art keywords
- barrier
- image
- road
- pixel
- parallax
- 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.)
- Active
Links
Classifications
-
- 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
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera, comprising the following steps: image S1, is shot by binocular camera；S2, image is pre-processed；S3, the final matching cost for calculating each pixel in image；S4, v- disparity map is calculated by final matching cost；S5, horizon and road surface range are determined using v- disparity map；S6, within the scope of road surface, barrier-road intersection is calculated；S7, current lane and outermost lane range are determined；S8, in barrier-road intersection above section, obstacle height is calculated；S9, the image-region for belonging to same barrier is determined；S10, output road region information and obstacle information.This programme can avoid unnecessary binocular parallax figure from calculating the real-time that algorithm is substantially improved under the premise of not sacrificing parallax precision, be suitable for driving the fields such as early warning, automatic Pilot.
Description
Technical field
The present invention relates to road image identifying processing field, more particularly, to it is a kind of with high accuracy based on vehicle-mounted
The barrier and drivable region detection method of binocular camera.
Background technique
The acquisition and analysis of vehicle periphery traffic information are to drive the important evidence of early warning, auxiliary driving and automatic Pilot.
Mainly based on monocular camera, part system is assisted with distance measuring sensors such as millimetre-wave radars existing relevant sensing technology.It is existing
There is system that can measure at a distance from front vehicles under certain good roads and operating condition, but there are still certain to lack
It falls into:
1. the barrier ranging based on monocular vision: such method passes through the camera heights demarcated mostly and pitch angle is believed
Breath, based on surface road it is assumed that the boundary image feature of Use barriers object and road intersection, calculates the distance letter of the barrier
Breath.This method is suitable for that light condition is good, the gentle unobstructed detection of obstacles of road, for road grade (climb and fall) with
And influence caused by shade can generate larger measurement error or failure；
2. the Lane detection based on monocular vision: such method is mostly based on surface road it is assumed that for road grade
And estimation effect is poor in the case of turning；
3. the barrier ranging based on binocular vision: such method is mostly based on binocular parallax, is calculated using parallax information
Barrier size and distance.The efficiency of algorithm of accurate disparity computation is lower, though Part Methods real-time is fine, leave for it is subsequent its
He is relatively limited the operation time of real-time application system, and such algorithm for barrier scanning range there is no limit, effect
Rate is not high and invalid information is more；
4. the barrier range-measurement system merged based on millimetre-wave radar with monocular camera: such method utilizes millimetre-wave radar
Provided range information determines the location information of barrier, passes through the projection of the real coordinate system and image coordinate system demarcated
Relationship, determines the location information of barrier, and such method is poor for the robustness of road grade.
Summary of the invention
The present invention mainly solve the prior art present in it is higher to environmental requirement, adaptability is weak, precision is more low
The technical issues of, provide it is a kind of can be adapted for more complicated environment, with degree of precision and robustness based on vehicle-mounted double
The barrier and drivable region detection method of mesh camera.
What the present invention was mainly addressed by following technical proposals in view of the above technical problems: one kind is based on vehicle-mounted double
The barrier and drivable region detection method of mesh camera, comprising the following steps:
S1, the image that rgb format is shot by binocular camera, obtained image are binocular image, including left figure and the right side
Figure；
S2, image is pre-processed；
S3, the final matching cost for calculating each pixel in image；In image coordinate system, u is that the horizontal axis of pixel is sat
Mark, v are the ordinate of orthogonal axes of pixel；
S4, the corresponding v- disparity map of longitudinal road plane is calculated by final matching cost；
S5, horizon and road surface range are determined using v- disparity map；
S6, within the scope of road surface, barrier-road intersection is calculated；
S7, the part below barrier-road intersection carry out lane detection, carry out secondary treatment to image, extract suddenly
Husband's straight line determines current lane and outermost lane range；
S8, in barrier-road intersection above section, obstacle height is calculated；
The threshold module of S9, place obstacles object width, height and depth filter barrier region adjacent in image, really
Surely belong to the image-region of same barrier；
S10, output are by the road region information and obstacle information after threshold filtering.
Preferably, including gray processing, except distortion and three-dimensional correction to the pretreatment that image carries out in the step S2.
Preferably, calculating the final matching cost of each pixel in image in the step S3 specifically:
S301, parallax d corresponding to pixel each in pretreated picture calculate the difference for being based on gray value absolute value
Matching cost Cv(ui,vi,di), calculation method is as follows:
Cv(ui,vi,di)=imgleft(ui,vi)-imgright(ui-di,vi)
Wherein, uiFor the abscissa of pixel i under image coordinate system, viFor the ordinate of pixel i under image coordinate system,
imgleft(ui,vi) it is gray value absolute value of the pixel i in left figure, imgright(ui-di,vi) be coordinate be (ui-di,vi)
Gray value absolute value of the pixel in right figure, diFor the corresponding parallax of pixel i, range 0-d, d are preset value, d and double
The correlations such as specification, setting position, the shooting angle of mesh camera；
S302, the sliding window convolution that window is n × n is carried out to the matching cost of the difference calculated based on gray value absolute value
Filtering, obtains final matching cost, and wherein n is parameter preset.
Preferably, the step S4 specifically:
S401, it sums to projection of the final disparity correspondence cost on the image coordinate system longitudinal axis (v axis), and calculates figure
As every a line longitudinal axis corresponds to the minimum value C of the sum of parallax costv,min:
Cv,min=min (Cv(vi,di))
It finds out every row and corresponds to the sum of parallax cost less than Cv,min+TdCorresponding parallax d, to obtain initial v- parallax
Figure；Parallax cost threshold value TdFor preset value；
S402, initial v- disparity map abscissa be vx, ordinate vy, by initial v- disparity map projection to real coordinate
It is the mapping of height y Yu depth x, calculation formula is as follows:
stereofIt is defined as binocular camera focal length；
stereov0It is defined as binocular image center；
stereobaselineIt is defined as binocular camera baseline length；
v0It is defined as v- parallax curve and y-axis focus, i.e. horizon ordinate；
Using B- spline curve fitting pavement-height and depth relationship, v- parallax plan is returned in last inverse mapping, that is, obtains
The corresponding v- disparity map of longitudinal road plane.
Preferably, in step S5, horizon v0The curve that the image-region that the time difference is 0 as in v- disparity map is constituted；
Image-region of the parallax greater than 0 is road surface range.
Preferably, step S6 specifically:
S601, within the scope of road surface, using the biaxial stress structure relationship of a line v every in v- disparity map and corresponding road surface parallax d,
Calculate barrier-road intersection matching cost CBoundary；The barrier-road intersection matching cost is by path adaptation cost and object
Body matching cost two parts composition, wherein path adaptation cost v and d meet v- disparity map mapping relations and object matches cost
Every a line then corresponds to identical parallax d；Its specific formula for calculation is as follows:
uiIt is defined as the abscissa of pixel i under image coordinate system；
viIt is defined as the ordinate of pixel i under image coordinate system；
diIt is defined as the parallax of pixel i；
f(vi) it is defined as parallax d in road surface in v- disparity mapiWith viMapping relations
S602, barrier-road intersection matching cost C is determined using 2 dimension dynamic programming methodsBoundaryCorresponding to minimum value
Sets of pixel values be barrier and road intersection, the corresponding parallax value of each column u is dBoundary(u)。
Preferably, the step S8 specifically:
S801, in barrier-road intersection above section, calculate obstacle height matching cost CHeight: first by general
Rate function m (u, v) calculates final matching cost Cm(u,v,dBoundary(u)) be local extremum a possibility that, the value is between -1 and 1
Between；Finally, obstacle height matching cost calculation formula is as follows:
m(ui, v) and more than road intersection certain pixel (u is defined as in image coordinate systemi, v) and road contact barrier
Probability；
vbotIt is defined as barrier in image coordinate system-road intersection and corresponds to ordinate；
S802, obstacle height matching cost C is determined using 2 dimension dynamic programming methodsHeightPixel corresponding to minimum value
It is worth (ui,vi) set be and barrier obstacle height information corresponding with the intersection of road.
The present invention is based on the calculating of disparity correspondence cost to solve corresponding global optimization problem by two-dimension dynamic programming
Realize the division of real-time traffic scene.The algorithm can avoid unnecessary binocular under the premise of not sacrificing parallax precision
Disparity map calculates the real-time that algorithm is substantially improved.
The output of this system is the location information in the picture of road area and barrier region, can be very easily
The location information being converted into real coordinate system.At the same time, the location information of barrier can be used as obstacle recognition algorithm
Input interface promote differentiating obstacle to greatly reduce the sliding window detection range of existing obstacle recognition algorithm
Real-time.
This method can be integrated with the hardware integration of binocular camera, make the output of vehicle-mounted biocular systems can traffic areas position
Confidence breath and potential barrier position and range information, the algorithm can also easily with other sensor-based system (such as millimeter waves
Radar sensor) fusion, it is that the intelligent algorithm of semi-automatic/autonomous driving vehicle is developed, the basis of environment sensing is provided.
The height and pitch angle information of this method real-time update biocular systems, and real-time estimation longitudinal direction road model, can
To eliminate influence of the change in road slope to range accuracy.In addition, binocular parallax information is stronger to the robustness of shadows on the road, it can
To determine barrier and road intersection in the presence of shade.
Bring substantial effect of the present invention is to be calculated using disparity correspondence cost instead of disparity map, calculation greatly improved
Method efficiency, by can traffic areas and barrier region division, optimizing detection range realized by a kind of new method
Can traffic areas detection with detection of obstacles and ranging, applicable working condition is more extensive, and efficiency of algorithm is higher, and versatility is stronger.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the invention.
Specific embodiment
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.
Embodiment: a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera of the present embodiment, with
Binocular camera image is input, can traffic areas and potential barrier range be output, as shown in Figure 1, comprising the following steps:
1. obtaining rgb format binocular image.
It mainly include gray processing 2. a pair image pre-processes, except distortion and three-dimensional correction.
3. the corresponding parallax d of each pixel (u, v) in pair pretreated picture, calculate based on gray value absolute value it
The matching cost C of poor (SAD)v(ui,vi,di), calculation method is as follows:
Cv(ui,vi,di)=imgleft(ui,vi)-imgright(ui-di,vi)
Wherein, uiFor the abscissa of pixel i under image coordinate system, viFor the ordinate of pixel i under image coordinate system,
imgleft(ui,vi) it is gray value absolute value of the pixel i in left figure, imgright(ui-di,vi) be coordinate be (ui-di,vi)
Gray value absolute value of the pixel in right figure, diFor the corresponding parallax of pixel i；The range of wherein u, v, d are that can set
Determine parameter.The calculating of binocular parallax matching cost can also be using based on the difference of two squares and (SSD) or other calculation methods.
4. pair matching cost calculated carries out the sliding window convolutional filtering of n × n, final matching cost C is obtainedm(u, v, d),
Wherein n be can setup parameter.
It sums 5. a pair final disparity correspondence cost is projected to the image longitudinal axis (v axis), and calculates the every a line v of image and correspond to parallax
The minimum value C of the sum of costv,min:
Cv,min=min (Cv(vi,di))
By setting parallax cost threshold value Td, find out every row and correspond to the sum of parallax cost less than Cv,min+TdCorresponding view
Difference d, to obtain v- disparity map (v-d mapping).
6. v- disparity map projection to the mapping of real coordinate system height and depth is utilized B- spline curve fitting road surface height
Degree and depth relationship, last inverse mapping return v- parallax plan, can be obtained the corresponding v- disparity map of longitudinal road plane.In addition to
B- spline curve can also use the spline curve of other forms, such as segmented linear or single straight line.
7. utilizing v- disparity map, horizon v is determinedo(parallax d=0) and road surface range (the corresponding image district in parallax d > 0
Domain).
8. utilizing the biaxial stress structure of a line v every in v- disparity map and corresponding road surface parallax d within the scope of the road surface that 7 determine
Relationship calculates barrier-road intersection matching cost CBoundary.The barrier-road intersection matching cost is by path adaptation generation
Valence and object matches cost two parts form, wherein path adaptation cost v and d meet v- disparity map mapping relations (f:v<->
D) and the every a line of object matches cost then corresponds to identical parallax d.Its specific formula for calculation is as follows:
Barrier-road intersection matching cost can also be taken using road surface matching cost (ignoring object matches cost) approximation
It is calculated for scheduling algorithm.
9. determining barrier-road intersection matching cost C using 2 dimension dynamic programming methodsBoundaryCorresponding to minimum value
Pixel value (ubot,vbot) set be barrier and road intersection, the corresponding parallax value of each column u is dBoundary(u)。
10. the part below barrier-road intersection, lane detection carry out secondaryization processing to image, extract Hough
Straight line determines current lane and outermost lane range.
11. calculating obstacle height matching cost C in barrier-road intersection above sectionHeight.Pass through probability first
Function m (u, v) calculates Cm(u,v,dBoundary(u)) be local extremum a possibility that, the value is between -1 and 1.Finally, obstacle
Object matched cost calculation formula is as follows:
Calculate obstacle height matching cost used in probability function m (u, v) can there are many output area 0 to 1
Or functional form between -1 to 1 indicates.
12. determining obstacle height matching cost C using 2 dimension dynamic programming methodsHeightPixel value corresponding to minimum value
(ui,vi) set be and barrier obstacle height information corresponding with the intersection of road.
Choose optimization barrier-road intersection, the method for obstacle height is not unique, such as can using greedy algorithm its
His global optimization method.
13. object width of placing obstacles, height, depth threshold module filter barrier region adjacent in image, determines and belong to
In the image-region of same barrier.
14. exporting road information and obstacle information after threshold filtering.
Choose can behind traffic areas can also using machine learning method to can traffic areas optimize.
Specific embodiment described herein is only an example for the spirit of the invention.The neck of technology belonging to the present invention
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Although the terms such as disparity correspondence, barrier, road surface range are used more herein, it is not precluded using other
A possibility that term.The use of these items is only for be more convenient to describe and explain essence of the invention；They are explained
It at any additional limitation is disagreed with spirit of that invention.
Claims (4)
1. a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera, which is characterized in that including following step
It is rapid:
S1, the image that rgb format is shot by binocular camera, obtained image are binocular image；
S2, image is pre-processed；
S3, the final matching cost for calculating each pixel in image；In image coordinate system, u is the horizontal axis coordinate of pixel, v
For the ordinate of orthogonal axes of pixel；
S4, the corresponding v- disparity map of longitudinal road plane is calculated by final matching cost；
S5, horizon and road surface range are determined using v- disparity map；
S6, within the scope of road surface, barrier-road intersection is calculated；
S7, the part below barrier-road intersection carry out lane detection, carry out secondary treatment to image, it is straight to extract Hough
Line determines current lane and outermost lane range；
S8, in barrier-road intersection above section, obstacle height is calculated；
The threshold module of S9, place obstacles object width, height and depth filter barrier region adjacent in image, determine and belong to
In the image-region of same barrier；
S10, output pass through road region information and obstacle information after threshold filtering,
It include gray processing, except distortion and solid are corrected to the pretreatment that image carries out in the step S2,Detection side's step
In rapid S3, the final matching cost of each pixel in image is calculated specifically:
S301, parallax d corresponding to pixel each in pretreated picture calculate of the difference based on gray value absolute value
With cost Cv(ui, vi, di), calculation method is as follows:
Cv(ui, vi, di)=imgleft(ui, vi)-imgright(ui-di, vi)
Wherein, uiFor the abscissa of pixel i under image coordinate system, viFor the ordinate of pixel i under image coordinate system,
imgleft(ui, vi) it is gray value absolute value of the pixel i in left figure, imgright(ui-di, vi) be coordinate be (ui-di, vi)
Gray value absolute value of the pixel in right figure, di is the corresponding parallax of pixel i；
S302, the sliding window convolution filter that window is n × n is carried out to the matching cost of the difference calculated based on gray value absolute value
Wave obtains final matching cost, and wherein n is parameter preset；
The step S4 specifically:
S401, it sums to projection of the final disparity correspondence cost on the image coordinate system longitudinal axis, and calculates the every a line of image
The longitudinal axis corresponds to the minimum value C of the sum of parallax costV, min:
CV, min=min (Cv(vi, di))
It finds out every row and corresponds to the sum of parallax cost less than CV, min+TdCorresponding parallax d, to obtain initial v- disparity map；Depending on
Poor cost threshold value TdFor preset value；
S402, initial v- disparity map abscissa be vx, ordinate vy, initial v- disparity map projection is high to real coordinate system
The mapping of y and depth x are spent, calculation formula is as follows:
stereofIt is defined as binocular camera focal length；
stereov0It is defined as binocular image center；
stereobaselineIt is defined as binocular camera baseline length；
v0It is defined as v- parallax curve and y-axis focus, i.e. horizon ordinate；
Using B- spline curve fitting pavement-height and depth relationship, v- parallax plan is returned in last inverse mapping, that is, obtains longitudinal
The corresponding v- disparity map of road plane.
2. a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera according to claim 1,
It is characterized in that, in step S5, horizon v0The curve that the image-region that the time difference is 0 as in v- disparity map is constituted；Parallax is greater than 0
Image-region be road surface range.
3. a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera according to claim 2,
It is characterized in that, step S6 specifically:
S601, within the scope of road surface, utilize the biaxial stress structure relationship of a line v every in v- disparity map and corresponding road surface parallax d, calculate
Barrier-road intersection matching cost CBoundarv；The barrier-road intersection matching cost is by path adaptation cost and object
It is formed with cost two parts, wherein path adaptation cost v and d meets the mapping relations of v- disparity map and object matches cost is each
It is capable then correspond to identical parallax d；Its specific formula for calculation is as follows:
uiIt is defined as the abscissa of pixel i under image coordinate system；
viIt is defined as the ordinate of pixel i under image coordinate system；
diIt is defined as the parallax of pixel i；
f(vi) it is defined as parallax d in road surface in v- disparity mapiWith viMapping relations；
S602, barrier-road intersection matching cost C is determined using 2 dimension dynamic programming methodsBoundarvPicture corresponding to minimum value
Plain value set is the intersection of barrier and road, and the corresponding parallax value of each column u is dBoundarv(u)。
4. a kind of barrier and drivable region detection method based on vehicle-mounted binocular camera according to claim 3,
It is characterized in that, the step S8 specifically:
S801, in barrier-road intersection above section, calculate obstacle height matching cost CHeight: pass through probability letter first
Number m (u, v) calculates final matching cost Cm (u, v, dBoundary(u)) be local extremum a possibility that, the value between -1 and 1 it
Between；Finally, obstacle height matching cost calculation formula is as follows:
m(ui, v) and more than road intersection certain pixel (u is defined as in image coordinate systemi, v) and it is general with road contact barrier
Rate；
vbotIt is defined as barrier in image coordinate system-road intersection and corresponds to ordinate；
S802, obstacle height matching cost C is determined using 2 dimension dynamic programming methodsHeightPixel value corresponding to minimum value
(ui, vi) set be and barrier obstacle height information corresponding with the intersection of road.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610538582.8A CN106228110B (en) | 2016-07-07 | 2016-07-07 | A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610538582.8A CN106228110B (en) | 2016-07-07 | 2016-07-07 | A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106228110A CN106228110A (en) | 2016-12-14 |
| CN106228110B true CN106228110B (en) | 2019-09-20 |
Family
ID=57520331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610538582.8A Active CN106228110B (en) | 2016-07-07 | 2016-07-07 | A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106228110B (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109215044B (en) * | 2017-06-30 | 2020-12-15 | 京东方科技集团股份有限公司 | Image processing method and system, storage medium, and mobile system |
| CN109426760A (en) * | 2017-08-22 | 2019-03-05 | 聚晶半导体股份有限公司 | Road image processing method and road image processing device |
| CN107610152B (en) * | 2017-08-31 | 2020-02-28 | 杭州视氪科技有限公司 | Passage detection method for avoiding water surface and obstacles |
| CN109472180A (en) * | 2017-09-07 | 2019-03-15 | 聚晶半导体股份有限公司 | Road image processing method and road image processing device |
| CN107909009B (en) * | 2017-10-27 | 2021-09-17 | 北京中科慧眼科技有限公司 | Obstacle detection method and device based on road surface learning |
| CN107909010B (en) * | 2017-10-27 | 2022-03-18 | 北京中科慧眼科技有限公司 | Road obstacle detection method and device |
| CN107990878B (en) * | 2017-11-08 | 2019-08-27 | 中国科学院西安光学精密机械研究所 | Ranging system and ranging method based on low-light binocular camera |
| CN107729878A (en) * | 2017-11-14 | 2018-02-23 | 智车优行科技（北京）有限公司 | Obstacle detection method and device, equipment, vehicle, program and storage medium |
| CN108009511A (en) * | 2017-12-14 | 2018-05-08 | 元橡科技（北京）有限公司 | Method for detecting area and device are exercised based on RGBD |
| CN107958226B (en) * | 2017-12-15 | 2020-05-22 | 海信集团有限公司 | Road curve detection method, device and terminal |
| CN108319910B (en) * | 2018-01-30 | 2021-11-16 | 海信集团有限公司 | Vehicle identification method and device and terminal |
| CN110378168B (en) * | 2018-04-12 | 2023-05-30 | 海信集团有限公司 | Method, device and terminal for fusing multiple types of barriers |
| WO2019232804A1 (en) * | 2018-06-08 | 2019-12-12 | 珊口（深圳）智能科技有限公司 | Software updating method and system, and mobile robot and server |
| CN110598505B (en) * | 2018-06-12 | 2023-05-02 | 海信集团有限公司 | Obstacle suspension state judging method, device and terminal |
| CN110807347B (en) * | 2018-08-06 | 2023-07-25 | 海信集团有限公司 | Obstacle detection method, obstacle detection device and terminal |
| CN109241855B (en) * | 2018-08-10 | 2022-02-11 | 西安交通大学 | A detection method of intelligent vehicle driving area based on stereo vision |
| CN109145798B (en) * | 2018-08-13 | 2021-10-22 | 浙江零跑科技股份有限公司 | Driving scene target identification and travelable region segmentation integration method |
| CN109460709B (en) * | 2018-10-12 | 2020-08-04 | 南京大学 | RTG visual barrier detection method based on RGB and D information fusion |
| CN109445592B (en) * | 2018-10-30 | 2020-07-31 | 北京小米移动软件有限公司 | Passage determination method and device, electronic equipment and computer readable storage medium |
| CN111209770B (en) * | 2018-11-21 | 2024-04-23 | 北京三星通信技术研究有限公司 | Lane line recognition method and device |
| CN109732591B (en) * | 2018-12-24 | 2021-09-24 | 济南大学 | A multi-robot cluster control method in an environment with obstacles |
| CN109703569B (en) * | 2019-02-21 | 2021-07-27 | 百度在线网络技术（北京）有限公司 | Information processing method, device and storage medium |
| CN110038666B (en) * | 2019-05-13 | 2024-03-29 | 闽江师范高等专科学校 | Intelligent crushing machine |
| CN112036210B (en) * | 2019-06-03 | 2024-03-08 | 杭州海康机器人股份有限公司 | Method and device for detecting obstacle, storage medium and mobile robot |
| CN110488816B (en) * | 2019-08-06 | 2021-12-31 | 华为技术有限公司 | Automatic driving longitudinal planning method and related equipment |
| CN112776777A (en) * | 2019-11-08 | 2021-05-11 | 郑州宇通客车股份有限公司 | Vehicle and height-limiting anti-collision control method thereof |
| JP7293100B2 (en) * | 2019-12-17 | 2023-06-19 | 日立Astemo株式会社 | camera system |
| CN111443704B (en) * | 2019-12-19 | 2021-07-06 | 苏州智加科技有限公司 | Obstacle positioning method and device for automatic driving system |
| CN111488812B (en) * | 2020-04-01 | 2022-02-22 | 腾讯科技（深圳）有限公司 | Obstacle position recognition method and device, computer equipment and storage medium |
| CN112097732A (en) * | 2020-08-04 | 2020-12-18 | 北京中科慧眼科技有限公司 | Binocular camera-based three-dimensional distance measurement method, system, equipment and readable storage medium |
| CN112150452B (en) * | 2020-09-30 | 2022-04-15 | 西南交通大学 | Millimeter wave imaging-based railway sleeper defect detection method |
| CN113297939B (en) * | 2021-05-17 | 2024-04-16 | 深圳市优必选科技股份有限公司 | Obstacle detection method, obstacle detection system, terminal device and storage medium |
| CN113554646B (en) * | 2021-09-17 | 2021-12-10 | 江苏正金建筑装饰工程有限公司 | Intelligent urban road pavement detection method and system based on computer vision |
| CN114219992B (en) * | 2021-12-14 | 2022-06-03 | 杭州古伽船舶科技有限公司 | Unmanned ship obstacle avoidance system based on image recognition technology |
| CN114708567B (en) * | 2022-06-06 | 2022-09-06 | 济南融瓴科技发展有限公司 | Road surface hollow detection and avoidance method and system based on binocular camera |
| CN119130985A (en) * | 2024-09-09 | 2024-12-13 | 元橡科技（北京）有限公司 | Method, device, system and storage medium for determining ground parallax false detection |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101999972A (en) * | 2010-11-24 | 2011-04-06 | 上海理工大学 | Stereoscopic vision based auxiliary walking device for blindmen and auxiliary method thereof |
| WO2013077338A1 (en) * | 2011-11-21 | 2013-05-30 | 株式会社ニコン | Display device, and display control program |
| CN103185571A (en) * | 2011-12-30 | 2013-07-03 | 财团法人车辆研究测试中心 | Drivable space detection system and its detection method |
| CN103679707A (en) * | 2013-11-26 | 2014-03-26 | 西安交通大学 | Binocular camera disparity map based road obstacle detection system and method |
| EP2838051A2 (en) * | 2013-08-12 | 2015-02-18 | Ricoh Company, Ltd. | Linear road marking detection method and linear road marking detection apparatus |
-
2016
- 2016-07-07 CN CN201610538582.8A patent/CN106228110B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101999972A (en) * | 2010-11-24 | 2011-04-06 | 上海理工大学 | Stereoscopic vision based auxiliary walking device for blindmen and auxiliary method thereof |
| WO2013077338A1 (en) * | 2011-11-21 | 2013-05-30 | 株式会社ニコン | Display device, and display control program |
| CN103185571A (en) * | 2011-12-30 | 2013-07-03 | 财团法人车辆研究测试中心 | Drivable space detection system and its detection method |
| EP2838051A2 (en) * | 2013-08-12 | 2015-02-18 | Ricoh Company, Ltd. | Linear road marking detection method and linear road marking detection apparatus |
| CN103679707A (en) * | 2013-11-26 | 2014-03-26 | 西安交通大学 | Binocular camera disparity map based road obstacle detection system and method |
Non-Patent Citations (1)
| Title |
|---|
| 基于双目视觉的障碍物识别研究;李宁;《中国优秀硕士学位论文全文数据库 信息科技辑》;20071115(第5期);第2-4章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106228110A (en) | 2016-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106228110B (en) | A kind of barrier and drivable region detection method based on vehicle-mounted binocular camera | |
| CN112396650B (en) | Target ranging system and method based on fusion of image and laser radar | |
| EP3732657B1 (en) | Vehicle localization | |
| CN108802785B (en) | Vehicle self-positioning method based on high-precision vector map and monocular vision sensor | |
| CN115032651B (en) | Target detection method based on laser radar and machine vision fusion | |
| CN106681353B (en) | Obstacle avoidance method and system for UAV based on binocular vision and optical flow fusion | |
| CN106225723B (en) | A kind of hinged angle measuring method of multiple row vehicle based on backsight binocular camera | |
| CN111443704B (en) | Obstacle positioning method and device for automatic driving system | |
| CN109359409A (en) | A kind of vehicle passability detection system of view-based access control model and laser radar sensor | |
| CN110378202B (en) | Omnidirectional pedestrian collision early warning method based on fisheye lens | |
| CN113327296B (en) | Laser radar and camera online combined calibration method based on depth weighting | |
| US7623700B2 (en) | Stereoscopic image processing apparatus and the method of processing stereoscopic images | |
| CN109631896A (en) | A kind of parking lot autonomous parking localization method based on vehicle vision and motion information | |
| CN107577996A (en) | A method and system for identifying vehicle driving path deviation | |
| CN106256606A (en) | A kind of lane departure warning method based on vehicle-mounted binocular camera | |
| CN107590438A (en) | A kind of intelligent auxiliary driving method and system | |
| CN115937810A (en) | A sensor fusion method based on binocular camera guidance | |
| CN110334678A (en) | A Pedestrian Detection Method Based on Vision Fusion | |
| CN104700414A (en) | Rapid distance-measuring method for pedestrian on road ahead on the basis of on-board binocular camera | |
| Shunsuke et al. | GNSS/INS/on-board camera integration for vehicle self-localization in urban canyon | |
| CN109410264A (en) | A kind of front vehicles distance measurement method based on laser point cloud and image co-registration | |
| Labayrade et al. | A reliable and robust lane detection system based on the parallel use of three algorithms for driving safety assistance | |
| CN114463303B (en) | Road target detection method based on fusion of binocular camera and laser radar | |
| CN114200442B (en) | Road target detection and association method based on millimeter wave radar and vision | |
| CN113920183A (en) | Monocular vision-based vehicle front obstacle distance measurement method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 6 / F, Xintu building, 451 Internet of things street, Binjiang District, Hangzhou City, Zhejiang Province, 310051 Patentee after: Zhejiang Zero run Technology Co.,Ltd. Address before: 6 / F, Xintu building, 451 Internet of things street, Binjiang District, Hangzhou City, Zhejiang Province, 310051 Patentee before: ZHEJIANG LEAPMOTOR TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |