CN109754415A - A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision - Google Patents
A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision Download PDFInfo
- Publication number
- CN109754415A CN109754415A CN201711061178.7A CN201711061178A CN109754415A CN 109754415 A CN109754415 A CN 109754415A CN 201711061178 A CN201711061178 A CN 201711061178A CN 109754415 A CN109754415 A CN 109754415A
- Authority
- CN
- China
- Prior art keywords
- camera
- mainboard
- module
- vehicle
- follows
- 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
Landscapes
- Studio Devices (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision, hardware components include: four groups of binocular camera shooting modules and mainboard module.Wherein the main component of binocular camera shooting module includes shell, binocular camera shooting mould group, adjustment knob；The main component of mainboard module includes mainboard module shell, mainboard, mainboard module pedestal, cooling fin.Binocular camera shooting module housing can be used for its fixation between installation car body there are screw hole, and adjustment knob can be used for adjusting dual camera towards angle, and effective adjustable range is 20 degree.Mainboard module shell side provides power interface, network interface, coaxial cable interface and video output interface, it is respectively used to communicate between system power supply, external extension information transmission, mainboard and camera module and results for video visualizes output, wherein coaxial cable interface can power simultaneously for camera module.Mainboard module pedestal can be used for there are 4 fixation holes and install fixation between car body, the pedestal is detachable.
Description
Technical field
The present invention relates to automobile assistant driving technical field, specifically: a kind of vehicle-mounted panoramic based on multiple groups binocular vision
Three-dimensional sensory perceptual system.
Background technique
In recent years, Global Auto industry development situation is good, and China's automobile volume of production and marketing speedup is continuously improved.Wherein, in
The data statistics of Automobile Associations, state shows that China's auto output in 2015 is 2450.3 ten thousand, and 2372.3 amplification compared with 2014 are
3.3%.China's automobile sales volume in 2015 is 2459.7 ten thousand amounts, is 4.68% compared with 2349.9 ten thousand amplification in 2014.2016 I
State's automobile sales volume realizes that rapid growth, production and marketing total amount record high again, and annual auto output completes 2,811.9 ten thousand, whole year
Automobile sales volume reaches 2,802.8 ten thousand, increases by 14.76% and 13.7% respectively than same period last year, is higher than same period last year (2015 years
Degree) 11.5 and 9.0 percentage points.
Advanced DAS (Driver Assistant System) (ADAS) technology be using the various kinds of sensors that is installed on vehicle, as GPS receiver,
Inertial Measurement Unit, imaging sensor, radar range finder etc. measure the critical environments on car body periphery, and utilize carrying
Computing platform carries out such as quiet, dynamic object identification, tracking scheduling algorithm and handles, and Calculation results are notified driver in time,
To reduce the perception blind area of driver, auxiliary driver carries out most safely and effectively Driving Decision-making under current environmental condition.With
Continuing to optimize for the development of sensor technology in recent years and chip manufacturing process, sensor measurement stability constantly improves, cost
It reduces year by year, the computing capability of Industry Control grade computing platform is continuously improved, and advanced DAS (Driver Assistant System) initially enters middle-end vapour
Car market, technical prospect and good market prospect.
Dual camera stereovision technique is the developing typical case technology of machine vision technique in recent years, core
Component, that is, dual camera sensor manufacturing cost is lower, and perception information is abundant, is therefore widely used in industrial detection, biology doctor
It learns, military affairs detect and multiple key areas such as robot perception unit making.Dual camera vision technique can get environment
Three-dimensional information, but also there is basic problem existing for camera vision, if it is sensitive to illumination variation, measurement accuracy is vulnerable to calibration
Parameter accuracy influences, and waits interference effects etc. vulnerable to blocking.Since the above problem and automobile industry want sensor technology stability
Ask higher, dual camera vision technique is currently few in the typical case of automobile industry, especially still locates in market development at home
In the starting stage.
Advanced DAS (Driver Assistant System) belongs to active security system, i.e., simultaneously by onboard sensor actively perceive external environment
Reaction is judged, perception target reduces driver as far as possible and perceives blind area, then actively perceive range is wider, ilities
It is better, but relative skill cost is higher, and difficulty is bigger.Currently marketed main perception class drives auxiliary product and is often laid in vehicle
Front end or rear vehicle end, for perceiving small range scene information, perception blind area is larger, for current complicated urban transportation
Environment easily causes missing inspection or erroneous judgement, and it is larger that user experience optimizes space.
Summary of the invention
Goal of the invention: a kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision is provided, multisensor is passed through
Information fusion technology realizes the peripheral vehicle environment sensing for being bordering on 360 degree of gamuts.
The present invention is achieved by the following technical solutions:
A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision, hardware components include: four groups of binocular camera shootings
Module and mainboard module.Wherein the main component of binocular camera shooting module includes shell, binocular camera shooting mould group, adjustment knob；Mainboard
The main component of module includes mainboard module shell, mainboard, mainboard module pedestal, cooling fin.Binocular camera shooting module housing there are
Screw hole, can be used for its fixation between installation car body, and adjustment knob can be used for adjusting dual camera towards angle, effectively adjusting model
Enclose is 20 degree.Mainboard module shell side provides power interface, network interface, coaxial cable interface and video output interface, point
It Yong Yu be communicated between system power supply, external extension information transmission, mainboard and camera module and results for video visualizes output,
Standard coaxial cable interface can power simultaneously for camera module.Mainboard module pedestal can be used for and installation car body there are 4 fixation holes
Between it is fixed, the pedestal is detachable.
Four groups of binocular camera shooting modules are connect with mainboard module by coaxial cable；
The mainboard module is powered using 24V DC voltage-stabilizing；
Function system of the invention mainly includes environment sensing algorithm and perception target characteristic database two parts.
Environment sensing strategy of the invention mainly includes following two part:
1. the comprehensive ranging based on dual camera vision；
2. the interest object detection based on camera image；
Wherein, the key step of the comprehensive ranging based on dual camera vision are as follows:
The first step obtains image and is pre-processed, which can segment are as follows:
Step 1, image gray processing；
Step 2, image histogram are balanced；
Step 3, image denoising；
Second step obtains dual camera two images characteristic point；
Third step realizes that Feature Points Matching, the step can be segmented are as follows:
Step 1, two feature point sets slightly match；
Step 2, Mismatching point screen out；
4th step, matching double points disparity computation；
5th step calculates the practical three-dimensional coordinate of matching double points.
The key step of dynamic object detection based on camera image are as follows:
The first step, image segmentation；
Second step, interest Object Extraction；
Third step, interest object space calculate；
4th step, interest object and main body vehicle relative velocity calculate；
5th step, interest object risk assessment, the step can be segmented are as follows:
Step 1, the prediction of interest object space；
Step 2, by predicted position corresponding thereto speed to interest object grade of risk；
6th step generates early warning information according to the above results.
The present invention provides video output interface, can visualize to current real-time detection result.
The present invention also provides network interface, carry out after more equipment timestamps synchronize, it can be in real time to other vehicle-mounted controls, decision
Unit sends environment sensing information.
Beneficial effect by adopting the above technical scheme is:
1. facility environment sensing range is wide, more reliable analysis result can be provided for Evaluation Environment safety.
2. equipment simple installation realizes basic warning function, and can provide data for other vehicle control syetems and connect
Mouthful.
3. this product modularity is good, at low cost, it is suitble to produce in enormous quantities.
4. each component of this product can reach Industry Control rank, wherein dual camera photographing module can meet substantially vehicle-mounted
Waterproof requirement, mainboard module can meet vehicle-mounted vibration-resistant, dust-proof requirement.
Detailed description of the invention
Fig. 1 present system structure and scheme of installation；
Fig. 2 present system work frame chart；
Fig. 3 Binocular Vision Principle figure of the present invention；
The present invention is based on the comprehensive ranging flow charts of dual camera vision by Fig. 4；
The present invention is based on the interest object detection flow charts of camera image by Fig. 5；
In figure: 1. photographing module shells, 21. right cameras, 22. left cameras, 3. adjustment knobs, 4. mainboards, 5. mainboards
Shell, 6. mainboard module pedestals, 7. cooling fins, 8. coaxial cable interfaces, 9.VGA video output interface, 10. motherboard power supplies and net
Network interface, 11. mainboard module mounting holes.
Specific embodiment
The present invention will be further described with reference to the accompanying drawing；
As shown in figure 1, figure 2, figure 3, figure 4 and figure 5, it is vertical to provide a kind of vehicle-mounted panoramic based on multiple groups binocular vision by the present invention
Body sensory perceptual system includes: photographing module shell (1), binocular camera shooting module (2), adjustment knob (3), mainboard (4), mainboard shell
(5), mainboard module pedestal (6), cooling fin (7) and a series of power supplys, communication and output interface (8,9,10)；The camera shooting mould
The adjustment knob (3) of one end installation of block (2) can be used for adjusting dual camera towards angle, and effective adjustable range is 20 degree；Institute
State the inside that mainboard (4) is fixedly installed in mainboard shell (5)；Cooling fin (7) are installed at the top of the mainboard shell (5)；Institute
Mainboard shell (5) bottom end can be fixedly installed in by stating mainboard module pedestal (6), detachably, be mainly used for mainboard module and installation vehicle
Fixation between, 4 jiaos of the pedestal are furnished with mounting hole (11)；The camera module (2) is pacified by right camera (21) and left camera (22)
Dress composition, and right camera (21) is parallel with the two axial lines of left camera (22)；
4 groups of binocular camera shooting modules (2) are connect with mainboard (4) by coaxial cable；
The mainboard module is powered using 24V DC voltage-stabilizing；
Each step technique details of the comprehensive ranging unit based on dual camera vision in the environment sensing strategy is such as
Under.
The first step, image preprocessing step details are as follows:
1. image gray processing；
2. it is balanced to carry out image histogram in the case where removing strong light interference condition；
3. carrying out image denoising using median filtering scheduling algorithm；
Second step, obtains dual camera two images characteristic point, and details is as follows；
1. obtaining right camera (21) and left camera (22) picture after pretreatment, it is synchronous to carry out timestamp.
2. extracting image Harris angle point as characteristic point.
Third step realizes that Feature Points Matching, the step can be segmented are as follows:
Step 1, two feature point sets are according to the thick matching of characteristic point feature vector distance；
Step 2 is searched original image colouring information and is verified to matching result, screens out Mismatching point；
4th step, matching double points disparity computation.
5th step obtains the practical three-dimensional coordinate of blip pixel according to disparity computation result, and calculation formula is as follows:
Wherein (x, y, z) indicates space characteristics point the to be sought coordinate under camera coordinate system；
Dis=Xleft-Xright, indicate characteristic point base line projection distance, that is, reflect the projector distance amount of parallax；
B are as follows: the axial line distance of right camera (21) and left camera (22)；
XleftWith XrightAre as follows: the matching characteristic point abscissa of right camera (21) and left camera (22) acquisition:
Y are as follows: the ordinate of right camera (21) and left camera (22)；
F are as follows: the focal length of right camera (21) and left camera (22)；
The information of 4 groups of binocular images is handled according to the above process.
The dynamic object test section based on camera image based on dual camera vision in the environment sensing strategy
The major technique details divided is as follows:
The first step realizes cluster formula image point using the feature vector that image pixel gray level, color and location of pixels are constituted
It cuts；
Second step, extracts the interest object such as people, vehicle using SVM classifier and labeling；
Third step, interest object space calculate, and main details are；
1. obtaining all characteristic point three-dimensional positions of interest object area by comprehensive distance measurement result；
2. removing noise position, relative distance of the nearest characteristic point of selected distance main body vehicle as the interest object
4th step calculates interest object and main body vehicle relative velocity using two continuous frames data difference；
5th step, interest object risk assessment, the step can be segmented are as follows:
Step 1 predicts interest object space using above-mentioned speed calculated result；
Step 2, by predicted position corresponding thereto speed to interest object grade of risk；
6th step generates early warning information according to the above results.
The mainboard nucleus module of the present invention onboard installation site can be located at standby box space, 4 groups of dual cameras peaces
Holding position can pass through binocular camera shooting module knob tune as shown in Figure 1, be respectively arranged on the four edges of car body top planar rectangular
Its whole camera angle, and carry out automatic Calibration.
In running car when on road, starting warning function may be selected, and mating source of early warning is connected in the present invention
It is connected to mainboard module.
The present invention can send current detection result to other equipment by RJ-45 Ethernet interface.
Claims (4)
1. the present invention provides a kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision, comprising: photographing module shell
(1), binocular camera shooting module (2), adjustment knob (3), mainboard (4), mainboard shell (5), mainboard module pedestal (6), cooling fin (7)
With a series of power supplys, communication and output interface (8,9,10)；It is characterized by:
The adjustment knob (3) of one end installation of the photographing module (2) can be used for adjusting dual camera towards angle, effectively adjust
Adjusting range is 20 degree；
The mainboard (4) is fixedly installed in the inside of mainboard shell (5)；
Cooling fin (7) are installed at the top of the mainboard shell (5)；
The mainboard module pedestal (6) can be fixedly installed in mainboard shell (5) bottom end, detachably, be mainly used for mainboard module with
Fixation between vehicle is installed, 4 jiaos of the pedestal are furnished with mounting hole (11)；
The camera module (2) is made of right camera (21) and left camera (22) installation, and right camera (21) and left camera (22)
Two axial lines it is parallel；
4 groups of binocular camera shooting modules (2) are connect with mainboard (4) by coaxial cable；
The mainboard module is powered using 24V DC voltage-stabilizing.
2. a kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision as described in right wants 1, it is characterised in that institute
Each step technique details of the comprehensive ranging unit based on dual camera vision stated in environment sensing strategy is as follows；
The first step, image preprocessing step details are as follows:
Step 1, image gray processing；
It is balanced to carry out image histogram in the case where removing strong light interference condition for step 2；
Step 3 carries out image denoising using median filtering scheduling algorithm；
Second step, obtains dual camera two images characteristic point, and details is as follows；
Step 1 obtains right camera (21) and left camera (22) picture after pretreatment, and it is synchronous to carry out timestamp；
Step 2 extracts image Harris angle point as characteristic point；
Third step realizes that Feature Points Matching, the step can be segmented are as follows:
Step 1, two feature point sets are according to the thick matching of characteristic point feature vector distance；
Step 2 is searched original image colouring information and is verified to matching result, screens out Mismatching point；
4th step, matching double points disparity computation；
5th step obtains the practical three-dimensional coordinate of blip pixel according to disparity computation result, and calculation formula is as follows:
Wherein (x, y, z) indicates space characteristics point the to be sought coordinate under camera coordinate system；
Dis=Xleft-Xright, indicate characteristic point base line projection distance, that is, reflect the projector distance amount of parallax；
B are as follows: the axial line distance of right camera (21) and left camera (22)；
XleftWith XrightAre as follows: the matching characteristic point abscissa of right camera (21) and left camera (22) acquisition:
Y are as follows: the ordinate of right camera (21) and left camera (22)；
F are as follows: the focal length of right camera (21) and left camera (22)；
The information of 4 groups of binocular images is handled according to the above process.
3. a kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision as described in right wants 1, it is characterised in that institute
State the main skill of the dynamic object detection part based on camera image based on dual camera vision in environment sensing strategy
Art details is as follows:
The first step realizes cluster formula image segmentation using the feature vector that image pixel gray level, color and location of pixels are constituted；
Second step, extracts the interest object such as people, vehicle using SVM classifier and labeling；
Third step, interest object space calculate, and main details are；
Step 1 is obtained all characteristic point three-dimensional positions of interest object area by comprehensive distance measurement result；
Step 2 removes noise position, relative distance of the nearest characteristic point of selected distance main body vehicle as the interest object；
4th step calculates interest object and main body vehicle relative velocity using two continuous frames data difference；
5th step, interest object risk assessment, the step can be segmented are as follows:
Step 1 predicts interest object space using above-mentioned speed calculated result；
Step 2, by predicted position corresponding thereto speed to interest object grade of risk；
6th step generates early warning information according to the above results.
4. a kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision as described in right wants 1, it is characterised in that institute
Mainboard nucleus module of the present invention onboard installation site is stated, can be located at standby box space, 4 groups of dual camera installation sites are as schemed
It shown in 1, is respectively arranged on the four edges of car body top planar rectangular, it can be adjusted by binocular camera shooting module knob and images angle
Degree, and carry out automatic Calibration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711061178.7A CN109754415A (en) | 2017-11-02 | 2017-11-02 | A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711061178.7A CN109754415A (en) | 2017-11-02 | 2017-11-02 | A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109754415A true CN109754415A (en) | 2019-05-14 |
Family
ID=66398322
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711061178.7A Pending CN109754415A (en) | 2017-11-02 | 2017-11-02 | A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109754415A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110944107A (en) * | 2019-12-29 | 2020-03-31 | 徐书诚 | Computer system for realizing single-screen restoration display of all-round network camera set |
| CN111559314A (en) * | 2020-04-27 | 2020-08-21 | 长沙立中汽车设计开发股份有限公司 | Depth and image information fused 3D enhanced panoramic looking-around system and implementation method |
| CN112001974A (en) * | 2020-08-25 | 2020-11-27 | 中国水产科学研究院渔业机械仪器研究所 | Calibration device and calibration method for underwater stereoscopic observation system |
| CN112360285A (en) * | 2020-10-27 | 2021-02-12 | 浙江清华长三角研究院 | Automatic door three-dimensional machine vision sensor and sensing method thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318561A (en) * | 2014-10-22 | 2015-01-28 | 上海理工大学 | Method for detecting vehicle motion information based on integration of binocular stereoscopic vision and optical flow |
| CN205395928U (en) * | 2016-01-19 | 2016-07-27 | 华晨汽车集团控股有限公司 | On -vehicle three -dimensional panorama monitored control system |
| CN105912998A (en) * | 2016-04-05 | 2016-08-31 | 辽宁工业大学 | Vehicle collision prevention early warning method based on vision |
| CN106686341A (en) * | 2015-11-05 | 2017-05-17 | 深圳市祈飞科技有限公司 | Interactive vehicle-mounted intelligent terminal system and monitoring method |
| CN106799993A (en) * | 2017-01-09 | 2017-06-06 | 智车优行科技（北京）有限公司 | Streetscape acquisition method and system, vehicle |
| CN106878687A (en) * | 2017-04-12 | 2017-06-20 | 吉林大学 | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor |
| CN106926800A (en) * | 2017-03-28 | 2017-07-07 | 重庆大学 | The vehicle-mounted visually-perceptible system of multi-cam adaptation |
-
2017
- 2017-11-02 CN CN201711061178.7A patent/CN109754415A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318561A (en) * | 2014-10-22 | 2015-01-28 | 上海理工大学 | Method for detecting vehicle motion information based on integration of binocular stereoscopic vision and optical flow |
| CN106686341A (en) * | 2015-11-05 | 2017-05-17 | 深圳市祈飞科技有限公司 | Interactive vehicle-mounted intelligent terminal system and monitoring method |
| CN205395928U (en) * | 2016-01-19 | 2016-07-27 | 华晨汽车集团控股有限公司 | On -vehicle three -dimensional panorama monitored control system |
| CN105912998A (en) * | 2016-04-05 | 2016-08-31 | 辽宁工业大学 | Vehicle collision prevention early warning method based on vision |
| CN106799993A (en) * | 2017-01-09 | 2017-06-06 | 智车优行科技（北京）有限公司 | Streetscape acquisition method and system, vehicle |
| CN106926800A (en) * | 2017-03-28 | 2017-07-07 | 重庆大学 | The vehicle-mounted visually-perceptible system of multi-cam adaptation |
| CN106878687A (en) * | 2017-04-12 | 2017-06-20 | 吉林大学 | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor |
Non-Patent Citations (5)
| Title |
|---|
| 刘冬 等: "3D车载环视全景生成方法", 《计算机科学》 * |
| 刘志强 等: "基于多信息检测的车辆智能防撞预警技术研究", 《中国安全科学学报》 * |
| 同志学 等: "《基于双目序列图像的测距定位与自车速度估计》", 《应用光学》 * |
| 高一宁 等: "基于Canny算子和Harris角点提取的三维重建方法", 《电子测试》 * |
| 黄慧玲 等: "基于前方车辆行为识别的碰撞预警系统", 《华中科技大学学报(自然科学版)》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110944107A (en) * | 2019-12-29 | 2020-03-31 | 徐书诚 | Computer system for realizing single-screen restoration display of all-round network camera set |
| CN111559314A (en) * | 2020-04-27 | 2020-08-21 | 长沙立中汽车设计开发股份有限公司 | Depth and image information fused 3D enhanced panoramic looking-around system and implementation method |
| CN111559314B (en) * | 2020-04-27 | 2021-08-24 | 长沙立中汽车设计开发股份有限公司 | Depth and image information fused 3D enhanced panoramic looking-around system and implementation method |
| CN112001974A (en) * | 2020-08-25 | 2020-11-27 | 中国水产科学研究院渔业机械仪器研究所 | Calibration device and calibration method for underwater stereoscopic observation system |
| CN112360285A (en) * | 2020-10-27 | 2021-02-12 | 浙江清华长三角研究院 | Automatic door three-dimensional machine vision sensor and sensing method thereof |
| CN112360285B (en) * | 2020-10-27 | 2023-07-11 | 浙江清华长三角研究院 | Automatic door stereoscopic machine vision sensor and sensing method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10726576B2 (en) | System and method for identifying a camera pose of a forward facing camera in a vehicle | |
| Zhu et al. | The multivehicle stereo event camera dataset: An event camera dataset for 3D perception | |
| US7587081B2 (en) | Method for processing stereo vision data using image density | |
| CN109754415A (en) | A kind of vehicle-mounted panoramic solid sensory perceptual system based on multiple groups binocular vision | |
| EP2445221B1 (en) | Correcting frame-to-frame image changes due to motion for three dimensional (3-d) persistent observations | |
| CN106960454B (en) | Depth of field obstacle avoidance method and equipment and unmanned aerial vehicle | |
| JP6574611B2 (en) | Sensor system for obtaining distance information based on stereoscopic images | |
| CN109263637B (en) | Collision prediction method and device | |
| CN105069804B (en) | Threedimensional model scan rebuilding method based on smart mobile phone | |
| AU2018282302A1 (en) | Integrated sensor calibration in natural scenes | |
| CN108805910A (en) | More mesh Train-borne recorders, object detection method, intelligent driving system and automobile | |
| CN108645375B (en) | Rapid vehicle distance measurement optimization method for vehicle-mounted binocular system | |
| CN106033614B (en) | A kind of mobile camera motion object detection method under strong parallax | |
| US11380111B2 (en) | Image colorization for vehicular camera images | |
| CN114495064A (en) | Monocular depth estimation-based vehicle surrounding obstacle early warning method | |
| CN114463303A (en) | Road target detection method based on fusion of binocular camera and laser radar | |
| CN112561996A (en) | Target detection method in autonomous underwater robot recovery docking | |
| EP3782363B1 (en) | Method for dynamic stereoscopic calibration | |
| EP3005238B1 (en) | Method and system for coordinating between image sensors | |
| John et al. | Automatic calibration and registration of lidar and stereo camera without calibration objects | |
| CN117406234A (en) | Target ranging and tracking method based on single-line laser radar and vision fusion | |
| CN116385997A (en) | Vehicle-mounted obstacle accurate sensing method, system and storage medium | |
| CN116151320A (en) | Visual odometer method and device for resisting dynamic target interference | |
| WO2021232222A1 (en) | Ranging method and apparatus | |
| CN113269857A (en) | Coordinate system relation obtaining method and device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190514 |