CN107145825A - Ground level fitting, camera calibration method and system, car-mounted terminal - Google Patents
Ground level fitting, camera calibration method and system, car-mounted terminal Download PDFInfo
- Publication number
- CN107145825A CN107145825A CN201710209423.8A CN201710209423A CN107145825A CN 107145825 A CN107145825 A CN 107145825A CN 201710209423 A CN201710209423 A CN 201710209423A CN 107145825 A CN107145825 A CN 107145825A
- Authority
- CN
- China
- Prior art keywords
- camera
- plane
- vehicle
- ground level
- escape
- 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
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The present invention provides a kind of ground level fitting, camera calibration method and system, car-mounted terminal, vehicle, including：The public visual field based on the multi-path camera being arranged on vehicle obtains the lane line of some groups of vehicle both sides, and the multi-path camera includes preceding road camera, way of escape camera, left camera and right wing camera respectively positioned at the front of vehicle, rear, left side and right side；The preceding road camera has the public visual field respectively with the left camera, the preceding road camera and the right wing camera, the way of escape camera and the left camera, the way of escape camera and the right wing camera.The nominal data of camera is generated according to the height of fitting ground level.The ground level of the present invention is fitted, camera calibration method and system, car-mounted terminal, vehicle ensure the accuracy of ground level fitting by rejecting the track data of mistake, ensure the accuracy of camera calibration so that the image mosaic position that each camera is shot is more accurate.
Description
Technical field
The present invention relates to the technical field of image procossing, more particularly to a kind of ground level fitting, camera calibration method
And system, car-mounted terminal, vehicle.
Background technology
In recent years, based on looking around the senior DAS (Driver Assistant System) (also cry and look around ADAS systems) of camera by increasingly
Many users are received.By the installation wide-angle camera all around in automobile, the image of vehicle body surrounding is gathered, by algorithm
Analyzing and processing, provides various senior including panorama parks auxiliary, deviation alarm, blind area vehicle detection to driver
Auxiliary drives function, has great importance for the main security for improving automobile.
Before beginning to use and looking around ADAS systems, it is necessary first to which camera is demarcated, to ensure that each camera is clapped
The stitching position for the image taken the photograph is more accurate, and error is smaller.In the prior art, generally imaged using fixed scaling method
Leader is determined, and vehicle is carried out after camera calibration in parking lot, and nominal data is fixed, is not settled with vehicle body and changes and become
Change, it is impossible to meet the actual demand of vehicle traveling.
The content of the invention
The shortcoming of prior art in view of the above, it is an object of the invention to provide a kind of fitting of ground level, camera
Scaling method and system, car-mounted terminal, vehicle, ensure the accuracy that ground level is fitted by rejecting the track data of mistake,
And then ensureing the accuracy of camera calibration so that the image mosaic position that each camera is shot is more accurate, and error is smaller.
In order to achieve the above objects and other related objects, the present invention provides a kind of ground level approximating method, including following step
Suddenly：The public visual field based on the multi-path camera being arranged on vehicle obtains the lane line of some groups of vehicle both sides, the multichannel
Camera includes preceding road camera, way of escape camera, left shooting respectively positioned at the front of vehicle, rear, left side and right side
Head and right wing camera；The preceding road camera and the left camera, the preceding road camera and the right wing camera,
The way of escape camera has public regard respectively with the left camera, the way of escape camera and the right wing camera
It is wild；Judge the lane line of acquired vehicle both sides whether parallel and in same level；Based on parallel and in same
The lane line generation ground point cloud of horizontal plane, and it is fitted generation ground level.
In one embodiment of the invention, the lane line bag of vehicle both sides is obtained based on the multi-path camera being arranged on vehicle
Include following steps：
Any straight line in preceding road camera and its visual field determines the first plane；
Any straight line in way of escape camera and its visual field determines the second plane；
Any straight line in left camera and its visual field determines the 3rd plane；
Any straight line in right wing camera and its visual field determines fourth plane；
By the intersection or second plane and the intersection of the 3rd plane of first plane and the 3rd plane
Determine left-hand lane line；
By the intersection of the intersection or second plane and the fourth plane of first plane and the fourth plane
Determine right-hand lane line.
Meanwhile, the present invention also provides a kind of camera calibration method, comprises the following steps：
According to any of the above-described ground level approximating method fitting generation ground level；
According to the height for the ground level for being fitted generation, the nominal data of camera is generated.
Correspondingly, the present invention also provides a kind of ground level fitting system, including acquisition module, judge module and fitting mould
Block；
The acquisition module is used for the public visual field based on the multi-path camera being arranged on vehicle and obtains some groups of vehicles
The lane line of both sides, the multi-path camera includes preceding road shooting respectively positioned at the front of vehicle, rear, left side and right side
Head, way of escape camera, left camera and right wing camera；The preceding road camera and the left camera, the preceding road
Camera and the right wing camera, the way of escape camera and the left camera, the way of escape camera and the right side
Road camera has the public visual field respectively；
Whether the judge module is used for the lane line for judging acquired vehicle both sides parallel and in same level
On；
The fitting module is used to generate ground point cloud based on the lane line parallel and in same level, and is fitted life
Into ground level.
In one embodiment of the invention, the acquisition module is imaged by following steps based on the multichannel being arranged on vehicle
The lane line of head acquisition vehicle both sides：
Any straight line in preceding road camera and its visual field determines the first plane；
Any straight line in way of escape camera and its visual field determines the second plane；
Any straight line in left camera and its visual field determines the 3rd plane；
Any straight line in right wing camera and its visual field determines fourth plane；
By the intersection or second plane and the intersection of the 3rd plane of first plane and the 3rd plane
Determine left-hand lane line；
By the intersection of the intersection or second plane and the fourth plane of first plane and the fourth plane
Determine right-hand lane line
Correspondingly, the present invention also provides a kind of camera calibration system, including any of the above-described ground level fitting system and
Adjusting module；
The ground level fitting system is used for based on the multi-path camera fitting generation ground level being arranged on vehicle；
The adjusting module is used for the height according to the ground level for being fitted generation, generates the nominal data of camera.
In addition, the present invention also provides a kind of car-mounted terminal, it is connected with multi-path camera, including above-mentioned camera calibration system
System, the multi-path camera includes being located at the preceding road camera on front, rear, left side and the right side of vehicle, way of escape shooting respectively
Head, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera with it is described
Right wing camera, the way of escape camera are distinguished with the left camera, the way of escape camera and the right wing camera
With the public visual field.
In addition, the present invention also provides a kind of vehicle, including above-mentioned car-mounted terminal and multi-path camera, the multichannel shooting
Head include respectively positioned at the preceding road camera on the front of vehicle, rear, left side and right side, way of escape camera, left camera and
Right wing camera；It is the preceding road camera and the left camera, the preceding road camera and the right wing camera, described
Way of escape camera has the public visual field respectively with the left camera, the way of escape camera and the right wing camera.
As described above, the ground level fitting of the present invention, camera calibration method and system, car-mounted terminal, vehicle, with
Lower beneficial effect：
(1) pavement image is gathered based on four-way CCD camera, and obtains the track data in pavement image, by rejecting mistake
Track data come ensure ground level fitting accuracy；
(2) height of measured fitting ground level, generates the nominal data of camera so that what each camera was shot
Image mosaic position is more accurate, and error is smaller；
(3) real-time is good, contentedly the time requirement of plane fitting and camera calibration.
Brief description of the drawings
Fig. 1 is shown as the flow chart of the ground level approximating method of the present invention；
Fig. 2 is shown as lane line and obtains principle schematic；
Fig. 3 is shown as the flow chart of the camera calibration method of the present invention；
Fig. 4 is shown as the structural representation of the ground level fitting system of the present invention；
Fig. 5 is shown as the structural representation of the camera calibration system of the present invention；
Fig. 6 is shown as the structural representation of the car-mounted terminal of the present invention；
Fig. 7 is shown as the structural representation of the vehicle of the present invention.
Component label instructions
1 ground level is fitted system
11 acquisition modules
12 judge modules
13 fitting modules
2 adjusting modules
7 vehicles
71 car-mounted terminals
72 multi-path cameras
Embodiment
Illustrate embodiments of the present invention below by way of specific instantiation, those skilled in the art can be by this specification
Disclosed content understands other advantages and effect of the present invention easily.The present invention can also pass through specific realities different in addition
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints with application, without departing from
Various modifications or alterations are carried out under the spirit of the present invention.It should be noted that, in the case where not conflicting, following examples and implementation
Feature in example can be mutually combined.
It should be noted that the diagram provided in following examples only illustrates the basic structure of the present invention in a schematic way
Think, then in schema only display with relevant component in the present invention rather than according to component count, shape and the size during actual implement
Draw, it is actual when implementing, and kenel, quantity and the ratio of each component can be a kind of random change, and its assembly layout kenel
It is likely more complexity.
The ground level of the present invention is fitted, camera calibration method and system, car-mounted terminal are based on multi-path camera and gather road
Face image, and the track data in pavement image are obtained, carry out ground level fitting by rejecting the track data of mistake；Base again
In the height for the ground level being fitted, the nominal data of camera is generated so that the image mosaic position that each camera is shot is more
Plus it is accurate, error is smaller；Program real-time is good, disclosure satisfy that ground level fitting and the time requirement of camera calibration.
Reference picture 1, ground level approximating method of the invention comprises the following steps：
Step S11, the public visual field based on the multi-path camera being arranged on vehicle obtain the car of some groups of vehicle both sides
Diatom, the multi-path camera is taken the photograph including preceding road camera respectively positioned at the front of vehicle, rear, left side and right side, the way of escape
As head, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera and institute
State right wing camera, the way of escape camera and the left camera, the way of escape camera and the right wing camera point
Ju You not the public visual field.
In ADAS systems, usually using four-way CCD camera, Chinese herbaceous peony, Che Hou, a vehicle body left side, the vehicle body right side are separately mounted to, point
Vehicle front, rear, left side and the scene on right side Yong Yu not observed.The scene splicing in the four-way CCD camera visual field can form 360
That spends looks around panorama.Wherein, existing between adjacent camera between the public visual field, relative camera does not have the public visual field.
The lane line of acquisition vehicle body left front is for example needed, the public visual field of preceding road camera and left camera can be used；Need
The lane line of vehicle body left back is obtained, the combination of left camera and way of escape camera can be used.
As shown in Fig. 2 the acquisition principle of lane line is as follows：The camera A and straight line a in its visual field determines that a space is put down
Face, the camera B and straight line b in its visual field determines a space plane, and the intersection of the two planes is the required car extracted
Diatom.Wherein, due to the collimation error, straight line a and straight line b are located in error plane, and lane line is then to be located at true ground level
On.
Based on above-mentioned principle, the lane line for obtaining vehicle both sides based on the multi-path camera being arranged on vehicle includes following
Step：
11) any straight line in preceding road camera and its visual field determines the first plane.
12) any straight line in way of escape camera and its visual field determines the second plane.
13) any straight line in left camera and its visual field determines the 3rd plane.
14) any straight line in right wing camera and its visual field determines fourth plane.
15) intersection or second plane by first plane and the 3rd plane and the friendship of the 3rd plane
Line determines left-hand lane line.
Specifically, the intersection of first plane and the 3rd plane is the lane line of left front；Described second is flat
The intersection of face and the 3rd plane is the lane line of left back.
16) by first plane and the fourth plane intersection or second plane and the fourth plane friendship
Line determines right-hand lane line.
Specifically, the intersection of first plane and the fourth plane is the lane line of right front；Described second is flat
The intersection of face and the fourth plane is the lane line of right back.
Whether the lane line of the acquired vehicle both sides of step S12, judgement is parallel and in same level.
Under normal circumstances, vehicle is always travelled along lane line, and the lane line on road surface is usually parallel.Therefore,
Need to judge whether lane line is parallel in the horizontal direction, if not parallel, remove the lane line；For parallel lane line also
Need in the vertical direction to judge whether to be in same level, if being not at same level, remove the lane line.This is
Because being easily collected into the wrong data such as curb when collecting rectilinear stretch data, due to curb with ground generally not same
In level height, using curb data ground level can be caused to be fitted mistake；And ground level fitting mistake can cause the mark of camera
Fixed number evidence it is inaccurate, the splicing dislocation between the image that each camera is shot is easily caused, for example, left and preceding road camera are adjusted
After whole, the image mosaic of left and the way of escape can misplace.
Step S13, ground point cloud generated based on the lane line parallel and in same level, and be fitted generation Horizon
Face.
Every straight line can be uniquely determined by two points in space.Therefore, two on a lane line correspondence ground level
Individual point.Three points on the same line can not uniquely determine a plane.Therefore, as long as gathering two different positions in theory
The lane line put can just calculate the space equation of ground level.
Point cloud refers to the set of many points in three dimensions.Using cloud data can fit straight line, plane or its
His geometry.Accordingly, ground point cloud refers to the set of many points on ground.Generation can be fitted using ground cloud data
Ground level.
In the present invention, ground level is fitted according to the ground point cloud of multigroup lane line acquisition, error can be reduced.
As shown in figure 3, the camera calibration method of the present invention comprises the following steps：
Step S21, according to above-mentioned ground level approximating method fitting generation ground level.
Step S22, the height according to the ground level for being fitted generation, generate the nominal data of camera.
Using shot by camera to image reduce during the object in space, it is assumed that the image that shot by camera is arrived
There is a kind of following simple linear relationship between the object in three dimensions：[as]=M [thing].Wherein, matrix M can be seen
Into the geometrical model for being video camera imaging.Parameter in M is exactly camera parameters.Generally, these parameters be will by experiment with
Calculate to obtain.This process for solving parameter is known as camera calibration.
When it is determined that during ground level, you can the nominal data of camera is generated based on the ground level, to cause each camera
The image mosaic position of shooting is more accurate, and error is smaller.To those skilled in the art, the demarcation number of camera is generated
According to belonging to ripe prior art, therefore it will not be repeated here.
Reference picture 4, ground level fitting system 1 of the invention includes acquisition module 11, judge module 12 and fitting module 13.
Acquisition module 11 is used for the public visual field based on the multi-path camera being arranged on vehicle and obtains some groups of vehicles two
The lane line of side, the preceding road camera on front, rear, left side and right side of the multi-path camera including being located at vehicle respectively,
Way of escape camera, left camera and right wing camera；The preceding road camera is imaged with the left camera, the preceding road
Head is taken the photograph with the right wing camera, the way of escape camera and the left camera, the way of escape camera with the right wing
As head has the public visual field respectively.
In ADAS systems, usually using four-way CCD camera, Chinese herbaceous peony, Che Hou, a vehicle body left side, the vehicle body right side are separately mounted to, point
Vehicle front, rear, left side and the scene on right side Yong Yu not observed.The scene splicing in the four-way CCD camera visual field can form 360
That spends looks around panorama.Wherein, existing between adjacent camera between the public visual field, relative camera does not have the public visual field.
The lane line of acquisition vehicle body left front is for example needed, the public visual field of preceding road camera and left camera can be used；Need
The lane line of vehicle body left back is obtained, the combination of left camera and way of escape camera can be used.
As shown in Fig. 2 the acquisition principle of lane line is as follows：The camera A and straight line a in its visual field determines that a space is put down
Face, the camera B and straight line b in its visual field determines a space plane, and the intersection of the two planes is the required car extracted
Diatom.Wherein, due to the collimation error, straight line a and straight line b are located in error plane, and lane line is then to be located at true ground level
On.
Based on above-mentioned principle, the lane line for obtaining vehicle both sides based on the multi-path camera being arranged on vehicle includes following
Step：
11) any straight line in preceding road camera and its visual field determines the first plane.
12) any straight line in way of escape camera and its visual field determines the second plane.
13) any straight line in left camera and its visual field determines the 3rd plane.
14) any straight line in right wing camera and its visual field determines fourth plane.
15) intersection or second plane by first plane and the 3rd plane and the friendship of the 3rd plane
Line determines left-hand lane line.
Specifically, the intersection of first plane and the 3rd plane is the lane line of left front；Described second is flat
The intersection of face and the 3rd plane is the lane line of left back.
16) by first plane and the fourth plane intersection or second plane and the fourth plane friendship
Line determines right-hand lane line.
Specifically, the intersection of first plane and the fourth plane is the lane line of right front；Described second is flat
The intersection of face and the fourth plane is the lane line of right back.
Judge module 12 is connected with acquisition module 11, the lane line for judging acquired vehicle both sides it is whether parallel and
In same level.
Under normal circumstances, vehicle is always travelled along lane line, and the lane line on road surface is usually parallel.Therefore,
Need to judge whether lane line is parallel in the horizontal direction, if not parallel, remove the lane line；For parallel lane line also
Need in the vertical direction to judge whether to be in same level, if being not at same level, remove the lane line.This is
Because being easily collected into the wrong data such as curb when collecting rectilinear stretch data, due to curb with ground generally not same
In level height, using curb data ground level can be caused to be fitted mistake；And ground level fitting mistake can cause the mark of camera
Fixed number evidence it is inaccurate, the splicing dislocation between the image that each camera is shot is easily caused, for example, left and preceding road camera are adjusted
After whole, the image mosaic of left and the way of escape can misplace.
Fitting module 13 is connected with judge module 12, for generating ground based on the lane line parallel and in same level
Millet cake cloud, and it is fitted generation ground level.
Every straight line can be uniquely determined by two points in space.Therefore, two on a lane line correspondence ground level
Individual point.Three points on the same line can not uniquely determine a plane.Therefore, as long as gathering two different positions in theory
The lane line put can just calculate the space equation of ground level.
Point cloud refers to the set of many points in three dimensions.Using cloud data can fit straight line, plane or its
His geometry.Accordingly, ground point cloud refers to the set of many points on ground.Generation can be fitted using ground cloud data
Ground level.
In the present invention, ground level is fitted according to the ground point cloud of multigroup lane line acquisition, error can be reduced.
As shown in figure 5, the camera calibration system of the present invention, which includes above-mentioned ground level, is fitted system 1 and adjusting module 2.
Ground level fitting system 1 is used for based on the multi-path camera fitting generation ground level being arranged on vehicle.
Adjusting module 2 is connected with ground level fitting system 1, for the height according to the ground level for being fitted generation, generation
The nominal data of camera.
Using shot by camera to image reduce during the object in space, it is assumed that the image that shot by camera is arrived
There is a kind of following simple linear relationship between the object in three dimensions：[as]=M [thing].Wherein, matrix M can be seen
Into the geometrical model for being video camera imaging.Parameter in M is exactly camera parameters.Generally, these parameters be will by experiment with
Calculate to obtain.This process for solving parameter is known as camera calibration.
When it is determined that during ground level, you can the height based on the ground level generates the nominal data of camera, to cause each
The image mosaic position that camera is shot is more accurate, and error is smaller.To those skilled in the art, camera is generated
Nominal data belongs to ripe prior art, therefore will not be repeated here.
As shown in fig. 6, the present invention also provides a kind of car-mounted terminal, it is connected with multi-path camera, including above-mentioned shooting leader
Determine system, the multi-path camera includes preceding road camera, the way of escape respectively positioned at the front of vehicle, rear, left side and right side
Camera, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera with
The right wing camera, the way of escape camera and the left camera, the way of escape camera and the right wing camera
There is the public visual field respectively.The car-mounted terminal of the present invention can generate the nominal data of camera so that what each camera was shot
Image mosaic position is more accurate, and error is smaller.
As shown in fig. 7, the present invention also provides a kind of vehicle 7, including above-mentioned connected car-mounted terminal 71 and multi-path camera
72, the multi-path camera includes being located at the preceding road camera on front, rear, left side and the right side of vehicle, way of escape shooting respectively
Head, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera with it is described
Right wing camera, the way of escape camera are distinguished with the left camera, the way of escape camera and the right wing camera
With the public visual field.
In summary, ground level of the invention fitting, camera calibration method and system, car-mounted terminal, vehicle are based on four
Road camera collection pavement image, and the track data in pavement image are obtained, ensured by rejecting the track data of mistake
The accuracy of ground level fitting；Measured fitting ground level, generates the nominal data of camera so that each camera is shot
Image mosaic position it is more accurate, error is smaller；Real-time is good, and contentedly the time of plane fitting and camera calibration will
Ask.So, the present invention effectively overcomes various shortcoming of the prior art and has high industrial utilization.
The above-described embodiments merely illustrate the principles and effects of the present invention, not for the limitation present invention.It is any ripe
Know the personage of this technology all can carry out modifications and changes under the spirit and scope without prejudice to the present invention to above-described embodiment.Cause
This, those of ordinary skill in the art is complete without departing from disclosed spirit and institute under technological thought such as
Into all equivalent modifications or change, should by the present invention claim be covered.
Claims (8)
1. a kind of ground level approximating method, it is characterised in that：Comprise the following steps：
The public visual field based on the multi-path camera being arranged on vehicle obtains the lane line of some groups of vehicle both sides, the multichannel
Camera includes preceding road camera, way of escape camera, left shooting respectively positioned at the front of vehicle, rear, left side and right side
Head and right wing camera；The preceding road camera and the left camera, the preceding road camera and the right wing camera,
The way of escape camera has public regard respectively with the left camera, the way of escape camera and the right wing camera
It is wild；
Judge the lane line of acquired vehicle both sides whether parallel and in same level；
Ground point cloud is generated based on the lane line parallel and in same level, and is fitted generation ground level.
2. ground level approximating method according to claim 1, it is characterised in that：Based on the multichannel shooting being arranged on vehicle
The lane line of head acquisition vehicle both sides comprises the following steps：
Any straight line in preceding road camera and its visual field determines the first plane；
Any straight line in way of escape camera and its visual field determines the second plane；
Any straight line in left camera and its visual field determines the 3rd plane；
Any straight line in right wing camera and its visual field determines fourth plane；
Determined by the intersection or second plane of first plane and the 3rd plane and the intersection of the 3rd plane
Left-hand lane line；
Determined by the intersection of the intersection or second plane and the fourth plane of first plane and the fourth plane
Right-hand lane line.
3. a kind of camera calibration method, it is characterised in that：Comprise the following steps：
Ground level approximating method fitting generation ground level according to claim 1 or 2；
According to the height for the ground level for being fitted generation, the nominal data of camera is generated.
4. a kind of ground level is fitted system, it is characterised in that：Including acquisition module, judge module and fitting module；
The acquisition module is used for the public visual field based on the multi-path camera being arranged on vehicle and obtains some groups of vehicle both sides
Lane line, the multi-path camera include preceding road camera respectively positioned at the front of vehicle, rear, left side and right side, after
Road camera, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera
Imaged with the right wing camera, the way of escape camera and the left camera, the way of escape camera and the right wing
Head has the public visual field respectively；
Whether the judge module is used for the lane line for judging acquired vehicle both sides parallel and in same level；
The fitting module is used to generate ground point cloud based on the lane line parallel and in same level, and is fitted generation ground
Plane.
5. ground level according to claim 4 is fitted system, it is characterised in that：The acquisition module passes through following steps base
The lane line of vehicle both sides is obtained in the multi-path camera being arranged on vehicle：
Any straight line in preceding road camera and its visual field determines the first plane；
Any straight line in way of escape camera and its visual field determines the second plane；
Any straight line in left camera and its visual field determines the 3rd plane；
Any straight line in right wing camera and its visual field determines fourth plane；
Determined by the intersection or second plane of first plane and the 3rd plane and the intersection of the 3rd plane
Left-hand lane line；
Determined by the intersection of the intersection or second plane and the fourth plane of first plane and the fourth plane
Right-hand lane line.
6. a kind of camera calibration system, it is characterised in that：Including the ground level fitting system and tune described in claim 4 or 5
Mould preparation block；
The ground level fitting system is used for based on the multi-path camera fitting generation ground level being arranged on vehicle；
The adjusting module is used for the height according to the ground level for being fitted generation, generates the nominal data of camera.
7. a kind of car-mounted terminal, it is characterised in that：It is connected with multi-path camera, including the camera calibration described in claim 6
System, the multi-path camera is taken the photograph including preceding road camera respectively positioned at the front of vehicle, rear, left side and right side, the way of escape
As head, left camera and right wing camera；The preceding road camera and the left camera, the preceding road camera and institute
State right wing camera, the way of escape camera and the left camera, the way of escape camera and the right wing camera point
Ju You not the public visual field.
8. a kind of vehicle, it is characterised in that：Including the car-mounted terminal and multi-path camera described in claim 7, the multichannel is taken the photograph
As head includes preceding road camera, way of escape camera, left camera respectively positioned at the front of vehicle, rear, left side and right side
With right wing camera；The preceding road camera and the left camera, the preceding road camera and the right wing camera, institute
State way of escape camera and the left camera, the way of escape camera and the right wing camera has the public visual field respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710209423.8A CN107145825A (en) | 2017-03-31 | 2017-03-31 | Ground level fitting, camera calibration method and system, car-mounted terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710209423.8A CN107145825A (en) | 2017-03-31 | 2017-03-31 | Ground level fitting, camera calibration method and system, car-mounted terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107145825A true CN107145825A (en) | 2017-09-08 |
Family
ID=59784197
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710209423.8A Pending CN107145825A (en) | 2017-03-31 | 2017-03-31 | Ground level fitting, camera calibration method and system, car-mounted terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107145825A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108162866A (en) * | 2017-12-21 | 2018-06-15 | 宁波吉利汽车研究开发有限公司 | A kind of lane recognition system and method based on Streaming Media external rearview mirror system |
| CN110555886A (en) * | 2018-05-31 | 2019-12-10 | 杭州海康威视数字技术股份有限公司 | Vehicle-mounted camera external parameter calibration method and device, electronic equipment and storage medium |
| CN110570475A (en) * | 2018-06-05 | 2019-12-13 | 上海商汤智能科技有限公司 | vehicle-mounted camera self-calibration method and device and vehicle driving method and device |
| CN110827337A (en) * | 2018-08-08 | 2020-02-21 | 深圳地平线机器人科技有限公司 | Method and device for determining posture of vehicle-mounted camera and electronic equipment |
| CN110858405A (en) * | 2018-08-24 | 2020-03-03 | 北京市商汤科技开发有限公司 | Attitude estimation method, device and system of vehicle-mounted camera and electronic equipment |
| CN111462243A (en) * | 2019-01-22 | 2020-07-28 | 上海欧菲智能车联科技有限公司 | Vehicle-mounted streaming media rearview mirror calibration method, system and device |
| CN111695379A (en) * | 2019-03-13 | 2020-09-22 | 长沙智能驾驶研究院有限公司 | Ground segmentation method and device based on stereoscopic vision, vehicle-mounted equipment and storage medium |
-
2017
- 2017-03-31 CN CN201710209423.8A patent/CN107145825A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108162866A (en) * | 2017-12-21 | 2018-06-15 | 宁波吉利汽车研究开发有限公司 | A kind of lane recognition system and method based on Streaming Media external rearview mirror system |
| CN110555886A (en) * | 2018-05-31 | 2019-12-10 | 杭州海康威视数字技术股份有限公司 | Vehicle-mounted camera external parameter calibration method and device, electronic equipment and storage medium |
| CN110555886B (en) * | 2018-05-31 | 2021-09-17 | 杭州海康威视数字技术股份有限公司 | Vehicle-mounted camera external parameter calibration method and device, electronic equipment and storage medium |
| CN110570475A (en) * | 2018-06-05 | 2019-12-13 | 上海商汤智能科技有限公司 | vehicle-mounted camera self-calibration method and device and vehicle driving method and device |
| CN110827337A (en) * | 2018-08-08 | 2020-02-21 | 深圳地平线机器人科技有限公司 | Method and device for determining posture of vehicle-mounted camera and electronic equipment |
| CN110827337B (en) * | 2018-08-08 | 2023-01-24 | 深圳地平线机器人科技有限公司 | Method and device for determining posture of vehicle-mounted camera and electronic equipment |
| CN110858405A (en) * | 2018-08-24 | 2020-03-03 | 北京市商汤科技开发有限公司 | Attitude estimation method, device and system of vehicle-mounted camera and electronic equipment |
| CN111462243A (en) * | 2019-01-22 | 2020-07-28 | 上海欧菲智能车联科技有限公司 | Vehicle-mounted streaming media rearview mirror calibration method, system and device |
| CN111695379A (en) * | 2019-03-13 | 2020-09-22 | 长沙智能驾驶研究院有限公司 | Ground segmentation method and device based on stereoscopic vision, vehicle-mounted equipment and storage medium |
| CN111695379B (en) * | 2019-03-13 | 2023-09-26 | 长沙智能驾驶研究院有限公司 | Ground segmentation method and device based on stereoscopic vision, vehicle-mounted equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107145825A (en) | Ground level fitting, camera calibration method and system, car-mounted terminal | |
| CN103413313B (en) | The binocular vision navigation system of electrically-based robot and method | |
| CN110207650A (en) | Automobile-used highway height-limiting frame height measurement method and device based on binocular vision | |
| CN103473762B (en) | A kind of method for detecting lane lines and device | |
| CN105300403B (en) | A kind of vehicle mileage calculating method based on binocular vision | |
| CN101789123B (en) | Method for creating distance map based on monocular camera machine vision | |
| CN109263637B (en) | Collision prediction method and device | |
| CN104902261B (en) | Apparatus and method for the road surface identification in low definition video flowing | |
| CN202071799U (en) | Intelligent automobile safety warning device | |
| CA2967872C (en) | Double stereoscopic sensor | |
| CN108877269A (en) | A kind of detection of intersection vehicle-state and V2X broadcasting method | |
| CN109657638A (en) | Barrier localization method, device and terminal | |
| CN110097591A (en) | The detection method of bow net state | |
| CN103313039B (en) | A kind of freeway tunnel entrance security prompt device on daytime and reminding method | |
| CN107767424A (en) | Scaling method, multicamera system and the terminal device of multicamera system | |
| CN104966062A (en) | Video monitoring method and device | |
| CN103692993A (en) | Binocular far infrared intelligent assistant safety driving system | |
| CN104890574A (en) | Car A column blind zone eliminating device based on machine vision | |
| CN107133587A (en) | Obstacle distance judges image generating method and device | |
| CN107146255A (en) | Panoramic picture error calibration method and device | |
| CN106570487A (en) | Method and device for predicting collision between objects | |
| CN114937249A (en) | Grid management-based vehicle-road cooperative visual perception processing method | |
| CN105096284A (en) | Method, device and system of generating road orthographic projection image | |
| CN114067287A (en) | Foreign matter identification and early warning system based on vehicle side road side data perception fusion | |
| CN104091450B (en) | A kind of video frequency velometer and video frequency speed-measuring method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170908 |
|
| RJ01 | Rejection of invention patent application after publication |