CN106289159A - The vehicle odometry method and device compensated based on range finding - Google Patents
The vehicle odometry method and device compensated based on range finding Download PDFInfo
- Publication number
- CN106289159A CN106289159A CN201610608255.5A CN201610608255A CN106289159A CN 106289159 A CN106289159 A CN 106289159A CN 201610608255 A CN201610608255 A CN 201610608255A CN 106289159 A CN106289159 A CN 106289159A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- distance
- camera
- spacing
- estimate
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C3/00—Measuring distances in line of sight; Optical rangefinders
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
- G01S17/08—Systems determining position data of a target for measuring distance only
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Traffic Control Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Measurement Of Optical Distance (AREA)
Abstract
The invention provides the vehicle odometry method compensated based on range finding, the method includes: use camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；Utilize gyroscope to measure the angle of pitch of in-vehicle camera, utilize tape measure in-vehicle camera true altitude overhead；N number of different position will be parked in respectively with reference to vehicle, utilize laser range finder, monocular vision vehicle odometry method obtain respectively be parked in N number of diverse location with reference to vehicle actual distance collection, estimate distance set, and obtain distance difference collection；Gather video image, use vehicle detecting algorithm to detect target vehicle from video image；Utilize monocular vision vehicle odometry method acquisition target vehicle estimates spacing, obtains the compensated distance value corresponding with estimating spacing；According to estimating spacing and compensated distance value, calculate the spacing of target vehicle and export.Compared with prior art, the present invention by range finding compensate improve vehicle odometry accuracy and range finding speed fast.
Description
Technical field
The present invention relates to image procossing, video monitoring and intelligent transportation, particularly to vehicle odometry method and device.
Background technology
Ensure that the automaticity of car steering, comfortableness and safety are the targets that intelligent vehicle is pursued always, and wherein car
Safety the most extremely pay attention to.The security performance of vehicle is mainly reflected in security system, danger early warning system now
The aspects such as system, CAS.These mobile units under control of the computer, are carried to driver by the mode such as sound, image
For auxiliary driving information, and can automatically or semi-automatically carry out wagon control, thus the accident that is effectively prevented occurs.Vehicle odometry
Technology is one of emphasis of automobile collision preventing technology.
At present, the method for vehicle odometry mainly has ultrasonic ranging, millimetre-wave radar range finding, range finding compensates and vision is surveyed
Away from etc. method.Due to the decay that square is directly proportional of ultrasonic energy and distance, therefore ultrasonic ranging is only applicable to relatively short distance.
Radar range finder holds easy electromagnetic wave interference, can cause the maloperation of vehicle.Although range finding compensates instrument measuring speed quickly, but survey
Accuracy of measurement is the highest.Visual token is the measurement side that image is used as detection and the transmission means of information or carrier
Method, has the feature such as good stability, untouchable measurement, and therefore the distance-finding method of view-based access control model has obtained great concern.
The Chinese invention patent application of Publication No. CN104392629A discloses a kind of method detecting spacing, this bag
Include: the image of tail-light before gathering；Determine described front tail-light positional information in described image；According to described positional information
Calculate the spacing between this car and front truck.But said method degree of accuracy is relatively low.The middle promulgated by the State Council of Publication No. CN105488454A
Bright patent application discloses the detection of a kind of front vehicles based on monocular vision and distance-finding method, by histogram equalization, classification
The methods such as device tentatively obtain the rectangular area of information of vehicles, utilize priori to remove flase drop, utilize vehicle bottom shade to determine
The accurate location of vehicle, carries out vehicle odometry using vehicle location and lane detection result as prior information.But, above-mentioned side
Method operand is relatively big, runs the time longer.
In sum, high vehicle odometry method and device is spent fast and accurately in the urgent need to proposition is a kind of at present.
Summary of the invention
In view of this, present invention is primarily targeted at and realize rapid vehicle range finding, and ranging accuracy is higher.
For reaching above-mentioned purpose, according to the first aspect of the invention, it is provided that the vehicle odometry side compensated based on range finding
Method, the method includes:
First step, uses camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Second step, utilizes gyroscope to measure the angle of pitch of in-vehicle camera, utilizes tape measure in-vehicle camera overhead
True altitude；
Third step, by being parked in N number of different position respectively with reference to vehicle, utilizes laser range finder, monocular vision car
Distance-finding method obtain respectively be parked in N number of diverse location with reference to vehicle actual distance collection, estimate distance set, and obtain distance
Difference collection；
4th step, gathers video image, uses vehicle detecting algorithm to detect target vehicle from video image；
5th step, utilize monocular vision vehicle odometry method acquisition target vehicle estimates spacing, obtains and estimate car
Away from corresponding compensated distance value；And
6th step, calculates estimate spacing and compensated distance value and value, using be worth as the spacing of target vehicle defeated
Go out.
Described third step farther includes:
Different distance selecting step, the N number of different position being parked in respectively in in-vehicle camera visual range with reference to vehicle
Put；
Actual distance collection obtaining step, utilizes laser range finder to measure the reference vehicle being in N number of different position respectively
Actual distance collection L={l1, l2..., lN}；
Estimate distance set calculation procedure, use monocular vision vehicle odometry method to obtain respectively and be in N number of different position
Reference vehicle estimate distance set D={d1, d2..., dN}；
Distance difference collection obtaining step, calculates actual distance collection L={l1, l2..., lNAnd estimate distance set D={d1,
d2..., dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ..., N.
Described 5th step farther includes:
Estimating spacing calculation procedure, use monocular vision vehicle odometry method S330 calculating target vehicle estimates spacing
dE；
Compensated distance value obtaining step, is estimating distance set D={d1, d2..., dNSearch in } and estimate spacing dEConnect most
Near di, from distance difference collection Δ D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ di, and will
ΔdiAs compensated distance value dC, i.e. dC=Δ di。
Described monocular vision vehicle odometry method farther includes:
Camera focus obtaining step, utilizes the focal distance f in the internal reference of camera and the physical size dy on y-axis direction, calculates
Camera focus
Disappearance vertical coordinate calculation procedure, calculates the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0For in image
The vertical coordinate of heart point, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate calculation procedure, utilizes vehicle detecting algorithm to detect the rectangular area of vehicle from image,
Obtain value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；
Estimating distance calculation procedure, calculate vehicle estimates distanceHcamFor
Camera true altitude overhead.
According to another aspect of the present invention, it is provided that the vehicle distance measurement device compensated based on range finding, this device includes:
Camera internal reference computing module (1), for using camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Camera extrinsic acquisition module (2), including gyroscope (21) and scale (22)；
Distance difference collection acquisition module (3), for by being parked in N number of different position respectively with reference to vehicle, utilizing laser
Diastimeter, monocular vision vehicle odometry method obtain respectively and are parked in the actual distance collection of N number of diverse location reference vehicle, estimate
Distance set, and obtain distance difference collection；
Target vehicle detection module (4), is used for gathering video image, uses vehicle detecting algorithm to detect from video image
Go out target vehicle；
Compensated distance value acquisition module (5), for utilizing monocular vision vehicle odometry method to obtain estimating of target vehicle
Spacing, obtains the compensated distance value corresponding with estimating spacing；And
The spacing computing module (6) of target vehicle, for calculating estimate spacing and compensated distance value and value, will be with value work
For the spacing of target vehicle and export.
Described gyroscope (21) is for measuring the angle of pitch of in-vehicle camera；Described scale (22) be used for measuring in-vehicle camera from
The true altitude on ground.
Described distance difference collection acquisition module (3) farther includes:
Different distance chooses module (31), for N number of by be parked in respectively in in-vehicle camera visual range with reference to vehicle
Different positions；
Actual distance collection acquisition module (32), is in N number of different position for utilizing laser range finder to measure respectively
Actual distance collection L={l with reference to vehicle1, l2..., lN}；
Estimate distance set computing module (33), be used for using monocular vision vehicle odometry module (330) to obtain respectively and be in N
The reference vehicle of individual different position estimate distance set D={d1, d2..., dN}；
Distance difference collection acquisition module (34), is used for calculating actual distance collection L={l1, l2..., lNAnd estimate distance set D
={ d1, d2..., dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ..., N.
Described target vehicle range finder module (5) farther includes:
Estimate spacing computing module (51), for using monocular vision vehicle odometry method S330 to calculate the pre-of target vehicle
Estimate spacing dE；
Compensated distance value acquisition module (52), for estimating distance set D={d1, d2..., dNSearch in } and estimate car
Away from dEImmediate di, from distance difference collection Δ D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ
di, and by Δ diAs compensated distance value dC, i.e. dC=Δ di。
Described monocular vision vehicle odometry module (330) farther includes:
Camera focus acquisition module (331), the focal distance f in the internal reference utilizing camera and the physics chi on y-axis direction
Very little dy, calculates camera focus
Disappearance vertical coordinate computing module (332), for calculating the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0
For the vertical coordinate of image center, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate computing module (333), for utilizing vehicle detecting algorithm to detect vehicle from image
Rectangular area, obtains value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；
Estimate distance calculation module (334), estimate distance for calculate vehicle
HcamFor camera true altitude overhead.
Compared with existing vehicle odometry technology, the vehicle odometry method and device based on range finding compensation of the present invention is passed through
The range finding compensation method using laser ranging and visual token to combine, can quickly measure the distance of target vehicle, and essence
Exactness is higher.
Accompanying drawing explanation
Fig. 1 shows the flow chart of the vehicle odometry method compensated based on range finding according to the present invention.
Fig. 2 shows the frame diagram of the vehicle distance measurement device compensated based on range finding according to the present invention.
Detailed description of the invention
For making your auditor can further appreciate that the structure of the present invention, feature and other purposes, in conjunction with appended preferable reality
Executing example and describe in detail as follows, illustrated preferred embodiment is merely to illustrate technical scheme, and the non-limiting present invention.
Fig. 1 gives the flow chart of the vehicle odometry method compensated based on range finding according to the present invention.As it is shown in figure 1, press
Vehicle odometry method based on range finding compensation according to the present invention includes:
First step S1, uses camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Second step S2, utilizes gyroscope to measure the angle of pitch of in-vehicle camera, utilizes tape measure in-vehicle camera overhead
True altitude；
Third step S3, by being parked in N number of different position respectively with reference to vehicle, utilizes laser range finder, monocular vision
Vehicle odometry method obtain respectively be parked in N number of diverse location with reference to vehicle actual distance collection, estimate distance set, and obtain away from
Deviation value collection；
4th step S4, gathers video image, uses vehicle detecting algorithm to detect target vehicle from video image；
5th step S5, utilize monocular vision vehicle odometry method acquisition target vehicle estimates spacing, obtains and estimate
The compensated distance value that spacing is corresponding；And
6th step S6, calculates estimate spacing and compensated distance value and value, using be worth as target vehicle spacing also
Output.
Wherein, in described first step S1, camera internal reference calibration algorithm is prior art, can pass through Zhang Zhengyou standardizition
Or the innovatory algorithm demarcated realizes, and such as " Liu Yan, Li Tengfei. the linguistic term to Zhang Zhengyou camera calibration method, " optics skill
Art ", 2014 (6): 565-570 ".The internal reference of described acquisition includes the center point coordinate (u of focal distance f, image0,v0)。
Described second step S2 measures Angle Method by existing gyroscope, the camera that measurement is loaded on vehicle
The angle of pitch is θ.Described scale is the conventional scale that can measure height, and measuring camera true altitude overhead is Hcam。
Described third step S3 farther includes:
Different distance selecting step S31 is N number of different by be parked in respectively in in-vehicle camera visual range with reference to vehicle
Position；
Actual distance collection obtaining step S32, utilizes laser range finder to measure the reference car being in N number of different position respectively
Actual distance collection L={l1, l2..., lN}；
Estimate distance set calculation procedure S33, use monocular vision vehicle odometry method to obtain respectively and be in N number of different position
The reference vehicle put estimate distance set D={d1, d2..., dN}；
Distance difference collection obtaining step S34, calculates actual distance collection L={l1, l2..., lNAnd estimate distance set D=
{d1, d2..., dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ..., N.
Wherein, described laser range finder is prior art.
Described different distance selecting step S31 further for will be parked in respectively with reference to vehicle distance in-vehicle camera 10～
N number of different position in the range of 100m.
Described N ∈ [5,50].Preferably, N ∈ [10,20].Such as, N elects 15 as.
Described estimate distance calculation procedure S33 in monocular vision vehicle odometry method S330 farther include:
Camera focus obtaining step S331, utilizes the focal distance f in the internal reference of camera and the physical size dy on y-axis direction,
Calculate camera focus
Disappearance vertical coordinate calculation procedure S332, calculates the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0For figure
The vertical coordinate of inconocenter point, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate calculation procedure S333, utilizes vehicle detecting algorithm to detect the rectangle region of vehicle from image
Territory, obtains value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；
Estimating distance calculation procedure S334, calculate vehicle estimates distance
HcamFor camera true altitude overhead.
Wherein, in described camera focus obtaining step S331, the physical size dy on y-axis direction is that camera photo-sensitive cell is solid
There is size.
In described vehicle intersection vertical coordinate calculation procedure S333, vehicle detecting algorithm is the vehicle inspection of existing view-based access control model
Survey method.Such as, " Yang Xianfeng, Yang Yan. a kind of high efficiency vehicle detection method based on HOG-LBP. " computer engineering ", 2014,
40(09):210-214”。
Described 4th step S4 gathers video image by in-vehicle camera.
In described 4th step S4, vehicle detecting algorithm is the vehicle checking method of existing view-based access control model, can use with
Vehicle detecting algorithm similar in vehicle intersection vertical coordinate calculation procedure S333.
Described 5th step S5 farther includes:
Estimating spacing calculation procedure S51, use monocular vision vehicle odometry method S330 calculating target vehicle estimates car
Away from dE；
Compensated distance value obtaining step S52, is estimating distance set D={d1, d2..., dNSearch in } and estimate spacing dE?
Close di, from distance difference collection Δ D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ di, and
By Δ diAs compensated distance value dC, i.e. dC=Δ di。
The described monocular vision vehicle odometry method estimated in spacing calculation procedure S51 uses and estimates distance calculation procedure
Monocular vision vehicle odometry method S330 identical in S33.
Described compensated distance value obtaining step S52 is estimating distance set D={d1, d2..., dNSearch in } and estimate car
Away from dEImmediate diDistance set D={d is estimated particularly as follows: calculate respectively1, d2..., dNEach in } estimate distance diWith dEExhausted
To difference di-dE|, | di-dE| the d that value is minimumiIt is and estimates spacing dEImmediate di。
Described 6th step S6 estimates spacing d for calculating furtherEWith compensated distance value dCAnd value dT=dE+dC, will be with
Value dTAs the spacing of target vehicle and export.
Fig. 2 gives the frame diagram of the camera calibration device compensated based on range finding according to the present invention.As in figure 2 it is shown, press
Camera calibration device based on range finding compensation according to the present invention includes:
Camera internal reference computing module 1, for using camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Camera extrinsic acquisition module 2, including gyroscope 21 and scale 22；
Distance difference collection acquisition module 3, for by being parked in N number of different position respectively with reference to vehicle, utilizing Laser Measuring
Distance meter, monocular vision vehicle odometry method obtain respectively be parked in N number of diverse location with reference to vehicle actual distance collection, estimate away from
From collection, and obtain distance difference collection；
Target vehicle detection module 4, is used for gathering video image, uses vehicle detecting algorithm to detect from video image
Target vehicle；
Compensated distance value acquisition module 5, for utilize monocular vision vehicle odometry method obtain target vehicle estimate car
Away from, obtain the compensated distance value corresponding with estimating spacing；And
The spacing computing module 6 of target vehicle, for calculating estimate spacing and compensated distance value and value, will be worth conduct
The spacing of target vehicle also exports.
Described gyroscope 21 is for measuring the angle of pitch of in-vehicle camera；Described scale 22 is used for measuring in-vehicle camera overhead
True altitude.
Wherein, in described camera internal reference computing module 1, camera internal reference calibration algorithm is prior art, can pass through Zhang Zhengyou
The innovatory algorithm of standardizition or demarcation realizes, such as " Liu Yan, Li Tengfei. the linguistic term to Zhang Zhengyou camera calibration method,
" optical technology ", 2014 (6): 565-570 ".The internal reference of described acquisition includes the center point coordinate (u of focal distance f, image0,v0)。
In described Camera extrinsic acquisition module 2, the method for angle measured by gyroscope 21 is prior art, measures and is loaded in car
The angle of pitch of the camera on is θ.Described scale 22 is the conventional scale that can measure height, measures camera overhead true
Real height is Hcam。
Described compensated distance value acquisition module 3 farther includes:
Different distance chooses module 31, for by with reference to vehicle be parked in respectively in in-vehicle camera visual range N number of not
Same position；
Actual distance collection acquisition module 32, for utilizing laser range finder to measure the ginseng being in N number of different position respectively
Examine the actual distance collection L={l of vehicle1, l2..., lN}；
Estimate distance set computing module 33, be used for using monocular vision vehicle odometry module 330 obtain respectively be in N number of not
With position reference vehicle estimate distance set D={d1, d2..., dN}；
Distance difference collection acquisition module 34, is used for calculating actual distance collection L={l1, l2..., lNAnd estimate distance set D=
{d1, d2..., dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ,=, N.
Wherein, described laser range finder is prior art.
Described different distance choose module 31 further for will be parked in respectively with reference to vehicle distance in-vehicle camera 10～
N number of different position in the range of 100m.
Described N ∈ [5,50].Preferably, N ∈ [10,20].Such as, N elects 15 as.
Described estimate monocular vision vehicle odometry module 330 in distance calculation module 33 and farther include:
Camera focus acquisition module 331, the focal distance f in the internal reference utilizing camera and the physical size on y-axis direction
Dy, calculates camera focus
Disappearance vertical coordinate computing module 332, for calculating the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0For
The vertical coordinate of image center, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate computing module 333, for utilizing vehicle detecting algorithm to detect the square of vehicle from image
Shape region, obtains value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；
Estimate distance calculation module 334, estimate distance for calculate vehicle
HcamFor camera true altitude overhead.
Wherein, in described camera focus acquisition module 331, the physical size dy on y-axis direction is that camera photo-sensitive cell is intrinsic
Size.
Described vehicle detecting algorithm is the vehicle checking method of existing view-based access control model.Such as, " Yang Xianfeng, Yang Yan. a kind of
High efficiency vehicle detection method based on HOG-LBP. " computer engineering ", 2014,40 (09): 210-214 ".
Described target vehicle detection module 4 gathers video image by in-vehicle camera.
In described target vehicle detection module 4, vehicle detecting algorithm is the vehicle checking method of existing view-based access control model, can
To use the vehicle detecting algorithm similar with vehicle intersection vertical coordinate computing module 333.
Described compensated distance value acquisition module 5 farther includes:
Estimate spacing computing module 51, for using monocular vision vehicle odometry method S330 to calculate estimating of target vehicle
Spacing dE；
Compensated distance value acquisition module 52, for estimating distance set D={d1, d2..., dNSearch in } and estimate spacing
dEImmediate di, from distance difference collection D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ di,
And by Δ diAs compensated distance value dC, i.e. dC=Δ di。
The described monocular vision vehicle odometry module estimated in spacing computing module 51 uses and estimates distance calculation module
Monocular vision vehicle odometry module 330 identical in 33.
Described compensated distance value acquisition module 52 is estimating distance set D={d1, d2..., dNSearch in } and estimate spacing
dEImmediate diDistance set D={d is estimated particularly as follows: calculate respectively1, d2..., dNEach in } estimate distance diWith dEAbsolute
Difference | di-dE|, | di-dE| the d that value is minimumiIt is and estimates spacing dEImmediate di。
The spacing computing module 6 of described target vehicle estimates spacing d for calculating furtherEWith compensated distance value dCAnd value
dT=dE+dC, will be with value dTAs the spacing of target vehicle and export.
Compared with existing vehicle odometry technology, the vehicle odometry method and device based on range finding compensation of the present invention is passed through
The range finding compensation method using laser ranging and visual token to combine, can quickly measure the distance of target vehicle, and essence
Exactness is higher.
The above, only presently preferred embodiments of the present invention, it is not intended to limit protection scope of the present invention, it should
Understanding, the present invention is not limited to implementation as described herein, and the purpose that these implementations describe is to help this area
In technical staff put into practice the present invention.Any those of skill in the art are easy to without departing from spirit and scope of the invention
In the case of be further improved and perfect, therefore the present invention is only by the content of the claims in the present invention and limiting of scope
System, its be intended to contain the alternative in all spirit and scope of the invention being included in and being defined by the appended claims and etc.
Same scheme.
Claims (10)
1. the vehicle odometry method compensated based on range finding, it is characterised in that the method includes:
First step, uses camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Second step, utilizes gyroscope to measure the angle of pitch of in-vehicle camera, utilizes tape measure in-vehicle camera overhead true
Highly；
Third step, by being parked in N number of different position respectively with reference to vehicle, utilizes laser range finder, monocular vision vehicle to survey
Obtain respectively away from method and be parked in the actual distance collection of N number of diverse location reference vehicle, estimate distance set, and obtain distance difference
Collection；
4th step, gathers video image, uses vehicle detecting algorithm to detect target vehicle from video image；
5th step, utilize monocular vision vehicle odometry method acquisition target vehicle estimates spacing, obtains and estimate spacing phase
Corresponding compensated distance value；And
6th step, calculates estimate spacing and compensated distance value and value, as the spacing of target vehicle and will export with value.
2. the method for claim 1, it is characterised in that described third step includes:
Different distance selecting step, the N number of different position being parked in respectively in in-vehicle camera visual range with reference to vehicle；
Actual distance collection obtaining step, utilizes laser range finder to measure reference vehicle true being in N number of different position respectively
Real distance set L={l1, l2..., lN}；
Estimate distance set calculation procedure, use monocular vision vehicle odometry method to obtain the ginseng being in N number of different position respectively
That examines vehicle estimates distance set D={d1, d2..., dN}；
Distance difference collection obtaining step, calculates actual distance collection L={l1, l2..., lNAnd estimate distance set D={d1, d2...,
dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ..., N.
3. the method for claim 1, it is characterised in that described 5th step includes:
Estimating spacing calculation procedure, use monocular vision vehicle odometry method S330 calculating target vehicle estimates spacing dE；
Compensated distance value obtaining step, is estimating distance set D={d1, d2..., dNSearch in } and estimate spacing dEImmediate
di, from distance difference collection Δ D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ di, and by Δ di
As compensated distance value dC, i.e. dC=Δ di。
4. the method as described in claims 1 to 3, it is characterised in that described monocular vision vehicle odometry method includes: camera is burnt
Away from obtaining step, utilize the focal distance f in the internal reference of camera and the physical size dy on y-axis direction, calculate camera focus
Disappearance vertical coordinate calculation procedure, calculates the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0For image center
Vertical coordinate, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate calculation procedure, utilizes vehicle detecting algorithm to detect the rectangular area of vehicle from image, obtains
Value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；
Estimating distance calculation procedure, calculate vehicle estimates distanceHcamFor camera
True altitude overhead.
5. method as claimed in claim 2, described N ∈ [5,50].
6. the vehicle distance measurement device compensated based on range finding, it is characterised in that this device includes:
Camera internal reference computing module (1), for using camera internal reference calibration algorithm to calculate the internal reference of in-vehicle camera；
Camera extrinsic acquisition module (2), including gyroscope (21) and scale (22)；
Distance difference collection acquisition module (3), for by being parked in N number of different position respectively with reference to vehicle, utilizing laser ranging
Instrument, monocular vision vehicle odometry method obtain respectively and are parked in the actual distance collection of N number of diverse location reference vehicle, estimate distance
Collection, and obtain distance difference collection；
Target vehicle detection module (4), is used for gathering video image, uses vehicle detecting algorithm to detect mesh from video image
Mark vehicle；
Compensated distance value acquisition module (5), for utilize monocular vision vehicle odometry method obtain target vehicle estimate spacing,
Obtain the compensated distance value corresponding with estimating spacing；And
The spacing computing module (6) of target vehicle, for calculating estimate spacing and compensated distance value and value, will be worth as mesh
Mark the spacing of vehicle and export；
Wherein, described gyroscope (21) is for measuring the angle of pitch of in-vehicle camera；Described scale (22) is used for measuring in-vehicle camera
True altitude overhead.
7. device as claimed in claim 6, it is characterised in that described distance difference collection acquisition module (3) including: different distance
Choose module (31), for the N number of different position being parked in respectively in in-vehicle camera visual range with reference to vehicle；
Actual distance collection acquisition module (32), for utilizing laser range finder to measure the reference being in N number of different position respectively
The actual distance collection L={l of vehicle1, l2..., lN}；
Estimate distance set computing module (33), be used for using monocular vision vehicle odometry module (330) obtain respectively be in N number of not
With position reference vehicle estimate distance set D={d1, d2..., dN}；
Distance difference collection acquisition module (34), is used for calculating actual distance collectionWith estimate distance set D={d1,
d2..., dNDistance difference collection Δ D={ Δ d1, Δ d2..., Δ dN, wherein Δ di=li-di, i=1,2 ..., N.
8. device as claimed in claim 6, it is characterised in that described target vehicle range finder module (5) including: estimates spacing meter
Calculate module (51), for use monocular vision vehicle odometry method S330 calculate target vehicle estimate spacing dE；
Compensated distance value acquisition module (52), for estimating distance set D={d1, d2..., dNSearch in } and estimate spacing dE?
Close di, from distance difference collection Δ D={ Δ d1, Δ d2..., Δ dNObtain and d in }iCorresponding distance difference Δ di, and
By Δ diAs compensated distance value dC, i.e. dC=Δ di。
9. the device as described in claim 6～8, it is characterised in that described monocular vision vehicle odometry module (330) is further
Including:
Camera focus acquisition module (331), the focal distance f in the internal reference utilizing camera and the physical size dy on y-axis direction,
Calculate camera focus
Disappearance vertical coordinate computing module (332), for calculating the vertical coordinate y of vanishing line in imageh=v0-Fcam* tan θ, v0For figure
The vertical coordinate of inconocenter point, θ is the angle of pitch of camera；
Vehicle intersection vertical coordinate computing module (333), for utilizing vehicle detecting algorithm to detect the rectangle of vehicle from image
Region, obtains value y of the lower boundary of rectangular areab, by ybIt is set to the intersection vertical coordinate of vehicle and ground；Estimate distance and calculate mould
Block (334), estimates distance for calculate vehicleHcamFor camera overhead
True altitude.
10. device as claimed in claim 7, described N ∈ [5,50].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610608255.5A CN106289159B (en) | 2016-07-28 | 2016-07-28 | Vehicle distance measurement method and device based on distance measurement compensation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610608255.5A CN106289159B (en) | 2016-07-28 | 2016-07-28 | Vehicle distance measurement method and device based on distance measurement compensation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106289159A true CN106289159A (en) | 2017-01-04 |
| CN106289159B CN106289159B (en) | 2019-12-10 |
Family
ID=57663132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610608255.5A Active CN106289159B (en) | 2016-07-28 | 2016-07-28 | Vehicle distance measurement method and device based on distance measurement compensation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106289159B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107389026A (en) * | 2017-06-12 | 2017-11-24 | 江苏大学 | A kind of monocular vision distance-finding method based on fixing point projective transformation |
| CN110031829A (en) * | 2019-04-18 | 2019-07-19 | 北京联合大学 | A kind of targeting accuracy distance measuring method based on monocular vision |
| CN110966982A (en) * | 2018-09-28 | 2020-04-07 | 成都家有为力机器人技术有限公司 | Monocular camera ranging system and method for cleaning robot |
| CN111623776A (en) * | 2020-06-08 | 2020-09-04 | 昆山星际舟智能科技有限公司 | Method for measuring distance of target by using near infrared vision sensor and gyroscope |
| CN111829484A (en) * | 2020-06-03 | 2020-10-27 | 江西江铃集团新能源汽车有限公司 | Target distance measuring and calculating method based on vision |
| CN112254821A (en) * | 2020-10-16 | 2021-01-22 | 盈合(深圳)机器人与自动化科技有限公司 | Method and device for compensating body temperature detection precision through visual ranging |
| CN113178005A (en) * | 2021-05-26 | 2021-07-27 | 国网河南省电力公司南阳供电公司 | Efficient photographing modeling method and device for power equipment |
| CN113643355A (en) * | 2020-04-24 | 2021-11-12 | 广州汽车集团股份有限公司 | Method and system for detecting position and orientation of target vehicle and storage medium |
| US11226200B2 (en) | 2017-06-28 | 2022-01-18 | Boe Technology Group Co., Ltd. | Method and apparatus for measuring distance using vehicle-mounted camera, storage medium, and electronic device |
| CN114295099A (en) * | 2021-12-28 | 2022-04-08 | 合肥英睿系统技术有限公司 | Monocular camera-based distance measurement method, vehicle-mounted distance measurement equipment and storage medium |
| CN116363623A (en) * | 2023-01-28 | 2023-06-30 | 苏州飞搜科技有限公司 | Vehicle detection method based on millimeter wave radar and vision fusion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104792302A (en) * | 2015-04-29 | 2015-07-22 | 深圳市保千里电子有限公司 | Modeling method for measuring car distance |
| CN105291880A (en) * | 2014-06-19 | 2016-02-03 | 福特全球技术公司 | Distance to empty prediction with short term distance compensation |
| US20160098839A1 (en) * | 2006-01-04 | 2016-04-07 | Mobileye Vision Technologies Ltd | Estimating distance to an object using a sequence of images recorded by a monocular camera |
| KR20160059149A (en) * | 2014-11-18 | 2016-05-26 | 한밭대학교 산학협력단 | Road distance estimation method using image information of black box camera |
| CN105691453A (en) * | 2014-12-09 | 2016-06-22 | 株式会社万都 | Steering torque compensating apparatus and steering torque compensating method |
-
2016
- 2016-07-28 CN CN201610608255.5A patent/CN106289159B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160098839A1 (en) * | 2006-01-04 | 2016-04-07 | Mobileye Vision Technologies Ltd | Estimating distance to an object using a sequence of images recorded by a monocular camera |
| CN105291880A (en) * | 2014-06-19 | 2016-02-03 | 福特全球技术公司 | Distance to empty prediction with short term distance compensation |
| KR20160059149A (en) * | 2014-11-18 | 2016-05-26 | 한밭대학교 산학협력단 | Road distance estimation method using image information of black box camera |
| CN105691453A (en) * | 2014-12-09 | 2016-06-22 | 株式会社万都 | Steering torque compensating apparatus and steering torque compensating method |
| CN104792302A (en) * | 2015-04-29 | 2015-07-22 | 深圳市保千里电子有限公司 | Modeling method for measuring car distance |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107389026B (en) * | 2017-06-12 | 2019-10-01 | 江苏大学 | A kind of monocular vision distance measuring method based on fixed point projective transformation |
| CN107389026A (en) * | 2017-06-12 | 2017-11-24 | 江苏大学 | A kind of monocular vision distance-finding method based on fixing point projective transformation |
| US11226200B2 (en) | 2017-06-28 | 2022-01-18 | Boe Technology Group Co., Ltd. | Method and apparatus for measuring distance using vehicle-mounted camera, storage medium, and electronic device |
| CN110966982A (en) * | 2018-09-28 | 2020-04-07 | 成都家有为力机器人技术有限公司 | Monocular camera ranging system and method for cleaning robot |
| CN110031829A (en) * | 2019-04-18 | 2019-07-19 | 北京联合大学 | A kind of targeting accuracy distance measuring method based on monocular vision |
| CN113643355B (en) * | 2020-04-24 | 2024-03-29 | 广州汽车集团股份有限公司 | Target vehicle position and orientation detection method, system and storage medium |
| CN113643355A (en) * | 2020-04-24 | 2021-11-12 | 广州汽车集团股份有限公司 | Method and system for detecting position and orientation of target vehicle and storage medium |
| CN111829484A (en) * | 2020-06-03 | 2020-10-27 | 江西江铃集团新能源汽车有限公司 | Target distance measuring and calculating method based on vision |
| CN111829484B (en) * | 2020-06-03 | 2022-05-03 | 江西江铃集团新能源汽车有限公司 | Target distance measuring and calculating method based on vision |
| CN111623776A (en) * | 2020-06-08 | 2020-09-04 | 昆山星际舟智能科技有限公司 | Method for measuring distance of target by using near infrared vision sensor and gyroscope |
| CN112254821A (en) * | 2020-10-16 | 2021-01-22 | 盈合(深圳)机器人与自动化科技有限公司 | Method and device for compensating body temperature detection precision through visual ranging |
| CN113178005A (en) * | 2021-05-26 | 2021-07-27 | 国网河南省电力公司南阳供电公司 | Efficient photographing modeling method and device for power equipment |
| CN114295099A (en) * | 2021-12-28 | 2022-04-08 | 合肥英睿系统技术有限公司 | Monocular camera-based distance measurement method, vehicle-mounted distance measurement equipment and storage medium |
| CN114295099B (en) * | 2021-12-28 | 2024-01-30 | 合肥英睿系统技术有限公司 | Ranging method based on monocular camera, vehicle-mounted ranging equipment and storage medium |
| CN116363623A (en) * | 2023-01-28 | 2023-06-30 | 苏州飞搜科技有限公司 | Vehicle detection method based on millimeter wave radar and vision fusion |
| CN116363623B (en) * | 2023-01-28 | 2023-10-20 | 苏州飞搜科技有限公司 | Vehicle detection method based on millimeter wave radar and vision fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106289159B (en) | 2019-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106289159A (en) | The vehicle odometry method and device compensated based on range finding | |
| EP3792660B1 (en) | Method, apparatus and system for measuring distance | |
| US10565867B2 (en) | Detection and documentation of tailgating and speeding violations | |
| CN110531376B (en) | Obstacle detection and tracking method for port unmanned vehicle | |
| US10260889B2 (en) | Position estimation device and position estimation method | |
| CN108444390B (en) | Unmanned automobile obstacle identification method and device | |
| US8378851B2 (en) | Fusion of images in enhanced obstacle detection | |
| CN104111058B (en) | Vehicle distance measurement method and device, vehicle relative velocity measuring method and device | |
| US10591277B2 (en) | Method and system for measuring outermost dimension of a vehicle positioned at an inspection station | |
| CN103630122B (en) | Monocular vision lane line detection method and distance measurement method thereof | |
| CN108692719B (en) | Object detection device | |
| CN107577988A (en) | Realize the method, apparatus and storage medium, program product of side vehicle location | |
| JP3727400B2 (en) | Crossing detection device | |
| CN103487034A (en) | Method for measuring distance and height by vehicle-mounted monocular camera based on vertical type target | |
| US10832428B2 (en) | Method and apparatus for estimating a range of a moving object | |
| CN114724110A (en) | Target detection method and device | |
| CN107688174A (en) | A kind of image distance-finding method, system, storage medium and vehicle-mounted visually-perceptible equipment | |
| CN111998780B (en) | Target ranging method, device and system | |
| CN106127736A (en) | One is parked detection method and processor | |
| Wei et al. | Research on preceding vehicle distance measurement with monocular vision based on lane plane geometric model | |
| CN115320603A (en) | Shooting elevation angle correction method and device and vehicle | |
| CN104155470A (en) | Detecting method and system based on binocular camera for real-time vehicle speed |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |