CN109345593A - A kind of detection method and device of video camera posture - Google Patents
A kind of detection method and device of video camera posture Download PDFInfo
- Publication number
- CN109345593A CN109345593A CN201811025833.8A CN201811025833A CN109345593A CN 109345593 A CN109345593 A CN 109345593A CN 201811025833 A CN201811025833 A CN 201811025833A CN 109345593 A CN109345593 A CN 109345593A
- Authority
- CN
- China
- Prior art keywords
- video camera
- image
- candidate
- pitch angle
- end point
- 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
- 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
Abstract
The application provides a kind of detection method and device of video camera posture, this method comprises: determining coordinate position of the end point of lane line in carriageway image in image coordinate system；Wherein, the end point is the intersection point of two lane lines；The candidate pitch angle and candidate's yaw angle of video camera are determined based on the coordinate position and preset camera parameter；The confidence rate of the confidence rate for the candidate pitch angle that the carriageway image of calculating specified quantity is determined and candidate yaw angle, the maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera, the maximum candidate yaw angle of confidence rate is determined as the actual yaw angle of video camera.In technical scheme, the candidate pitch angle and candidate's yaw angle of video camera are determined by the coordinate position and camera parameter of the end point of lane line in carriageway image, the carriageway image for being then based on specified quantity determines the actual pitch angle of video camera and actual yaw angle, to determine video camera posture according to above-mentioned pitch angle and yaw angle.
Description
Technical field
This application involves field of image processing, in particular to a kind of detection method and device of video camera posture.
Background technique
Lane detection is then to extract the vehicle in image using the video camera acquisition carriageway image for being mounted on vehicle front
Road pixel characteristic point, to the lane information for obtaining vehicle front after the fitting of lane pixel characteristic point.By judging vehicle and lane
Whether the distance of the lane line on lane both sides in information determines that current vehicle travels on lane and shifts.
Under normal circumstances, the optical axis of video camera is parallel to lane line, level ground, therefore, the collected vehicle of video camera
Road image is effectively used for lane detection.
However, if when installation video camera do not adjust video camera posture, alternatively, because external force in vehicle travel process (for example,
Vehicle jolts the external force of generation) cause video camera posture to change, then the optical axis of video camera and lane line and/or level ground
There can be angle.When video camera posture not timing, will affect lane detection as a result, making the vehicle judged and two sides lane
The actual range inaccuracy of line.
Summary of the invention
In view of this, the application provides a kind of detection method and device of video camera posture, for detecting video camera posture,
In order to which subsequent adjustment is thus influence caused by avoiding because of posture not face lane detection.
Specifically, the application is achieved by the following technical solution:
A kind of detection method of video camera posture, comprising:
Determine coordinate position of the end point of lane line in carriageway image in image coordinate system；Wherein, the end point
For the intersection point of two lane lines；
Coordinate position and preset camera parameter based on the end point determine the candidate pitch angle and time of video camera
Select yaw angle；
The confidence rate of the confidence rate for the candidate pitch angle that the carriageway image of calculating specified quantity is determined and candidate yaw angle,
The maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera, the maximum candidate yaw angle of confidence rate is determined
For the actual yaw angle of video camera.
In the detection method of the video camera posture, the method also includes:
Pose adjustment is carried out to the video camera based on the actual pitch angle and actual yaw angle determined.
In the detection method of the video camera posture, the end point of lane line is sat in image in the determining carriageway image
Coordinate position in mark system, comprising:
The measurement parameter for obtaining onboard sensor, determines whether current road conditions meet preset vehicle based on the measurement parameter
Diatom detected rule；
If so, determining the difference of the lane line parameter in current lane image and the lane line parameter in former frame carriageway image
It is different whether to be less than default fractional threshold；
If so, determining coordinate position of the end point of lane line in image coordinate system from current lane image.
In the detection method of the video camera posture, the camera parameter includes that the picture centre of the carriageway image exists
The focal length of coordinate position, video camera in image coordinate system and the unit pixel size of the carriageway image；
The coordinate position based on the end point and preset camera parameter determine the candidate pitch angle of video camera
With candidate yaw angle, comprising:
The candidate pitch angle β of the video camera is calculated based on following first formula:
Wherein, u1For the abscissa of the end point, u0For the abscissa of the picture centre of the carriageway image, dx is institute
Unit pixel size of the carriageway image on abscissa direction is stated, f is the focal length of video camera；
The candidate yaw angle α of the video camera is calculated based on following second formula:
Wherein, v1For the ordinate of the end point, v0For the ordinate of the picture centre of the carriageway image, dy is institute
State carriageway image in the ordinate on unit pixel size, f be video camera focal length, β be video camera candidate pitch angle.
In the detection method of the video camera posture, the method also includes:
Based on the actual pitch angle and actual deviation angle determined, determine vehicle and two sides lane line it is practical away from
From.
A kind of detection device of video camera posture, comprising:
First determination unit, for determining coordinate bit of the end point of lane line in carriageway image in image coordinate system
It sets；Wherein, the end point is the intersection point of two lane lines；
Computing unit, for based on the end point coordinate position and preset camera parameter determine the time of video camera
Select pitch angle and candidate yaw angle；
Second determination unit, the confidence rate and time of the candidate pitch angle that the carriageway image for calculating specified quantity is determined
The maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera, most by confidence rate by the confidence rate for selecting yaw angle
Big candidate yaw angle is determined as the actual yaw angle of video camera.
In the detection device of the video camera posture, described device further include:
Adjustment unit, for carrying out appearance to the video camera based on the actual pitch angle and actual yaw angle determined
State adjustment.
In the detection device of the video camera posture, first determination unit is further used for:
The measurement parameter for obtaining onboard sensor, determines whether current road conditions meet preset vehicle based on the measurement parameter
Diatom detected rule；
If so, determining the difference of the lane line parameter in current lane image and the lane line parameter in former frame carriageway image
It is different whether to be less than default fractional threshold；
If so, determining coordinate position of the end point of lane line in image coordinate system from current lane image.
In the detection device of the video camera posture, the camera parameter includes that the picture centre of the carriageway image exists
The focal length of coordinate position, video camera in image coordinate system and the unit pixel size of the carriageway image；
The computing unit, is further used for:
The candidate pitch angle β of the video camera is calculated based on following first formula:
Wherein, u1For the abscissa of the end point, u0For the abscissa of the picture centre of the carriageway image, dx is institute
Unit pixel size of the carriageway image on abscissa direction is stated, f is the focal length of video camera；
The candidate yaw angle α of the video camera is calculated based on following second formula:
Wherein, v1For the ordinate of the end point, v0For the ordinate of the picture centre of the carriageway image, dy is institute
State carriageway image in the ordinate on unit pixel size, f be video camera focal length, β be video camera candidate pitch angle.
In the detection device of the video camera posture, described device further include:
Third determination unit, for determining vehicle and two based on the actual pitch angle and actual deviation angle determined
The actual range of side lane line.
In the embodiment of the present application, it is first determined coordinate of the end point of lane line in image coordinate system in carriageway image
Position, the coordinate position and preset camera parameter for being then based on above-mentioned end point determine the candidate pitch angle and time of video camera
The confidence of the confidence rate for the candidate pitch angle that the carriageway image for selecting yaw angle, and calculating specified quantity is determined and candidate yaw angle
The maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera by rate, by the maximum candidate yaw angle of confidence rate
It is determined as the actual yaw angle of video camera；
The time of video camera can be determined by the coordinate position and camera parameter of the end point of lane line in carriageway image
Pitch angle and candidate yaw angle are selected, the carriageway image for being then based on specified quantity determines the actual pitch angle of video camera and reality
Yaw angle, to determine video camera posture according to above-mentioned pitch angle and yaw angle.
Detailed description of the invention
Fig. 1 is a kind of schematic diagram of image coordinate system shown in the application；
Fig. 2 is a kind of schematic diagram of camera coordinate system shown in the application；
Fig. 3 is a kind of flow chart of video camera attitude detecting method shown in the application；
Fig. 4 is a kind of schematic diagram of lane line variation shown in the application；
Fig. 5 is the schematic diagram of another lane line variation shown in the application；
Fig. 6 is a kind of embodiment block diagram of video camera Attitute detecting device shown in the application；
Fig. 7 is a kind of hardware structure diagram of video camera Attitute detecting device shown in the application.
Specific embodiment
Technical solution in embodiment in order to enable those skilled in the art to better understand the present invention, and make of the invention real
The above objects, features, and advantages for applying example can be more obvious and easy to understand, with reference to the accompanying drawing to prior art and the present invention
Technical solution in embodiment is described in further detail.
The embodiment of the present application can be related to the space geometry relationship of video camera imaging during detecting video camera posture
Related content.For the principle for being illustrated more clearly that detection means in the application, three coordinates relevant to video camera imaging first
System is illustrated.Wherein, three coordinate systems include image coordinate system, camera coordinate system and world coordinate system.
It is a kind of schematic diagram of image coordinate system shown in the application referring to Fig. 1.As shown in Figure 1, image coordinate system is to scheme
As the upper left corner is that origin O0 establishes rectangular coordinate system u-v as unit of pixel, the abscissa u of any pixel point and vertical in image
Coordinate v is respectively its locating columns and line number in the picture.
Since u-v coordinate only represents the columns and line number of pixel, and the position of pixel in the picture need to also be with physics list
Position indicates, so, the image seat indicated with physical unit (such as: physical unit can be millimeter) is also established in the plane of delineation
Mark system X-Y.
Typically, the intersection point of the optical axis of video camera and the plane of delineation can be defined as to the origin O1 of X-Y coordinate, the friendship
Point is usually located at the center of the plane of delineation.In addition, X-axis is parallel with u axis, Y-axis is parallel with v axis.Exist assuming that (u0, v0) represents O1
Coordinate under u-v coordinate system, dx and dy respectively indicate physical size of each pixel in X-direction and Y direction.Then image
In each pixel can convert in the coordinate in u-v coordinate system and between the coordinate in X-Y coordinate, specific phase as detailed below
Close description.
It referring to fig. 2, is a kind of schematic diagram of camera coordinate system shown in the application.As shown in Fig. 2, camera coordinate system
Origin O point be video camera luminous point (projection centre), the Xc axis and Yc axis of camera coordinate system respectively with plane of delineation X-Y coordinate
The X-axis of system is parallel with Y-axis.The Zc axis of camera coordinate system and the optical axis coincidence of video camera, perpendicular to the plane of delineation, and and image
Plane intersects at O1 point.The distance between the origin of camera coordinate system and the origin of the plane of delineation X-Y coordinate O-O1 is
For the focal length f of video camera.
World coordinate system is introduced to describe the position of video camera, in the application, is not related to video camera entirety position
The calculating set, therefore, using the origin of camera coordinate system as the origin of world coordinate system, by video camera posture taking the photograph when correct
Camera coordinate system is as world coordinate system.
It is a kind of flow chart of video camera attitude detecting method shown in the application, as shown in figure 3, this method referring to Fig. 3
The following steps are included:
Step 301: determining coordinate position of the end point of lane line in carriageway image in image coordinate system；Wherein, institute
State the intersection point that end point is two lane lines.
Wherein, the above method can be applied to electronic equipment, which can be video camera, or dock with video camera
Smart machine.Hereafter this programme for ease of description, using video camera as executing subject.
Wherein, which is mounted on the suitable position in front side, and video camera is allowed to acquire the image of vehicle front.
In vehicle travel process, video camera can acquire the carriageway image of vehicle front in real time, then pass through lane line
Detection technique extracts the lane pixel characteristic point in image, believes the lane obtained in front of lane after the fitting of lane pixel characteristic point
Breath.As one embodiment, lane information can be the grayscale image for fitting the lane line on lane both sides.
Lane detection can be realized by the methods of Hough transformation, least square method, RANSAC, specifically can refer to existing phase
Pass technology, details are not described herein.
The lane line of both sides of the road is considered as two straight lines in the picture, two lane lines in the picture from the near to the distant to
Preceding extension, eventually intersects at a point, and does not extend further along.Therefore, the intersection point of two lane lines is also referred to as lane line
End point.It is forthright in road, and when vehicle is towards road ahead, video camera can be installed onboard suitable position, so that
End point is located at image center in video camera acquired image.In other words, so that end point is in plane of delineation u-v coordinate system
In coordinate be (u0,v0)。
When video camera posture changes, i.e. the optical axis of video camera and lane line there are angle, alternatively, optical axis with horizontally
There are angles in face, or, optical axis and lane line and level ground are respectively present angle, then the end point of lane line is in the picture
Coordinate position can change.
It referring to fig. 4 and Fig. 5, is the schematic diagram of two kinds of lane lines variation shown in the application.
As shown in figure 4, when video camera is there are when pitch angle, if the pitch angle is greater than 0, lane line meeting in carriageway image
It moves up, correspondingly, the end point of lane line also can be moved up, and the pitch angle the big, what lane line and end point moved up
Distance is bigger；If the pitch angle, less than 0, lane line can move down in carriageway image, correspondingly, the end point of lane line also can
It moves down, and pitch angle is smaller, the distance that lane line and end point move down is bigger.
As shown in figure 5, when video camera is there are when yaw angle, if the yaw angle is greater than 0, lane line meeting in carriageway image
It moves to right, correspondingly, the end point of lane line also can be toward moving right, and the yaw angle the big, what lane line and end point moved right
Distance is bigger；If the yaw angle, less than 0, lane line can move to left in carriageway image, correspondingly, the end point of lane line also can
Toward moving left, and yaw angle is smaller, and the distance that lane line and end point are moved to the left is bigger.
Based on the above principles, technical scheme determines that video camera becomes by the coordinate position of end point in image
The attitude angle (including pitch angle and deviation angle) of change.It therefore, can be into one after obtaining lane information by lane detection technology
Walk the coordinate position for determining end point of the lane line in the lane information (grayscale image), the coordinate position also lane line
Coordinate position in carriageway image.
In a kind of embodiment shown, it is contemplated that there are a series of pairs of video camera attitude detections impact because
Element need to determine the coordinate position of the end point of lane line when meeting specified requirements.
Firstly, the road conditions of vehicle driving may impact video camera attitude detection.When vehicle driving in detour or
When the road of slope, lane line is not straight line, therefore the position of the end point of lane line can also change in carriageway image.
Secondly, vehicle running state may also impact video camera attitude detection.When vehicle driving shakiness, for example,
Vehicle pitches, then so that imaging position of the lane line in carriageway image is changed, lead to the disappearance of lane line
The position of point changes.
As it can be seen that being so that the coordinate position of the end point for the lane line determined in carriageway image can accurately be used for video camera
Attitude detection need to meet condition identified below:
First, vehicle driving is on smooth forthright；
Second, vehicle running state is stablized.
To meet above-mentioned specified requirements, the measurement ginseng of video camera available onboard sensor first (such as: gyroscope)
Number, and determine whether current road conditions meet preset lane detection rule based on above-mentioned measurement parameter.Wherein, above-mentioned lane line
Detected rule defines measurement parameter range, when the measurement parameter of onboard sensor is within the scope of the policing parameter, then illustrates vehicle
Traveling on smooth forthright.As one embodiment, if above-mentioned onboard sensor is three-axis gyroscope, above-mentioned measurement ginseng
Number can be the angular acceleration that gyroscope surrounds three axis.
On the one hand, if above-mentioned measurement parameter is unsatisfactory for above-mentioned lane detection rule, it can determine that current road conditions are curved
Road or slope road, the coordinate position of the end point without determining lane line from carriageway image；
On the other hand, if above-mentioned policing parameter meets above-mentioned lane detection rule, it can determine that current road conditions are flat
Whole forthright then may further determine that whether traveling state of vehicle is stable.Specific associated description as detailed below.
Video camera can determine whether current vehicle driving condition is stable by the change rate of the lane line parameter detected.
Wherein, above-mentioned lane line parameter may include slope and intercept of the lane line in plane of delineation coordinate system.
Specifically, video camera can determine the lane line parameter in current lane image and the vehicle in former frame carriageway image
Whether the difference of diatom is less than default fractional threshold.Wherein, which can be in application process based on empirically determined
Lane line difference out is to the acceptable numerical value of video camera attitude detection, for example, can be 10%.
If lane line parameter is slope and intercept, by the slope of two lane lines in current lane image and can cut
Away from the slope and intercept of corresponding two lane lines in former frame carriageway image is individually subtracted.Further, by difference again divided by
The slope and intercept of corresponding two lane lines, obtain four ratios in former frame carriageway image.It can be to four ratio calculations
Then average value is compared using the average value as difference with above-mentioned fractional threshold；Alternatively, directly taking in four ratios most
Big value, is then compared using the maximum value as difference with above-mentioned fractional threshold.
On the one hand, if difference is not less than the fractional threshold, it is determined that current vehicle driving condition is unstable, is not necessarily to from lane
The coordinate position of the end point of lane line is determined in image；
On the other hand, if difference is less than the fractional threshold, it is determined that current vehicle driving condition is stablized, at this point it is possible to from
The coordinate position of the end point of lane line is determined in carriageway image.
Certainly, if needing for vehicle to be parked in the center of smooth forthright in the initial stage of video camera installation, then passing through
After video camera acquires carriageway image, the coordinate position of lane line end point in carriageway image is determined.When the seat of lane line end point
When mark is set to image center, illustrates the pitch angle of video camera and yaw angle is 0, determining current camera Installation posture just
Really.
Step 302: coordinate position and preset camera parameter based on the end point determine that the candidate of video camera bows
The elevation angle and candidate yaw angle.
Wherein, above-mentioned camera parameter may include coordinate bit of the picture centre of carriageway image in described image coordinate system
It sets, the size of the unit pixel of the focal length of video camera and carriageway image.Above-mentioned candidate's pitch angle is referred to through a frame lane figure
As the pitch angle for the video camera determined, above-mentioned candidate's yaw angle refers to the video camera determined by a frame carriageway image
Yaw angle.
Illustrate that the application determines the candidate pitch angle of video camera and the principle of candidate yaw angle below.
When video camera posture is normal, end point carriageway image picture centre in plane of delineation u-v coordinate system
Coordinate is (u0,v0).When video camera posture changes, the candidate pitch angle of video camera is β, and candidate yaw angle is α.
Then coordinate (u of the end point in plane of delineation u-v coordinate system1,v1) can be indicated by following formula (1):
Wherein, f is the focal length of video camera, and dx indicates pixel in the size of X-direction (u axis direction), and dy indicates pixel
Size of the point in Y direction (v axis direction).
Therefore, the candidate pitch angle of video camera can be indicated by following formula (2):
The candidate yaw angle of video camera can be indicated by following formula (3):
Step 303: calculating the confidence rate and candidate yaw angle of the candidate pitch angle that the carriageway image of specified quantity is determined
Confidence rate, the maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera, by the maximum candidate of confidence rate
Yaw angle is determined as the actual yaw angle of video camera.
For the calculating error for reducing pitch angle and yaw angle, it can recorde the candidate determined based on several carriageway images and bow
The elevation angle and candidate offset angle, after the candidate pitch angle and candidate offset angle that the carriageway image for having recorded specified quantity is determined,
The above-mentioned candidate confidence rate of pitch angle and the confidence rate at above-mentioned candidate offset angle can be calculated separately.Specific calculation can be with
Using data processing means such as gauss of distribution function and Euclidean distances, this will not be repeated here.
After the completion of calculating, the maximum candidate pitch angle of confidence rate can be determined as the actual pitch angle of video camera, it will
The maximum candidate yaw angle of confidence rate is determined as the actual yaw angle of video camera.
It in the embodiment of the present application, can be based on the actual pitch angle and actual yaw angle determined to above-mentioned camera shooting
Machine carries out pose adjustment.
Such as: if yaw angle is 1 degree, illustrates that video camera deviates to the right 1 degree, therefore, video camera can be adjusted 1 to the left
Degree.If yaw angle is -1 degree, illustrates that video camera deviates 1 degree to the left, therefore, video camera can be deviated to the right to 1 degree.
If pitch angle is 1 degree, illustrates that video camera offsets up 1 degree, therefore, video camera can be adjusted downwards to 1 degree.If bowing
The elevation angle is -1 degree, then illustrates that video camera offsets downward 1 degree, therefore, video camera can be adjusted upward to 1 degree.
It, can also be to having generated after determining the yaw angle and pitch angle of video camera in addition, in the embodiment of the present application
The vehicle of error and the actual range of two sides lane line are adjusted, so that lane detection available for vehicle shift one
A accurate result.Specific calculating process is as follows:
If the coordinate a little in world coordinate system in space is denoted as (Xw,Yw,Zw), u-v of this in the plane of delineation
Coordinate in coordinate system is denoted as (u, v), and coordinate of this in camera coordinate system is (Xc,Yc,Zc).Then in world coordinate system
The conversion of coordinate coordinate into image coordinate system can be indicated by following formula (4):
Wherein, f is the focal length of video camera, and dx indicates pixel in the size of X-direction (u axis direction), and dy indicates pixel
For point in the size of Y direction (v axis direction), t is translation matrix, and R is that spin matrix can based on the pitch angle that has determined that and partially
Boat angle is calculated, and calculation can be indicated by following formula (5):
Wherein, α is the yaw angle of video camera, and β is the pitch angle of video camera.
Translation matrix t is represented byWherein, h indicates the height apart from ground of video camera.
Any two points on pick-up diatom calculate the slope of lane line in the coordinate that image coordinate is fastened, then slope k can lead to
Cross following formula (6) expression:
Wherein, (u1,v1) and (u2,v2) be image coordinate system on lane line two points.
The abscissa u of the point in image coordinate system and ordinate v can be passed through in L matrix by above-mentioned formula (4)
Coordinate in element and world coordinate system indicates, is then brought into above-mentioned formula (6).
When video camera is in correct posture, camera optical axis is parallel with lane line, therefore, point and light on lane line
The distance of axis is identical, that is to say, that X of the point in world coordinate system on lane linewIt is identical, and the point on lane line is in the world
Y in coordinate systemwIt is 0, in addition, the coordinate of Z-direction of two points in world coordinate system can disappear in calculating process.
Therefore, final slope can be indicated by following formula (7):
After bringing each coefficient into calculating, reduce, final lane line and video camera distance XwFollowing formula can be passed through
(8) it indicates:
Wherein, slope k can actually be directly based upon the position determination of lane line in image, the slope for two lane lines
Therefore difference the actual range of vehicle Yu two sides lane line can be calculated by formula (8)；Each element in L matrix can also
It is obtained in the calculating process of formula (4).
Therefore, after determining actual pitch angle and actual deviation angle, formula (8) can be arrived by above-mentioned formula (4)
Determine the actual range of vehicle Yu two sides lane line.
In conclusion the application by coordinate position of the end point of lane line in carriageway image in image coordinate system and
Camera parameter determines that the candidate pitch angle of video camera and candidate yaw angle, the carriageway image for being then based on specified quantity are determined
The actual pitch angle of video camera and actual yaw angle, to be taken the photograph according to the actual pitch angle and the actual yaw angle determination
Camera posture；
By above-mentioned measure, video camera posture can be adjusted, avoid video camera posture not face lane detection
Caused by influence.
Corresponding with the embodiment of the detection method of aforementioned video camera posture, present invention also provides the inspections of video camera posture
Survey the embodiment of device.
It is a kind of embodiment block diagram of the detection device of video camera posture shown in the application referring to Fig. 6:
As shown in fig. 6, the detection device 60 of the video camera posture includes:
First determination unit 610, for determining coordinate of the end point of lane line in carriageway image in image coordinate system
Position；Wherein, the end point is the intersection point of two lane lines.
Computing unit 620, for based on the end point coordinate position and preset camera parameter determine video camera
Candidate pitch angle and candidate yaw angle.
Second determination unit 630, the confidence rate for the candidate pitch angle that the carriageway image for calculating specified quantity is determined
With the confidence rate of candidate yaw angle, the maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera, by confidence
The maximum candidate yaw angle of rate is determined as the actual yaw angle of video camera.
In this example, described device further include:
640 (not shown) of adjustment unit, for based on the actual pitch angle and actual yaw angle pair determined
The video camera carries out pose adjustment.
In this example, first determination unit 610, is further used for:
The measurement parameter for obtaining onboard sensor, determines whether current road conditions meet preset vehicle based on the measurement parameter
Diatom detected rule；
If so, determining the difference of the lane line parameter in current lane image and the lane line parameter in former frame carriageway image
It is different whether to be less than default fractional threshold；
If so, determining coordinate position of the end point of lane line in image coordinate system from current lane image.
In this example, the camera parameter includes coordinate bit of the picture centre of the carriageway image in image coordinate system
It sets, the unit pixel size of the focal length of video camera and the carriageway image；
The computing unit 620, is further used for:
The candidate pitch angle β of the video camera is calculated based on following first formula:
Wherein, u1For the abscissa of the end point, u0For the abscissa of the picture centre of the carriageway image, dx is institute
Unit pixel size of the carriageway image on abscissa direction is stated, f is the focal length of video camera；
The candidate yaw angle α of the video camera is calculated based on following second formula:
Wherein, v1For the ordinate of the end point, v0For the ordinate of the picture centre of the carriageway image, dy is institute
State carriageway image in the ordinate on unit pixel size, f be video camera focal length, β be video camera candidate pitch angle.
In this example, described device further include:
650 (not shown) of third determination unit, for based on the actual pitch angle and actual offset determined
Angle determines the actual range of vehicle Yu two sides lane line.
The embodiment of the application video camera Attitute detecting device can be using on an electronic device.Installation practice can lead to
Software realization is crossed, can also be realized by way of hardware or software and hardware combining.Taking software implementation as an example, as a logic
Device in meaning is to be referred to computer program corresponding in nonvolatile memory by the processor of electronic equipment where it
It enables and is read into memory what operation was formed.For hardware view, as shown in fig. 7, being the application video camera Attitute detecting device
A kind of hardware structure diagram of place electronic equipment in addition to processor shown in Fig. 7, memory, network interface and non-volatile is deposited
Except reservoir, the electronic equipment in embodiment where device is gone back generally according to the actual functional capability of the video camera Attitute detecting device
It may include other hardware, this repeated no more.
The function of each unit and the realization process of effect are specifically detailed in the above method and correspond to step in above-mentioned apparatus
Realization process, details are not described herein.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, wherein described be used as separation unit
The unit of explanation may or may not be physically separated, and component shown as a unit can be or can also be with
It is not physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to actual
The purpose for needing to select some or all of the modules therein to realize application scheme.Those of ordinary skill in the art are not paying
Out in the case where creative work, it can understand and implement.
The foregoing is merely the preferred embodiments of the application, not to limit the application, all essences in the application
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the application protection.
Claims (10)
1. a kind of detection method of video camera posture characterized by comprising
Determine coordinate position of the end point of lane line in carriageway image in image coordinate system；Wherein, the end point is two
The intersection point of lane line；
Coordinate position and preset camera parameter based on the end point determine that the candidate pitch angle of video camera and candidate are inclined
Boat angle；
The confidence rate of the confidence rate for the candidate pitch angle that the carriageway image of calculating specified quantity is determined and candidate yaw angle, will set
The maximum candidate pitch angle of letter rate is determined as the actual pitch angle of video camera, and the maximum candidate yaw angle of confidence rate is determined as taking the photograph
The actual yaw angle of camera.
2. the method according to claim 1, wherein the method also includes:
Pose adjustment is carried out to the video camera based on the actual pitch angle and actual yaw angle determined.
3. the method according to claim 1, wherein the end point of lane line is being schemed in the determining carriageway image
As the coordinate position in coordinate system, comprising:
The measurement parameter for obtaining onboard sensor, determines whether current road conditions meet preset lane line based on the measurement parameter
Detected rule；
If so, determining that lane line parameter and the difference of the lane line parameter in former frame carriageway image in current lane image are
It is no to be less than default fractional threshold；
If so, from coordinate position of the end point of lane line in described image coordinate system is determined in current lane image.
4. the method according to claim 1, wherein the camera parameter includes in the image of the carriageway image
Coordinate position, the focal length of video camera and the unit pixel size of the carriageway image of the heart in described image coordinate system；
The coordinate position based on the end point and preset camera parameter determine the candidate pitch angle and time of video camera
Select yaw angle, comprising:
The candidate pitch angle β of the video camera is calculated based on following first formula:
Wherein, u1For the abscissa of the end point, u0For the abscissa of the picture centre of the carriageway image, dx is the vehicle
Unit pixel size of the road image on abscissa direction, f are the focal length of video camera；
The candidate yaw angle α of the video camera is calculated based on following second formula:
Wherein, v1For the ordinate of the end point, v0For the ordinate of the picture centre of the carriageway image, dy is the vehicle
Road image in the ordinate on unit pixel size, f be video camera focal length, β be video camera candidate pitch angle.
5. the method according to claim 1, wherein the method also includes:
Based on the actual pitch angle and actual deviation angle determined, the actual range of vehicle Yu two sides lane line is determined.
6. a kind of detection device of video camera posture characterized by comprising
First determination unit, for determining coordinate position of the end point of lane line in carriageway image in image coordinate system；Its
In, the end point is the intersection point of two lane lines；
Computing unit, for based on the end point coordinate position and preset camera parameter determine that the candidate of video camera bows
The elevation angle and candidate yaw angle；
Second determination unit, the confidence rate for the candidate pitch angle that the carriageway image for calculating specified quantity is determined and candidate are inclined
The maximum candidate pitch angle of confidence rate is determined as the actual pitch angle of video camera by the confidence rate at boat angle, and confidence rate is maximum
Candidate yaw angle is determined as the actual yaw angle of video camera.
7. device according to claim 6, which is characterized in that described device further include:
Adjustment unit, for carrying out posture tune to the video camera based on the actual pitch angle and actual yaw angle determined
It is whole.
8. device according to claim 6, which is characterized in that first determination unit is further used for:
The measurement parameter for obtaining onboard sensor, determines whether current road conditions meet preset lane line based on the measurement parameter
Detected rule；
If so, determining that lane line parameter and the difference of the lane line parameter in former frame carriageway image in current lane image are
It is no to be less than default fractional threshold；
If so, determining coordinate position of the end point of lane line in image coordinate system from current lane image.
9. device according to claim 6, which is characterized in that the camera parameter includes in the image of the carriageway image
Coordinate position, the focal length of video camera and the unit pixel size of the carriageway image of the heart in image coordinate system；
The computing unit, is further used for:
The candidate pitch angle β of the video camera is calculated based on following first formula:
Wherein, u1For the abscissa of the end point, u0For the abscissa of the picture centre of the carriageway image, dx is the vehicle
Unit pixel size of the road image on abscissa direction, f are the focal length of video camera；
The candidate yaw angle α of the video camera is calculated based on following second formula:
Wherein, v1For the ordinate of the end point, v0For the ordinate of the picture centre of the carriageway image, dy is the vehicle
Road image in the ordinate on unit pixel size, f be video camera focal length, β be video camera candidate pitch angle.
10. device according to claim 6, which is characterized in that described device further include:
Third determination unit, for determining vehicle and two sides vehicle based on the actual pitch angle and actual deviation angle determined
The actual range of diatom.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811025833.8A CN109345593B (en) | 2018-09-04 | 2018-09-04 | Camera posture detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811025833.8A CN109345593B (en) | 2018-09-04 | 2018-09-04 | Camera posture detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109345593A true CN109345593A (en) | 2019-02-15 |
| CN109345593B CN109345593B (en) | 2022-04-26 |
Family
ID=65293770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811025833.8A Active CN109345593B (en) | 2018-09-04 | 2018-09-04 | Camera posture detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109345593B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363819A (en) * | 2019-06-25 | 2019-10-22 | 华为技术有限公司 | The method and relevant device that image capture device is demarcated in intelligent automobile |
| CN110718068A (en) * | 2019-09-27 | 2020-01-21 | 华中科技大学 | Road monitoring camera installation angle estimation method |
| CN111739105A (en) * | 2020-08-07 | 2020-10-02 | 北京理工大学 | Automatic learning method for live broadcast station camera shooting scheme |
| CN112184822A (en) * | 2019-07-01 | 2021-01-05 | 北京地平线机器人技术研发有限公司 | Camera pitch angle adjusting method and device, storage medium and electronic equipment |
| CN112837352A (en) * | 2021-04-20 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Image-based data processing method, device and equipment, automobile and storage medium |
| CN112866629A (en) * | 2019-11-27 | 2021-05-28 | 深圳市大富科技股份有限公司 | Binocular vision application control method and terminal |
| CN113227708A (en) * | 2021-03-30 | 2021-08-06 | 深圳市锐明技术股份有限公司 | Method and device for determining pitch angle and terminal equipment |
| CN113490967A (en) * | 2020-09-22 | 2021-10-08 | 深圳市锐明技术股份有限公司 | Camera calibration method and device and electronic equipment |
| CN113515973A (en) * | 2020-04-09 | 2021-10-19 | 北京地平线机器人技术研发有限公司 | Data acquisition method, training method, adjustment method and device |
| CN114782549A (en) * | 2022-04-22 | 2022-07-22 | 南京新远见智能科技有限公司 | Camera calibration method and system based on fixed point identification |
| CN115103123A (en) * | 2022-07-07 | 2022-09-23 | 上海智能交通有限公司 | Automatic vehicle-mounted holder camera distribution and control method and system based on lane line identification |
| CN115164823A (en) * | 2022-05-16 | 2022-10-11 | 上海芯翌智能科技有限公司 | Method and device for acquiring gyroscope information of camera |
| CN116309814A (en) * | 2022-11-29 | 2023-06-23 | 北京斯年智驾科技有限公司 | Vehicle pose determination method, device, computing equipment and medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024139A (en) * | 2009-09-18 | 2011-04-20 | 富士通株式会社 | Device and method for recognizing character strings |
| CN102324017A (en) * | 2011-06-09 | 2012-01-18 | 中国人民解放军国防科学技术大学 | FPGA (Field Programmable Gate Array)-based lane line detection method |
| CN103903435A (en) * | 2012-12-30 | 2014-07-02 | 王方淇 | Road condition information obtaining method and mobile terminal |
| CN104008645A (en) * | 2014-06-12 | 2014-08-27 | 湖南大学 | Lane line predicating and early warning method suitable for city road |
| CN105320934A (en) * | 2014-07-11 | 2016-02-10 | 株式会社电装 | Lane boundary line recognition device |
| CN105447892A (en) * | 2015-11-05 | 2016-03-30 | 奇瑞汽车股份有限公司 | Method and device for determining yaw angle of vehicle |
| CN105740809A (en) * | 2016-01-28 | 2016-07-06 | 东南大学 | Expressway lane line detection method based on onboard camera |
| CN106875448A (en) * | 2017-02-16 | 2017-06-20 | 武汉极目智能技术有限公司 | A kind of vehicle-mounted monocular camera external parameter self-calibrating method |
| CN108052908A (en) * | 2017-12-15 | 2018-05-18 | 郑州日产汽车有限公司 | Track keeping method |
| CN108450058A (en) * | 2015-12-28 | 2018-08-24 | 英特尔公司 | Automatic vehicle-mounted camera calibrated in real time |
-
2018
- 2018-09-04 CN CN201811025833.8A patent/CN109345593B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024139A (en) * | 2009-09-18 | 2011-04-20 | 富士通株式会社 | Device and method for recognizing character strings |
| CN102324017A (en) * | 2011-06-09 | 2012-01-18 | 中国人民解放军国防科学技术大学 | FPGA (Field Programmable Gate Array)-based lane line detection method |
| CN103903435A (en) * | 2012-12-30 | 2014-07-02 | 王方淇 | Road condition information obtaining method and mobile terminal |
| CN104008645A (en) * | 2014-06-12 | 2014-08-27 | 湖南大学 | Lane line predicating and early warning method suitable for city road |
| CN105320934A (en) * | 2014-07-11 | 2016-02-10 | 株式会社电装 | Lane boundary line recognition device |
| CN105447892A (en) * | 2015-11-05 | 2016-03-30 | 奇瑞汽车股份有限公司 | Method and device for determining yaw angle of vehicle |
| CN108450058A (en) * | 2015-12-28 | 2018-08-24 | 英特尔公司 | Automatic vehicle-mounted camera calibrated in real time |
| CN105740809A (en) * | 2016-01-28 | 2016-07-06 | 东南大学 | Expressway lane line detection method based on onboard camera |
| CN106875448A (en) * | 2017-02-16 | 2017-06-20 | 武汉极目智能技术有限公司 | A kind of vehicle-mounted monocular camera external parameter self-calibrating method |
| CN108052908A (en) * | 2017-12-15 | 2018-05-18 | 郑州日产汽车有限公司 | Track keeping method |
Non-Patent Citations (1)
| Title |
|---|
| 罗瑾等: "基于车道线交点的车载视频稳像算法", 《计算机技术与发展》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363819A (en) * | 2019-06-25 | 2019-10-22 | 华为技术有限公司 | The method and relevant device that image capture device is demarcated in intelligent automobile |
| CN110363819B (en) * | 2019-06-25 | 2023-03-03 | 华为技术有限公司 | Method for calibrating image acquisition equipment in intelligent automobile and related equipment |
| CN112184822B (en) * | 2019-07-01 | 2024-01-30 | 北京地平线机器人技术研发有限公司 | Camera pitch angle adjusting method and device, storage medium and electronic equipment |
| CN112184822A (en) * | 2019-07-01 | 2021-01-05 | 北京地平线机器人技术研发有限公司 | Camera pitch angle adjusting method and device, storage medium and electronic equipment |
| CN110718068A (en) * | 2019-09-27 | 2020-01-21 | 华中科技大学 | Road monitoring camera installation angle estimation method |
| CN112866629A (en) * | 2019-11-27 | 2021-05-28 | 深圳市大富科技股份有限公司 | Binocular vision application control method and terminal |
| CN113515973A (en) * | 2020-04-09 | 2021-10-19 | 北京地平线机器人技术研发有限公司 | Data acquisition method, training method, adjustment method and device |
| CN111739105A (en) * | 2020-08-07 | 2020-10-02 | 北京理工大学 | Automatic learning method for live broadcast station camera shooting scheme |
| CN113490967A (en) * | 2020-09-22 | 2021-10-08 | 深圳市锐明技术股份有限公司 | Camera calibration method and device and electronic equipment |
| CN113227708A (en) * | 2021-03-30 | 2021-08-06 | 深圳市锐明技术股份有限公司 | Method and device for determining pitch angle and terminal equipment |
| CN112837352A (en) * | 2021-04-20 | 2021-05-25 | 腾讯科技（深圳）有限公司 | Image-based data processing method, device and equipment, automobile and storage medium |
| CN114782549A (en) * | 2022-04-22 | 2022-07-22 | 南京新远见智能科技有限公司 | Camera calibration method and system based on fixed point identification |
| CN114782549B (en) * | 2022-04-22 | 2023-11-24 | 南京新远见智能科技有限公司 | Camera calibration method and system based on fixed point identification |
| CN115164823A (en) * | 2022-05-16 | 2022-10-11 | 上海芯翌智能科技有限公司 | Method and device for acquiring gyroscope information of camera |
| CN115164823B (en) * | 2022-05-16 | 2024-04-02 | 上海芯翌智能科技有限公司 | Method and device for acquiring gyroscope information of camera |
| CN115103123A (en) * | 2022-07-07 | 2022-09-23 | 上海智能交通有限公司 | Automatic vehicle-mounted holder camera distribution and control method and system based on lane line identification |
| CN116309814A (en) * | 2022-11-29 | 2023-06-23 | 北京斯年智驾科技有限公司 | Vehicle pose determination method, device, computing equipment and medium |
| CN116309814B (en) * | 2022-11-29 | 2024-03-08 | 北京斯年智驾科技有限公司 | Vehicle pose determination method, device, computing equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109345593B (en) | 2022-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109345593A (en) | A kind of detection method and device of video camera posture | |
| US10964054B2 (en) | Method and device for positioning | |
| CN109902637B (en) | Lane line detection method, lane line detection device, computer device, and storage medium | |
| CN108885791B (en) | Ground detection method, related device and computer readable storage medium | |
| CN108805934B (en) | External parameter calibration method and device for vehicle-mounted camera | |
| Bensrhair et al. | A cooperative approach to vision-based vehicle detection | |
| US7889887B2 (en) | Lane recognition apparatus | |
| CN109931939B (en) | Vehicle positioning method, device, equipment and computer readable storage medium | |
| US10706582B2 (en) | Real-time monocular structure from motion | |
| US10260862B2 (en) | Pose estimation using sensors | |
| US9049396B2 (en) | Creating composite images based on image capture device poses corresponding to captured images | |
| CN107305632B (en) | Monocular computer vision technology-based target object distance measuring method and system | |
| US20150138310A1 (en) | Automatic scene parsing | |
| EP2736013A2 (en) | Calibration of multiple cameras on a moving object | |
| WO2020228694A1 (en) | Camera pose information detection method and apparatus, and corresponding intelligent driving device | |
| US20220319042A1 (en) | Detection, 3d reconstruction and tracking of multiple rigid objects moving in relation to one another | |
| CN112115980A (en) | Binocular vision odometer design method based on optical flow tracking and point line feature matching | |
| US10991105B2 (en) | Image processing device | |
| CN112150558A (en) | Method and device for acquiring three-dimensional position of obstacle for roadside computing equipment | |
| JP5310027B2 (en) | Lane recognition device and lane recognition method | |
| CN110109465A (en) | A kind of self-aiming vehicle and the map constructing method based on self-aiming vehicle | |
| CN106228531B (en) | Automatic vanishing point calibration method and system based on horizon line search | |
| CN111105695A (en) | Map making method and device, electronic equipment and computer readable storage medium | |
| CN109214254A (en) | A kind of method and device of determining robot displacement | |
| CN112489106A (en) | Video-based vehicle size measuring method and device, terminal and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |