CN111145263A - Vehicle-mounted-based automatic camera calibration method - Google Patents
Vehicle-mounted-based automatic camera calibration method Download PDFInfo
- Publication number
- CN111145263A CN111145263A CN201910976937.5A CN201910976937A CN111145263A CN 111145263 A CN111145263 A CN 111145263A CN 201910976937 A CN201910976937 A CN 201910976937A CN 111145263 A CN111145263 A CN 111145263A
- Authority
- CN
- China
- Prior art keywords
- coordinate system
- coordinates
- point
- vehicle body
- vehicle
- 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.)
- Withdrawn
Links
Images
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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the field of image processing, and discloses a vehicle-mounted camera-based automatic calibration method which comprises the following steps: acquiring image data and vehicle-mounted positioning world coordinates; judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate; performing corner detection and matching on the image data; calculating a vehicle body course vector based on the coordinates of the angular points; calculating a road plane equation of the corner points based on the coordinates of the corner points; establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the vehicle body coordinate system coordinates of the angular points; and calculating a transformation matrix for converting the pixel coordinate system into the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates. Some technical effects of the invention are as follows: and automatic calibration of the camera is realized.
Description
Technical Field
The invention relates to the field of image processing, in particular to a camera calibration technology in the field of image processing.
Background
In the field of computer vision and photogrammetry, camera calibration is necessary to obtain the correspondence between computer pixel points and actual physical space points. The camera calibration refers to a process of processing images of a certain camera model and solving internal and external parameters of the camera model by using a series of mathematical transformation and calculation methods. The conventional method using a calibration reference such as a calibration template, typically a two-step method of Tsai, has a great inconvenience to the photographing operation and the calibration method due to the use of the calibration reference and the large amount of manual involvement in the photographing and calibration processes.
Disclosure of Invention
In order to at least solve the problem of automatic calibration of camera calibration, the invention provides a vehicle-mounted camera-based automatic calibration method, which adopts the following technical scheme:
the method comprises the following steps: acquiring image data and vehicle-mounted positioning world coordinates; judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate; performing corner detection and matching on the image data; calculating a vehicle body course vector based on the coordinates of the angular points; calculating a road plane equation based on the coordinates of the corner points; establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the coordinates of the vehicle body coordinate system; and calculating a transformation matrix M from the pixel coordinate system to the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates.
Preferably, the determination as to whether the vehicle is traveling in a straight line is made by: taking any continuous two groups of positioning data, and fitting respectively; when the included angle of the two fitted straight lines is smaller than a first threshold value, the vehicle can be judged to run along the straight lines; if yes, the subsequent steps are entered.
Preferably, said first threshold value is 5 °.
Preferably, the corner detection is performed by filtering the image data and then performing Shi-Tomasi corner detection; the corner matching comprises the matching of the same-target corner and the matching of different-target corners, namely, for any target of the camera, frame image data at a plurality of different moments are obtained, and corner descriptors are calculated for matching; respectively acquiring frame images of each target at a certain moment for different targets of the camera, and calculating an angular point descriptor for matching; and when the number of the matched corner points is not less than a preset second threshold value, entering the subsequent step.
Preferably, said second threshold is 4.
Preferably, the vehicle body heading vector is calculated: the angular point is identified at the time of O', and the pixel coordinates of the left eye angular point are A respectively1′，A2′，A3′，A4′，A5′，A6' the pixel coordinates of the corresponding right eye corner point are A respectively1″，A2″，A3″，A4″，A5″，A6And obtaining the coordinates of the calibration point in a coordinate system of the binocular camera according to a binocular distance measuring principle as follows: a. the1，A2，A3，A4，A5，A6(ii) a Similarly, at the time O, the coordinates of the binocular camera coordinate system of the corner point are obtained as follows: b is1，B2，B3，B4，B5，B6The simultaneous equations have:
A1＝MoB1
A2＝MoB2
A3＝MoB3
A4＝MoB4
coordinate transformation matrix can be calculated by solving equation set
Wherein R isoIs a rotation matrix;
and the three-dimensional translation vector is the vehicle body heading vector.
Preferably, the road plane equation for the corner point is calculated: at time O, 6 coordinate points in the binocular camera coordinate system are known: b is1，B2，B3，B4，B5，B6Let the road plane equation of 6 points be:
Ax+By+Cz+D＝0
both sides are divided by D at the same time:
ax+by+cz+1＝0
namely:
according to the principle of least square method:
preferably, the body coordinate system coordinates of the corner points are calculated: the projection point of the optical center O point of the camera on the road plane at the time O is set as OpThe projection point of the optical center O 'point of the camera on the road plane at the time of O' is Op′；
With OpAs origin, vector
Establishing a vehicle body coordinate system on a road plane for a vehicle body coordinate system y axis;
for any point B of the road planeiCoordinate (x) of coordinate system of vehicle bodyi，yi0) is B 'as a projection point on the y-axis'iThen there is
Let vector quantity
And the y axis vector
Is a vector product of
Namely:
normal vector of road-setting plane
And
the angle of the vectors is x and,
then:
therefore, the method comprises the following steps:
if λ < 90, then xiNegative, and vice versa, positive.
Preferably, the transformation matrix of the pixel coordinate system to the body coordinate system
Wherein α, β and theta are rotation angles around the x-axis, y-axis and z-axis respectively, and Tx、Ty、TzRespectively, in the x-axis, y-axis, and z-axis directions.
Preferably, the method further comprises the following steps: converting from the body coordinates to corresponding world coordinates:
setting a point O' and the world coordinate system coordinate of the point O as (x)2，y2，z2)，(x1，y1，z1) And the course angle is delta, then:
namely:
①x2＞x1，y2＞y1then, δ is 2 π - μ
②x2＜x1，y2＞y1When δ is equal to μ
③x2＞x1，y2＜y1Then, δ is π + μ
④x2＜x1，y2＜y1Then, δ ═ pi- μ
Let the coordinate of any point p in the coordinate system of the vehicle body at any time be (x)c，yc，zc) The vehicle-mounted positioning world coordinate corresponding to the time is (x)o，yo，zo)，
The coordinates (x, y, z) of the point p in the world coordinate system are:
x＝xccosδ-ycsinδ+xo
y＝xccosδ+ycsinδ+yo
z＝zc+zo。
the method at least provides a technical scheme for automatically calibrating the camera based on the vehicle, does not need a calibration object, greatly reduces manual participation, and at least can well realize the automatic calibration of the camera.
Drawings
For a better understanding of the technical solution of the present invention, reference is made to the following drawings, which are included to assist in describing the prior art or embodiments. These drawings will selectively demonstrate articles of manufacture or methods related to either the prior art or some embodiments of the invention. The basic information for these figures is as follows:
FIG. 1 is a schematic flow chart of a vehicle-mounted monocular calibration method in one embodiment.
Detailed Description
The technical means or technical effects related to the present invention will be further described below, and it is obvious that the examples provided are only some embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step, will be within the scope of the present invention based on the embodiments of the present invention and the explicit or implicit representations or hints.
On the general idea, the invention discloses a vehicle-mounted monocular calibration method, which comprises the following steps: acquiring image data and vehicle-mounted positioning world coordinates; judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate; performing corner detection and matching on the image data; calculating a vehicle body course vector based on the coordinates of the angular points; calculating a road plane equation based on the corner point coordinates; establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the coordinates of the vehicle body coordinate system; and calculating a transformation matrix M from the pixel coordinate system to the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates.
Based on the general concept, those skilled in the art should understand that the vehicle in the vehicle of the present invention refers to a vehicle driven or towed by a power device, and the power is generally from an internal combustion engine or an electric motor. The positioning information refers to position information provided by GNSS, including but not limited to world coordinate system coordinates. GNSS, a satellite navigation system, includes, but is not limited to, GPS in the united states, GLONASS in russia, Galileo in the european union, and BDS in china. The corner points are also called feature points, interest points and extreme points, and refer to representative and robust points in the image.
Some technical effects of the invention are as follows: and a calibration object is not needed, and automatic calibration is realized.
In some embodiments, a camera or a device having a camera function is mounted and fixed on a vehicle. Generally, a camera or an apparatus having an image pickup function is mounted and fixed in front of a vehicle, particularly, right above a front window glass of the vehicle, so as to obtain a good working view environment. The device with the camera function refers to a device capable of shooting and acquiring image data such as videos or pictures, such as an acquisition terminal device for acquiring map data in the mapping field and a terminal device for visually identifying road conditions in the automatic driving field. Generally, a camera integrates a positioning function or is mounted with a positioning device. And in the working process of the camera, the positioning data of the camera is acquired at the same time to obtain world coordinates. Ideally, when the positioning device coincides with the optical center of the camera, the obtained positioning data is the positioning data of the optical center. In reality, the positioning data of the camera can be regarded as the positioning data of the optical center. Particularly low, if it is necessary to eliminate the aforementioned errors, the translation vector of the positioning device to the optical center of the camera can be calculated and compensated.
It will be appreciated that the foregoing embodiment may be disposable, i.e., set for the first time, and subsequently may not need to be reset without environmental change; when the calibration is performed again, the operation steps of the above embodiment can be omitted.
In some embodiments, image data and on-board positioning world coordinates are acquired; judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate; performing corner detection and matching on the image data; calculating a vehicle body course vector based on the coordinates of the angular points; calculating a road plane equation based on the corner point coordinates; establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the coordinates of the vehicle body coordinate system; and calculating a transformation matrix M for converting the pixel coordinate system into the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates.
In some embodiments, image data and on-board positioning world coordinates are acquired. The camera integrated with the positioning function is arranged in the middle upper part of the front window glass of the vehicle. When the vehicle moves forward, the camera works to continuously acquire image data and corresponding vehicle-mounted positioning world coordinates. The image data includes picture data and video data, and each frame of image data is associated with corresponding vehicle-mounted positioning world coordinates.
In some embodiments, whether the vehicle runs along a straight line is judged according to the vehicle-mounted positioning world coordinates. And acquiring vehicle-mounted positioning world coordinates corresponding to any 2n (n is more than or equal to 2) moments. And randomly dividing the positioning points into two groups, and respectively fitting each group of vehicle-mounted positioning points to obtain two corresponding straight lines. And when the included angle of the two straight lines is smaller than a first threshold value, judging that the vehicle runs along the straight line. Fitting as referred to herein means finding a straight line such that the sum of the distances of all points to the straight line is minimized.
In some embodiments, the vehicle-mounted positioning world coordinates corresponding to 10 moments are randomly acquired. And randomly dividing the positioning points into two groups, wherein each group comprises 5 vehicle-mounted positioning points. And respectively fitting the two groups of vehicle-mounted positioning points to obtain two straight lines. And when the included angle of the two straight lines is less than 5 degrees, judging that the vehicle runs along the straight line.
In some embodiments, corner detection and matching is performed from the image data. And acquiring a plurality of frames of image data, wherein each frame of image data corresponds to a certain moment and a specific world coordinate. Filtering (Gauss filtering, etc.) the image, and then performing Shi-Tomasi corner detection, that is, obtaining pixel points corresponding to the local maximum of the first derivative (i.e. the gradient of the gray level) of all pixel points in the image in the neighborhood, namely, obtaining the detected corner points. After the corner points are detected, sub-pixel level positioning is performed, i.e. curve fitting is performed to the image pixel values, and then the position of the peak value between the pixels is determined. Then, corner matching is performed: the corner matching comprises the matching of the same-eye corner and the matching of the different-eye corner. The former is to match the corner points of different frames of the same purpose; the latter refers to matching different destination corners of the same frame. In general, corner matching can be performed by vector comparison by computing a vector of corners in the image. And when the number of the matched corner points is not less than the second threshold value, judging that the corner point detection and the matching are successful. The number of matched corner points referred to herein refers to the number of corner points that are common to each frame of image and that are successfully matched.
In some embodiments, image data of frames corresponding to a left target and a right target at the time O are obtained, and corresponding angular points are respectively detected; and acquiring image data of frames corresponding to the left target and the right target at the time of O', and respectively detecting corresponding angular points. Matching angular points of the left eye at the O' moment and the O moment; and matching corner points of the right eye at the O' moment and the O moment. Matching the left eye angular point and the right eye angular point at the time of O'; and matching the left eye corner point and the right eye corner point at the time O. And when the matching result shows that the matching corner is not less than 4, judging that the corner detection and matching are successful.
In some embodiments, a body heading vector is calculated based on coordinates of the corner points. Specifically, based on the coordinates of the camera coordinate system of the angular point, the heading vector of the vehicle body is calculated: the angular point is identified at the time of O', and the pixel coordinates of the left eye angular point are A respectively1′，A2′，A3′，A4′，A5′，A6' the pixel coordinates of the corresponding right eye corner point are A respectively1″，A2″，A3″，A4″，A5″，A6And obtaining the coordinates of the calibration point in a coordinate system of the binocular camera according to a binocular distance measuring principle as follows: a. the1，A2，A3，A4，A5，A6(ii) a Similarly, at the time O, the coordinates of the binocular camera coordinate system of the corner point are obtained as follows: b is1，B2，B3，B4，B5，B6The simultaneous equations have:
A1＝MoB1
A2＝MoB2
A3＝MoB3
A4＝MoB4
coordinate transformation matrix can be calculated by solving equation set
Wherein R isoIs a rotation matrix;
and the three-dimensional translation vector is the vehicle body heading vector.
In some embodiments, the road plane equation is calculated based on the coordinates of the corner points. Specifically, based on the camera coordinate system coordinates of the corner points, the road plane equation of the corner points is calculated: at time O, 6 coordinate points in the binocular camera coordinate system are known: b is1，B2，B3，B4，B5，B6Let the road plane equation of 6 points be:
Ax+By+Cz+D＝0
both sides are divided by D at the same time:
ax+by+cz+1＝0
namely:
according to the principle of least square method:
in some embodiments, a body coordinate system is established based on coordinates of the corner points, and body coordinate system coordinates are calculated. Specifically, a vehicle body coordinate system is established based on world coordinate system coordinates of the angular points, and vehicle body coordinate system coordinates of the angular points are calculated: the projection point of the optical center O point of the camera on the road plane at the time O is set as OpThe projection point of the optical center O 'point of the camera on the road plane at the time of O' is Op′；
With OpAs origin, vector
Establishing a vehicle body coordinate system on a road plane for a vehicle body coordinate system y axis;
for any point B of the road planeiCoordinate (x) of coordinate system of vehicle bodyi，yi0) is B 'as a projection point on the y-axis'iThen there is
Let vector quantity
And the y axis vector
Is a vector product of
Namely:
normal vector of road-setting plane
And
the angle of the vectors is x and,
then:
therefore, the method comprises the following steps:
such as lambda<90, then xiIs negative; λ > 90, then xiIs positive.
In some embodiments, the transformation matrix M for the conversion of the pixel coordinate system into the body coordinate system is calculated based on the correspondence of the body coordinate system coordinates and the pixel coordinates of the corner points. Specifically, at time O, the angular point vehicle body coordinate is C1，C2，C3，C4，C5，C6In the image corresponding image coordinates C1′，C2′，C3′，C4′，C5′，C6′。
Calculating transformation matrix for converting pixel coordinates into coordinates of vehicle body coordinate system
Wherein, the matrix parameter fx，fy，cx，cyFor the intra-camera parameters, a, β, θ are rotation angles around the x-axis, y-axis, and z-axis, respectively, and T _ x, T _ y, and T _ z are translations in the directions of the x-axis, y-axis, and z-axis, respectively.
Specifically, the transformation matrix is Ci＝MCi'; wherein, CiAs coordinate points of the coordinate system of the vehicle body, Ci' is a pixel coordinate point.
Since the calibration points are all located on the same road plane, the z-coordinate value is a constant. Since the projection transformation matrix from the three-dimensional coordinate point in the vehicle body coordinate system to the two-dimensional point in the pixel coordinate system is a 3 × 4 irreversible matrix, and since z is a constant, two extra parameters are eliminated, the transformation matrix can be rewritten into a 3 × 3 reversible matrix. That is, the coordinate points in the pixel coordinate system may be obtained from the coordinate points in the vehicle coordinate system, or the coordinate points in the road plane where z is a constant in the vehicle coordinate system may be obtained from the coordinate points in the pixel coordinate system.
Wherein the 3 × 4 irreversible matrix is
And the following steps:
wherein: rx，Ry，RzIs a rotation matrix around the x-axis, y-axis, z-axis, respectively, Tx，Ty，TzThe translation is in the directions of an x axis, a y axis and a z axis respectively.
Assuming that the rotation angle around the x-axis, the y-axis and the z-axis is α, theta, then:
therefore, the method is simple and easy to operate.
The road surface z under the coordinate system of the vehicle body is constant 0, and the third row is eliminated to obtain a transformation matrix
Wherein α, β and theta are rotation angles around the x-axis, y-axis and z-axis respectively, and Tx、Ty、TzRespectively, in the x-axis, y-axis, and z-axis directions.
In some embodiments, the body coordinates are also converted to corresponding world coordinates: setting a point O' and the world coordinate system coordinate of the point O as (x)2，y2，z2)，(x1，y1，z1) And the corresponding heading angle is δ (i.e. the angle between the straight line in the clockwise direction and the true north direction), then:
namely:
①x2＞x1，y2＞y1then, δ is 2 π - μ
②x2＜x1，y2＞y1When δ is equal to μ
③x2＞x1，y2＜y1Then, δ is π + μ
④x2＜x1，y2＜y1Then, δ ═ pi- μ
Let the coordinate of any point p in the coordinate system of the vehicle body at any time be (x)c，yc，zc) The vehicle-mounted positioning world coordinate corresponding to the time is (x)o，yo，zo)，
The coordinates (x, y, z) of the point p in the world coordinate system are:
x＝xccosδ-ycsinδ+xo
y＝xccosδ+ycsinδ+yo
z＝zc+zo。
in another aspect, in some embodiments, a storage medium is provided. The storage medium stores computer program instructions that, when executed by the processor, repeatedly perform at least once the following steps: acquiring image data and vehicle-mounted positioning world coordinates; judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate; performing corner detection and matching on the image data; calculating a vehicle body course vector based on the coordinates of the angular points; calculating a road plane equation of the corner points based on the coordinates of the corner points; establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the vehicle body coordinate system coordinates of the angular points; and calculating a transformation matrix M for converting the pixel coordinate system into the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates.
In some embodiments, a storage medium is provided. The storage medium stores computer program instructions that, when executed by the processor, repeatedly perform at least once the following steps: and converting the coordinates of the vehicle body into corresponding world coordinates.
It should be understood that the storage medium may be a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a Random Access Memory (RAM), a usb disk, a removable hard disk, or the like.
The various embodiments or features mentioned herein may be combined with each other as additional alternative embodiments without conflict, within the knowledge and ability level of those skilled in the art, and a limited number of alternative embodiments formed by a limited number of combinations of features not listed above are still within the scope of the present disclosure, as understood or inferred by those skilled in the art from the figures and above.
Finally, it is emphasized that the above-mentioned embodiments, which are typical and preferred embodiments of the present invention, are only used for explaining and explaining the technical solutions of the present invention in detail for the convenience of the reader, and are not used to limit the protection scope or application of the present invention.
Therefore, any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be covered within the protection scope of the present invention.
Claims (10)
1. A vehicle-mounted camera automatic calibration method is characterized by comprising the following steps: the method comprises the following steps:
acquiring image data and vehicle-mounted positioning world coordinates;
judging whether the vehicle runs along a straight line or not by the vehicle-mounted positioning world coordinate;
performing corner detection and matching on the image data;
calculating a vehicle body course vector based on the coordinates of the angular points;
calculating a road plane equation based on the coordinates of the corner points;
establishing a vehicle body coordinate system based on the coordinates of the angular points, and calculating the coordinates of the vehicle body coordinate system;
and calculating a transformation matrix M from the pixel coordinate system to the vehicle body coordinate system based on the corresponding relation between the vehicle body coordinate system coordinates of the angular points and the pixel coordinates.
2. The method of claim 1, wherein: judging whether the vehicle runs along the straight line or not is that:
taking any continuous two groups of positioning data, and fitting respectively;
when the included angle of the two fitted straight lines is smaller than a first threshold value, the vehicle can be judged to run along the straight lines;
if yes, the subsequent steps are entered.
3. The method of claim 2, wherein: the first threshold value is 5 °.
4. The method of claim 1, wherein:
the angular point detection is to carry out Shi-Tomasi angular point detection after filtering the image data;
the corner matching comprises the matching of the same-target corner and the matching of different-target corners, namely, for any target of the camera, frame image data at a plurality of different moments are obtained, and corner descriptors are calculated for matching; respectively acquiring frame images of each target at a certain moment for different targets of the camera, and calculating an angular point descriptor for matching;
and when the number of the matched corner points is not less than a preset second threshold value, entering the subsequent step.
5. The method of claim 4, wherein: the second threshold is 4.
6. The method of claim 1, wherein: the calculation of the heading vector of the vehicle body is as follows:
the angular point is identified at the time of O', and the pixel coordinates of the left eye angular point are A respectively1′，A2′，A3′，A4′，A5′，A6' the pixel coordinates of the corresponding right eye corner point are A respectively1″，A2″，A3″，A4″，A5″，A6And obtaining the coordinates of the calibration point in a coordinate system of the binocular camera according to a binocular distance measuring principle as follows: a. the1，A2，A3，A4，A5，A6(ii) a Similarly, at the time O, the coordinates of the binocular camera coordinate system of the corner point are obtained as follows: b is1，B2，B3，B4，B5，B6The simultaneous equations have:
A1＝MoB1
A2＝MoB2
A3＝MoB3
A4＝MoB4
coordinate transformation matrix can be calculated by solving equation set
Wherein R isOIs a rotation matrix;
and the three-dimensional translation vector is the vehicle body heading vector.
7. The method of claim 1, wherein: the road plane equation of the corner point is calculated as follows:
at time O, 6 coordinate points in the binocular camera coordinate system are known: b is1，B2，B3，B4，B5，B6Let the road plane equation of 6 points be:
Ax+By+Cz+D＝0
both sides are divided by D at the same time:
ax+by+cz+1＝0
namely:
according to the principle of least square method:
8. the method of claim 1, wherein: and the coordinate of the vehicle body coordinate system of the angular point is calculated as follows:
the projection point of the optical center O point of the camera on the road plane at the time O is set as OpThe projection point of the optical center O 'point of the camera on the road plane at the time of O' is Op′；
With OpAs origin, vector
Establishing a vehicle body coordinate system on a road plane for a vehicle body coordinate system y axis;
for any point B of the road planeiCoordinate (x) of coordinate system of vehicle bodyi，yi0) is B 'as a projection point on the y-axis'iThen there is
Let vector quantity
And the y axis vector
Is a vector product of
Namely:
normal vector of road-setting plane
And
the angle of the vectors is x and,
then:
therefore, the method comprises the following steps:
if λ < 90, then xiNegative, and vice versa, positive.
9. The method of claim 1, wherein: the transformation matrix from the pixel coordinate system to the vehicle body coordinate system
Wherein α, β and theta are rotation angles around the x-axis, y-axis and z-axis respectively, and Tx、Ty、TzRespectively, in the x-axis, y-axis, and z-axis directions.
10. The method of claim 9, further comprising: converting from the body coordinates to corresponding world coordinates:
setting a point O' and the world coordinate system coordinate of the point O as (x)2，y2，z2)，(x1，y1，z1) And the course angle is delta, then:
namely:
①x2＞x1，y2＞y1then, δ is 2 π - μ
②x2＜x1，y2＞y1When δ is equal to μ
③x2＞x1，y2＜y1Then, δ is π + μ
④x2＜x1，y2＜y1Then, δ ═ pi- μ
Let the coordinate of any point p in the coordinate system of the vehicle body at any time be (x)c，yc，zc) The vehicle-mounted positioning world coordinate corresponding to the time is (x)o，yo，zo)，
The coordinates (x, y, z) of the point p in the world coordinate system are:
x＝xccosδ-ycsinδ+xo
y＝xccosδ+ycsinδ+yo
z＝zc+zo。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910976937.5A CN111145263A (en) | 2019-10-14 | 2019-10-14 | Vehicle-mounted-based automatic camera calibration method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910976937.5A CN111145263A (en) | 2019-10-14 | 2019-10-14 | Vehicle-mounted-based automatic camera calibration method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111145263A true CN111145263A (en) | 2020-05-12 |
Family
ID=70516846
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910976937.5A Withdrawn CN111145263A (en) | 2019-10-14 | 2019-10-14 | Vehicle-mounted-based automatic camera calibration method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111145263A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113420732A (en) * | 2021-08-23 | 2021-09-21 | 深圳市城市交通规划设计研究中心股份有限公司 | Pavement disease detection method and device and storage medium |
| CN113639782A (en) * | 2021-08-13 | 2021-11-12 | 北京地平线信息技术有限公司 | External parameter calibration method and device for vehicle-mounted sensor, equipment and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106558080A (en) * | 2016-11-14 | 2017-04-05 | 天津津航技术物理研究所 | Join on-line proving system and method outside a kind of monocular camera |
| DE102016104729A1 (en) * | 2016-03-15 | 2017-09-21 | Connaught Electronics Ltd. | Method for extrinsic calibration of a camera, computing device, driver assistance system and motor vehicle |
| CN110223354A (en) * | 2019-04-30 | 2019-09-10 | 惠州市德赛西威汽车电子股份有限公司 | A kind of Camera Self-Calibration method based on SFM three-dimensional reconstruction |
-
2019
- 2019-10-14 CN CN201910976937.5A patent/CN111145263A/en not_active Withdrawn
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102016104729A1 (en) * | 2016-03-15 | 2017-09-21 | Connaught Electronics Ltd. | Method for extrinsic calibration of a camera, computing device, driver assistance system and motor vehicle |
| CN106558080A (en) * | 2016-11-14 | 2017-04-05 | 天津津航技术物理研究所 | Join on-line proving system and method outside a kind of monocular camera |
| CN110223354A (en) * | 2019-04-30 | 2019-09-10 | 惠州市德赛西威汽车电子股份有限公司 | A kind of Camera Self-Calibration method based on SFM three-dimensional reconstruction |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113639782A (en) * | 2021-08-13 | 2021-11-12 | 北京地平线信息技术有限公司 | External parameter calibration method and device for vehicle-mounted sensor, equipment and medium |
| CN113420732A (en) * | 2021-08-23 | 2021-09-21 | 深圳市城市交通规划设计研究中心股份有限公司 | Pavement disease detection method and device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3751519B1 (en) | Method, apparatus, device and medium for calibrating pose relationship between vehicle sensor and vehicle | |
| CN108805934B (en) | External parameter calibration method and device for vehicle-mounted camera | |
| CN109631887B (en) | Inertial navigation high-precision positioning method based on binocular, acceleration and gyroscope | |
| CN107111879B (en) | Method and apparatus for estimating vehicle's own motion by panoramic looking-around image | |
| US11781863B2 (en) | Systems and methods for pose determination | |
| CN110842940A (en) | Building surveying robot multi-sensor fusion three-dimensional modeling method and system | |
| CN107677274B (en) | Unmanned plane independent landing navigation information real-time resolving method based on binocular vision | |
| CN111279354B (en) | Image processing method, apparatus and computer readable storage medium | |
| CN113850126A (en) | Target detection and three-dimensional positioning method and system based on unmanned aerial vehicle | |
| CN112789655A (en) | System and method for calibrating an inertial test unit and camera | |
| CN102692236A (en) | Visual milemeter method based on RGB-D camera | |
| CN112116651B (en) | Ground target positioning method and system based on monocular vision of unmanned aerial vehicle | |
| CN115187798A (en) | Multi-unmanned aerial vehicle high-precision matching positioning method | |
| CN111572633B (en) | Steering angle detection method, device and system | |
| CN113781562B (en) | Lane line virtual-real registration and self-vehicle positioning method based on road model | |
| US20220171060A1 (en) | Systems and methods for calibrating a camera and a multi-line lidar | |
| JP2023505891A (en) | Methods for measuring environmental topography | |
| WO2019061064A1 (en) | Image processing method and device | |
| CN107437264A (en) | In-vehicle camera external parameter automatic detection and bearing calibration | |
| CN116433737A (en) | Method and device for registering laser radar point cloud and image and intelligent terminal | |
| CN111145263A (en) | Vehicle-mounted-based automatic camera calibration method | |
| CN116385504A (en) | Inspection and ranging method based on unmanned aerial vehicle acquisition point cloud and image registration | |
| Huttunen et al. | A monocular camera gyroscope | |
| CN110197104B (en) | Distance measurement method and device based on vehicle | |
| CN111145262A (en) | Vehicle-mounted monocular calibration method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200512 |
|
| WW01 | Invention patent application withdrawn after publication |