CN112288743B - Automobile vision detection error evaluation method and system based on unconstrained light plane - Google Patents
Automobile vision detection error evaluation method and system based on unconstrained light plane Download PDFInfo
- Publication number
- CN112288743B CN112288743B CN202011335531.8A CN202011335531A CN112288743B CN 112288743 B CN112288743 B CN 112288743B CN 202011335531 A CN202011335531 A CN 202011335531A CN 112288743 B CN112288743 B CN 112288743B
- Authority
- CN
- China
- Prior art keywords
- camera
- light plane
- laser
- vision detection
- error evaluation
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 52
- 238000011156 evaluation Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims abstract description 6
- 238000007689 inspection Methods 0.000 claims abstract description 5
- 238000004458 analytical method Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000013519 translation Methods 0.000 claims description 10
- 238000009795 derivation Methods 0.000 claims description 6
- 229910000831 Steel Inorganic materials 0.000 claims description 3
- 239000010959 steel Substances 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses an automobile vision detection error evaluation method and system based on an unconstrained light plane, and aims to solve the problem of automobile vision detection error source evaluation based on the unconstrained light plane. The method for evaluating the automobile vision detection error based on the unconstrained light plane mainly comprises the steps of obtaining each factor affecting the detection precision of the automobile vision detection error evaluation method based on the unconstrained light plane, analyzing the image coordinate influence of laser characteristic points, analyzing the inner parameter K influence of a camera, analyzing the outer parameter influence of the camera, testing and verifying the accuracy of the automobile vision detection error evaluation method based on the unconstrained light plane and the like. The automobile vision detection error evaluation system based on the unconstrained light plane mainly comprises a scale (2). A method and system for evaluating performance-stable errors in an automotive vision inspection system based on unconstrained light planes is provided.
Description
Technical Field
The invention relates to a detection error evaluation method and a detection error evaluation system in the field of automobile detection, in particular to an automobile vision detection error evaluation method and a detection error evaluation system based on an unconstrained light plane.
Background
With the continuous development of detection technology in recent years, machine vision gradually becomes a key technology for improving automobile detection efficiency and guaranteeing automobile detection precision. The vehicle appearance detection based on the machine vision can provide important research basis for automatic identification of overrun and overload of the vehicle, classification of the vehicle type, detection of the size parameter of the whole vehicle, acquisition and reconstruction of vehicle information and the like. However, in order to solve the problem, the present invention provides an automobile vision detection error evaluation system and method based on an unconstrained light plane, which can analyze the influence factors which cause the reconstruction of laser feature point precision of the automobile vision detection system (publication number: CN 109540042A) based on the unconstrained light plane, explore the influence of each influence factor on the reconstruction precision of the automobile vision detection system based on the unconstrained light plane, and use the error evaluation system to perform corresponding experiment to verify the correctness of the error evaluation method.
Disclosure of Invention
Aiming at the lack of quantitative research on influence factors of detection precision in the process of detecting the appearance of an automobile by an automobile vision detection system based on an unconstrained light plane, the invention provides a method and a system with reliable performance, simple structure and simple and convenient operation, and the analysis of each factor influencing the precision of the reconstructed automobile feature points is realized. The system is mainly composed of a scale, and the method is composed of five steps. The influence of various factors such as camera internal parameters, external parameters, laser characteristic point image coordinates and the like of the automobile vision detection system based on the unconstrained light plane on reconstruction precision can be analyzed.
The invention is realized by adopting the following technical scheme in combination with the attached drawings of the specification:
the automobile vision detection error evaluation method based on the unconstrained light plane comprises the following specific steps of:
the first step: obtaining various factors affecting detection precision of an automobile vision detection error evaluation method based on an unconstrained light plane:
the three-dimensional coordinates of the laser characteristic points under the camera coordinate system areEach component is
Wherein,is->Corresponding image projection point coordinates +.>The elements of (a), (b), (c), (u), (v) being the matrix of parameters in the camera +.>Element of (a)>Is the laser plane coordinate under the camera coordinate system +.> Wherein->Is laser plane coordinate +.>Element of (a)>Is a camera external parameter rotation matrix +.>Element of (a)>Is the camera external parameter translation vector +.>Elements of (a) and (b);
and a second step of: the image coordinates of the laser feature points influence the precision of the three-dimensional coordinate values of the reconstructed laser feature points:
image coordinates of laser feature pointsElements in the direction of the horizontal axis->The derivation is carried out to obtain
Image sitting of laser feature pointsLabel (C)Elements in the direction of the longitudinal axis->The derivation is carried out to obtain
And a third step of: the three-dimensional coordinate value precision influence of the internal parameter K of the camera on the reconstructed laser characteristic point:
deriving alpha element in internal parameter K of camera to obtain
Deriving beta element in the internal parameter K of the camera to obtain
Deriving gamma element in internal parameter K of camera to obtain
Deriving u element in internal parameter K of camera internal parameter to obtain
Deriving v element in the internal parameter K of the camera to obtain
Fourth step: the external parameters of the camera influence the precision of the three-dimensional coordinate values of the reconstructed laser characteristic points:
rotation matrix R for external parameters of camera RC，q The nine elements in the three are derived to obtain
Translation vector t to external parameters of camera RC，q The three elements in the three are derived to obtain
Fifth step: test verification of accuracy of automobile vision detection error evaluation method based on unconstrained light plane:
the camera support of the automobile vision detection system based on the unconstrained light plane is placed on the ground, the two-dimensional target plate is placed on the q-th position, the distances between the two-dimensional target plate and the camera are different in different positions, when the two-dimensional target plate moves, the position of the scale is adjusted, the checkerboard feature points on the scale are always intersected with the laser plane, the camera continuously collects the image of the q-th position, and q=1, 2, the number of the points is n; according to q images acquired by the camera, the standard length of the two-dimensional target plate on the scale with different distances from the camera can be reconstructed, and the translation vector t of the camera in the third step can be verified by comparing the reconstructed length with the actual standard length RC，q The accuracy of derivative analysis is carried out on the three elements in the process.
The automobile vision detection error evaluation system based on the unconstrained light plane comprises a scale.
In the technical scheme, the scale is a rectangular steel plate, and the inner surface is stuck with a regular checkerboard pattern.
The beneficial effects of the invention are as follows:
(1) The system has the advantages of wide measurement range, reliable performance, simple structure, simple and convenient operation and wide application range, and can verify the correctness of the error evaluation method by only adding one scale 2 on the original detection system.
(2) The invention provides an error analysis and evaluation method for the automobile vision detection system based on the unconstrained light plane, can provide a theoretical basis for analyzing test errors and provides a reference basis for other active vision detection error evaluation methods.
(3) According to the invention, after the original detection test, the laser plane is intersected with the chessboard lines on the scale 2, the standard distance on the scale 2 is reconstructed under the camera coordinate system, and the accuracy of the error evaluation method is verified by comparing the actual standard distance with the reconstructed distance.
Drawings
FIG. 1 is a flow chart of solving three-dimensional coordinates of laser feature points in a camera coordinate system in an automobile vision detection error evaluation method based on an unconstrained light plane;
FIG. 2 is a flow chart of an analysis of the effect of image coordinates on reconstructed three-dimensional coordinates of feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 3 is a flow chart of an analysis of the impact of focal length elements in camera internal parameters on reconstructed three-dimensional coordinates of feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 4 is a flow chart of an analysis of the influence of distortion elements in parameters in a camera on three-dimensional coordinates of reconstructed feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 5 is a flow chart of an analysis of the influence of principal point coordinate elements in parameters in a camera on three-dimensional coordinates of reconstructed feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 6 is a flow chart of an analysis of the influence of elements in the camera external parameter rotation matrix on the reconstructed three-dimensional coordinates of feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 7 is a flow chart of an analysis of the influence of elements in the camera external parameter translation vector on the reconstructed three-dimensional coordinates of feature points in an automobile vision detection error evaluation method based on unconstrained light planes;
FIG. 8 is an isometric view of an automobile vision inspection error assessment method based on unconstrained light planes;
FIG. 9 is an isometric view of the scale 2 in the method of evaluating automobile visual inspection errors based on unconstrained light planes;
in the figure: 1. and 2, an automobile vision detection system based on an unconstrained light plane, and a scale.
Detailed Description
The invention is described in further detail below with reference to the attached drawing figures:
referring to fig. 1 to 7, the method for evaluating the visual inspection error of the automobile based on the unconstrained light plane can be divided into the following five steps:
the first step: obtaining various factors affecting detection precision of an automobile vision detection error evaluation method based on an unconstrained light plane:
the three-dimensional coordinates of the laser characteristic points under the camera coordinate system areEach component is
Wherein,is->Corresponding image projection point coordinates +.>The elements of (a), (b), (c), (u), (v) being the matrix of parameters in the camera +.>Element of (a)>Is the laser plane coordinate under the camera coordinate system +.> Wherein->Is laser plane coordinate +.>Element of (a)>Is a camera external parameter rotation matrix +.>Element of (a)>Is the camera external parameter translation vector +.>Elements of (a) and (b);
and a second step of: the image coordinates of the laser feature points influence the precision of the three-dimensional coordinate values of the reconstructed laser feature points:
image coordinates of laser feature pointsElements in the direction of the horizontal axis->The derivation is carried out to obtain
Image coordinates of laser feature pointsElements in the direction of the longitudinal axis->The derivation is carried out to obtain
And a third step of: the three-dimensional coordinate value precision influence of the internal parameter K of the camera on the reconstructed laser characteristic point:
deriving alpha element in internal parameter K of camera to obtain
Deriving beta element in the internal parameter K of the camera to obtain
Deriving gamma element in internal parameter K of camera to obtain
Deriving u element in internal parameter K of camera internal parameter to obtain
Deriving v element in the internal parameter K of the camera to obtain
Fourth step: the external parameters of the camera influence the precision of the three-dimensional coordinate values of the reconstructed laser characteristic points:
rotation matrix R for external parameters of camera RC，q The nine elements in the three are derived to obtain
Translation vector t to external parameters of camera RC，q The three elements in the three are derived to obtain
Fifth step: test verification of accuracy of automobile vision detection error evaluation method based on unconstrained light plane:
the camera support of the automobile vision detection system 1 based on the unconstrained light plane is placed on the ground, a two-dimensional target plate is placed at the q-th position, the distances between the two-dimensional target plate and a camera are different in different positions, when the two-dimensional target plate moves, the position of the scale 2 is adjusted, so that checkerboard feature points on the scale 2 always intersect with a laser plane, the camera continuously collects images at the q-th position, q=1, 2, the number of the points is n; according to q images acquired by the camera, the standard length of the two-dimensional target plate on the scale 2 when the distance between the target plate and the camera is different can be reconstructed, and the translation vector t of the camera in the third step can be verified by comparing the reconstructed length with the actual standard length RC，q The accuracy of derivative analysis is carried out on the three elements in the process.
Referring to fig. 8 to 9, an automobile vision inspection error evaluation system based on an unconstrained light plane includes a scale 2.
The scale 2 is a rectangular steel plate, and the inner surface is stuck with a regular checkerboard pattern.
Claims (2)
1. The automobile vision detection error evaluation method based on the unconstrained light plane is characterized by comprising the following specific steps of:
the first step: obtaining various factors affecting detection precision of an automobile vision detection error evaluation method based on an unconstrained light plane:
the three-dimensional coordinates of the laser characteristic points under the camera coordinate system areEach component is
Wherein,is->Corresponding image projection point coordinates +.>The elements of (a), (b), (c), (u), (v) being the matrix of parameters in the camera +.>Element of (a)>Is the laser plane coordinate under the camera coordinate system +.> Wherein->Is laser plane coordinate +.>Element of (a)>Is a camera external parameter rotation matrix +.>Element of (a)>Is the camera external parameter translation vector +.>Elements of (a) and (b);
and a second step of: the image coordinates of the laser feature points influence the precision of the three-dimensional coordinate values of the reconstructed laser feature points:
image coordinates of laser feature pointsElements in the direction of the horizontal axis->The derivation is carried out to obtain
Image coordinates of laser feature pointsElements in the direction of the longitudinal axis->The derivation is carried out to obtain
And a third step of: the three-dimensional coordinate value precision influence of the internal parameter K of the camera on the reconstructed laser characteristic point:
deriving alpha element in internal parameter K of camera to obtain
Deriving beta element in the internal parameter K of the camera to obtain
Deriving gamma element in internal parameter K of camera to obtain
Deriving u element in internal parameter K of camera internal parameter to obtain
Deriving v element in the internal parameter K of the camera to obtain
Fourth step: the external parameters of the camera influence the precision of the three-dimensional coordinate values of the reconstructed laser characteristic points:
rotation matrix R for external parameters of camera RC，q The nine elements in the three are derived to obtain
Translation vector t to external parameters of camera RC，q The three elements in the three are derived to obtain
Fifth step: test verification of accuracy of automobile vision detection error evaluation method based on unconstrained light plane:
the camera support of the automobile vision detection system (1) based on the unconstrained light plane is placed on the ground, the two-dimensional target plate is placed on the q-th position, the distances between the two-dimensional target plate and a camera are different in different positions, and when the two-dimensional target plate moves, the position of the scale (2) is adjusted to enable the scale (2) to be arranged onThe checkerboard feature points of (a) always intersect the laser plane, and the camera continues to acquire images of the q-th position, q=1, 2, n; according to q images acquired by the camera, the standard length of the two-dimensional target plate on the scale (2) with different distances from the camera can be reconstructed, and the translation vector t of the camera in the third step can be verified by comparing the reconstructed length with the actual standard length RC，q The accuracy of derivative analysis is carried out on the three elements in the process.
2. The evaluation system of the automobile vision inspection error evaluation method based on the unconstrained light plane according to claim 1, wherein the evaluation system of the automobile vision inspection error evaluation method based on the unconstrained light plane comprises a scale (2); the scale (2) is a rectangular steel plate, and the inner surface is stuck with a regular checkerboard pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011335531.8A CN112288743B (en) | 2020-11-22 | 2020-11-22 | Automobile vision detection error evaluation method and system based on unconstrained light plane |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011335531.8A CN112288743B (en) | 2020-11-22 | 2020-11-22 | Automobile vision detection error evaluation method and system based on unconstrained light plane |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112288743A CN112288743A (en) | 2021-01-29 |
| CN112288743B true CN112288743B (en) | 2024-02-13 |
Family
ID=74425308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011335531.8A Active CN112288743B (en) | 2020-11-22 | 2020-11-22 | Automobile vision detection error evaluation method and system based on unconstrained light plane |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112288743B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109540040A (en) * | 2019-01-14 | 2019-03-29 | 吉林大学 | Based on without constraint homocentric beam race automobile pattern Active visual inspection System and method for |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107121109B (en) * | 2017-06-12 | 2019-12-06 | 北京航空航天大学 | structural optical parameter calibration device and method based on front coated plane mirror |
-
2020
- 2020-11-22 CN CN202011335531.8A patent/CN112288743B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109540040A (en) * | 2019-01-14 | 2019-03-29 | 吉林大学 | Based on without constraint homocentric beam race automobile pattern Active visual inspection System and method for |
Non-Patent Citations (1)
| Title |
|---|
| 汽车车轮定位参数视觉测量模型的建立与检测方法;徐观;汽车技术;20191231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112288743A (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6539107B1 (en) | Machine vision method using search models to find features in three-dimensional images | |
| CN109883533B (en) | Low-frequency vibration measurement method based on machine vision | |
| US8111907B2 (en) | Method for repeatable optical determination of object geometry dimensions and deviations | |
| US7869026B2 (en) | Targeted artifacts and methods for evaluating 3-D coordinate system measurement accuracy of optical 3-D measuring systems using such targeted artifacts | |
| CN109087274A (en) | Electronic device defect inspection method and device based on multidimensional fusion and semantic segmentation | |
| CN108535097A (en) | A kind of method of triaxial test sample cylindrical distortion measurement of full field | |
| CN102207371A (en) | Three-dimensional point coordinate measuring method and measuring apparatus thereof | |
| CN107341824B (en) | Comprehensive evaluation index generation method for image registration | |
| CN113008158B (en) | Multi-line laser tire pattern depth measuring method | |
| CN109410175B (en) | SAR radar imaging quality rapid automatic evaluation method based on multi-subregion image matching | |
| CN111369484B (en) | Rail profile detection method and device | |
| CN110044291A (en) | A kind of method of camera battle array measurement local deformation | |
| US6898333B1 (en) | Methods and apparatus for determining the orientation of an object in an image | |
| CN112288743B (en) | Automobile vision detection error evaluation method and system based on unconstrained light plane | |
| CN112381892B (en) | Automobile vision detection error evaluation method and system for unconstrained concentric beam group | |
| CN113983951B (en) | Three-dimensional target measuring method, device, imager and storage medium | |
| US7257248B2 (en) | Non-contact measurement system and method | |
| CN115685164A (en) | Three-dimensional laser imager working parameter testing system and method | |
| CN108844977B (en) | Industrial CT system spatial resolution testing method and evaluation method capable of eliminating angle inclination influence | |
| CN114061480B (en) | Method for detecting appearance of workpiece | |
| Shao et al. | High-Speed and Accurate Method for the Gear Surface Integrity Detection Based on Visual Imaging | |
| CN110738180B (en) | Method for evaluating signal accuracy and system precision in detection process | |
| CN113503830B (en) | Aspheric surface shape measuring method based on multiple cameras | |
| CN112902878B (en) | Method and device for adjusting laser plane of track geometry detection system | |
| CN114897892B (en) | Method for calculating characteristic parameters of apparent cracks and holes of PC (polycarbonate) member |
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 |