CN111784643A - Tire tangent plane obtaining method and system based on cross line structured light - Google Patents
Tire tangent plane obtaining method and system based on cross line structured light Download PDFInfo
- Publication number
- CN111784643A CN111784643A CN202010523781.8A CN202010523781A CN111784643A CN 111784643 A CN111784643 A CN 111784643A CN 202010523781 A CN202010523781 A CN 202010523781A CN 111784643 A CN111784643 A CN 111784643A
- Authority
- CN
- China
- Prior art keywords
- light
- coordinate system
- plane
- image
- tire
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a method, a system and a storage medium for obtaining a tire tangent plane based on cross line structured light, belonging to the technical field of tire tangent plane calibration. A tire tangent plane obtaining method based on cross line structured light comprises the following steps: acquiring a light strip image of a crossed laser line on a calibration plate, denoising the light strip image, segmenting the denoised light strip image, extracting an interested region of a light strip, and acquiring the center of the light strip on the calibration plate according to the interested region; establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system; acquiring a light plane according to the conversion relation between the light bar center and the object space coordinate system and the conversion relation between the image space coordinate system and the object space coordinate system; and acquiring a characteristic positioning point, and fitting a tire tangent plane to be measured through the characteristic positioning point. The accuracy and the real-time performance of obtaining the tire tangent plane are improved.
Description
Technical Field
The invention relates to the technical field of tire tangent plane calibration, in particular to a tire tangent plane acquisition method and system based on cross line structured light and a storage medium.
Background
In the use process of the automobile, the steering mechanism, the axle and the frame are deformed and worn, so that the operating performance of the automobile is poor, and driving accidents are easy to occur; in addition, the rolling resistance of the wheels is increased, the dynamic property is reduced, and the running oil consumption is increased. The resulting abnormal wear of the tires also reduces the economy of use of the vehicle. Therefore, in order to maintain the straight-line driving stability and the operation portability of the automobile and reduce the abrasion of the tire and related parts thereof, the wheel needs to be periodically detected, the tire needs to be accurately positioned, and the tangent plane of the tire needs to be obtained.
Disclosure of Invention
In view of the above, the invention provides a tire tangent plane obtaining method, system and storage medium based on cross line structured light, which solve the technical problems of low precision and poor real-time performance in the existing tire tangent plane calibration scheme.
In one aspect, the invention provides a tire tangent plane obtaining method based on cross line structured light, which comprises the following steps:
acquiring a light strip image of a crossed laser line on a calibration plate, denoising the light strip image, segmenting the denoised light strip image, extracting an interested region of a light strip, and acquiring the center of the light strip on the calibration plate according to the interested region;
establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system;
acquiring a light plane according to the conversion relation between the light bar center and the object space coordinate system and the conversion relation between the image space coordinate system and the object space coordinate system;
the method comprises the steps of projecting light with a cross line structure on the side face of a tire to be tested to form a light strip, extracting the light strip center of the side face of the tire to be tested through the light strip, obtaining a light plane equation of the light strip according to the light plane, obtaining a characteristic curve corresponding to the light strip center according to the light strip center and the light plane equation, obtaining a characteristic locating point according to the characteristic curve, and fitting a tangent plane of the tire to be tested through the characteristic locating point.
Further, acquiring the center of the light bar according to the region of interest, specifically including performing gaussian filtering on the region of interest, calculating a first-order partial derivative of a point in the region of interest, acquiring a normal vector of the region of interest, acquiring a coordinate correction value of the point according to the first-order partial derivative and the correction coefficient, determining whether the point is the center point of the light bar according to the coordinate correction value, and acquiring the center points of all the light bars in the region of interest to form the center of the light bar.
Further, establishing a camera imaging model to obtain a conversion relation between an image space coordinate system and an object space coordinate system, specifically comprising,
establishing a camera imaging model, wherein in the camera imaging model, an object point M, an image point M and a photographing center O are expressed by a formula
Representing, wherein M and M are homogeneous coordinates of the image point and the object point in respective coordinate systems, R, T are a rotation matrix and a translation matrix, s is a scale factor converted by the coordinate systems, and A is an internal reference matrix of the camera;
acquiring an internal reference matrix, a radial distortion parameter and a tangential distortion parameter of a camera, determining a conversion relation of coordinates in an image plane coordinate system and an image space coordinate system according to the internal reference matrix, the radial distortion parameter and the tangential distortion parameter of the camera, and determining and acquiring a conversion relation of the image space coordinate system and an object space coordinate system according to the conversion relation of the coordinates in the image plane coordinate system and the image space coordinate system.
Further, the method for obtaining the tire tangent plane of the cross structured light further comprises the steps of establishing a transformation plane checkerboard target, transforming the posture of the transformation plane checkerboard target in the space, acquiring an image of the structured light projected on the checkerboard target, extracting checkerboard angular points in the image, obtaining a selection matrix and a translation matrix according to the checkerboard angular points in the image, and obtaining plane equation parameters of the plane checkerboard target in an image space coordinate system according to the selection matrix and the translation matrix.
Further, according to the selection matrix and the translation matrix, obtaining plane equation parameters of the planar checkerboard target in an image space coordinate system, and according to a formula
Plane equation parameters A, B, C, D are obtained, where R, T are the rotation matrix and the translation matrix, respectively.
The method specifically comprises the steps of fitting the light strip centers to form light strip line segments, obtaining an equation of the light strip line segments in an image space coordinate, randomly taking a set number of points on each light strip line segment as characteristic points, obtaining object point coordinates corresponding to characteristic points according to the conversion relation of the image space coordinate system and an object space coordinate system and plane equation parameters of a planar checkerboard target in the image space coordinate system, and fitting a light plane according to the object point coordinates corresponding to the characteristic points.
Further, a characteristic curve corresponding to the light strip center is obtained according to the light strip center and a light plane equation, a characteristic positioning point is obtained according to the characteristic curve, a tire tangent plane to be measured is fitted through the characteristic positioning point, and the method specifically comprises the following steps of,
obtaining a characteristic curve of the light strip of the tire according to the light strip center and the light plane equation of the light strip, and performing space ellipse fitting on the characteristic curve of the light strip of the tire to obtain an ellipse equation; and acquiring a common tangent point of the ellipse according to the ellipse equation, acquiring a characteristic positioning point according to the common tangent point of the ellipse, and fitting a tangent plane of the tire to be measured through the characteristic positioning point.
On the other hand, the invention also provides a tire tangent plane acquiring system based on the cross structured light, which comprises a processor and a memory, wherein the memory is stored with a computer program, and when the computer program is executed by the processor, the tire tangent plane acquiring method based on the cross structured light is realized.
In another aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for obtaining a tire tangent plane based on cross-line structured light according to any one of the above-mentioned technical solutions.
Compared with the prior art, the invention has the beneficial effects that: the method comprises the steps of denoising an optical strip image of a cross laser line on a calibration plate by obtaining the optical strip image, segmenting the denoised optical strip image, extracting an interested area of an optical strip, and obtaining the center of the optical strip on the calibration plate according to the interested area; establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system; acquiring a light plane according to the conversion relation between the light bar center and the object space coordinate system and the conversion relation between the image space coordinate system and the object space coordinate system; projecting the light with the cross line structure on the side face of the tire to be tested to form a light strip, extracting the light strip center of the side face of the tire to be tested through the light strip, acquiring a light plane equation of the light strip according to the light plane, acquiring a characteristic curve corresponding to the light strip center according to the light strip center and the light plane equation, acquiring a characteristic positioning point according to the characteristic curve, and fitting a tangent plane of the tire to be tested through the characteristic positioning point; the accuracy and the real-time performance of obtaining the tire tangent plane are improved.
Drawings
Fig. 1 is a schematic flow chart of a tire tangential plane acquisition method based on cross-line structured light according to embodiment 1 of the present invention;
fig. 2 is a schematic flow chart of acquiring the center of a light bar according to embodiment 1 of the present invention;
FIG. 3 is a schematic diagram of image acquisition according to embodiment 1 of the present invention;
FIG. 4 is a schematic view of image exposure according to embodiment 1 of the present invention;
FIG. 5 is a schematic view of the cross-shaped structure light projected on the tire side according to example 1 of the present invention;
fig. 6 is a schematic diagram of coordinate system conversion according to embodiment 1 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
The embodiment of the invention provides a tire tangent plane obtaining method based on cross line structured light, which comprises the following steps of:
step S1, acquiring a light strip image of a crossed laser line on a calibration plate, denoising the light strip image, segmenting the denoised light strip image, extracting an interesting region of a light strip, and acquiring the center of the light strip on the calibration plate according to the interesting region;
s2, establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system;
step S3, acquiring a light plane according to the conversion relation between the light bar center and the space coordinate system and the object coordinate system;
step S4, projecting the light with the cross line structure on the side surface of the tire to be measured to form a light strip, extracting the light strip center of the side surface of the tire to be measured through the light strip, obtaining a light plane equation of the light strip according to the light plane, obtaining a characteristic curve corresponding to the light strip center according to the light strip center and the light plane equation, obtaining a characteristic positioning point according to the characteristic curve, and fitting the tangent plane of the tire to be measured through the characteristic positioning point.
Preferably, the obtaining of the light bar center according to the region of interest specifically includes performing gaussian filtering on the region of interest, calculating a first-order partial derivative of a point in the region of interest, obtaining a normal vector of the region of interest, obtaining a coordinate correction value of the point according to the first-order partial derivative and a correction coefficient, determining whether the point is a light bar center point according to the coordinate correction value, and obtaining all light bar center points in the region of interest to form the light bar center.
In a specific embodiment, a schematic flow chart of obtaining the center of a light bar is shown in fig. 2, where the flow of obtaining the center of the light bar includes step S10, inputting an image, S11, binarizing and segmenting to extract an ROI, S12, gaussian filtering, S13, calculating first and second partial derivatives of an undetermined point, S14, calculating a law vector, S15, calculating a correction coefficient, S16, calculating a coordinate correction value, S17, determining whether (dx, dy) [ -e-0.5, 0.5] × [ -0.5,0.5] holds, if not, performing step S13, if yes, performing step S18, determining the determined center point, step S19, determining whether all points are involved, if yes, performing step S13, if yes, performing S20, and determining a series of center points of the light bar;
the flow chart for obtaining the center of the light bar is shown in FIG. 2, and the normal direction of the light bar is determined by the following matrix
In the above formula, g (x, y) represents a two-dimensional Gaussian function, z (x, y) represents an image,
representing convolution, a point (x) on a light bar0,y0) The normal direction of the point (x) is determined by the eigenvector corresponding to the absolute value of the maximum eigenvalue of the Hessian matrix of the image z (x, y) at the point, then the coordinate correction value is calculated, and the point (x) is set0,y0) The unit vector in the normal direction of the light stripe obtained from the Hessian matrix is (n)x,ny) The coordinate correction value is expressed as (dx, dy) ═ tnx,tny) Wherein t is a correction coefficient,
to point (x)0,y0) Is subjected to second-order Taylor expansion to obtain
Wherein,
the first partial derivatives of the image z (x, y) after gaussian convolution are obtained, and (dx, dy) is (tn)x,tny) Can be obtained by substituting the above formula
Obtaining t expression by taking derivative of z (t) as zero and obtaining extremum
If (dx, dy) ∈ [ -0.5,0.5 [ ]]×[-0.5,0.5]Then point (x)0,y0) Can be identified as the center point of the light strip and corrected to (x)0+dx,y0+ dy), calculating each preliminarily extracted light bar point to obtain a series of light bar central points with sub-pixel precision, segmenting the image by binarization, extracting an ROI (region of interest) of the light bars, and extracting the light bar centers only in the ROI by a Steger algorithm; the ROI basically only contains information of light bars, little salt and pepper noise exists, and much Gaussian noise exists, so that Gaussian filtering can well reduce noise of an image;
preferably, the establishing of the camera imaging model and the obtaining of the transformation relation between the image space coordinate system and the object space coordinate system specifically comprise,
establishing a camera imaging model, wherein in the camera imaging model, an object point M, an image point M and a photographing center O are expressed by a formula
Representing, wherein M and M are homogeneous coordinates of the image point and the object point in respective coordinate systems, R, T are a rotation matrix and a translation matrix, s is a scale factor converted by the coordinate systems, and A is an internal reference matrix of the camera;
acquiring an internal reference matrix, a radial distortion parameter and a tangential distortion parameter of a camera, determining a conversion relation of coordinates in an image plane coordinate system and an image space coordinate system according to the internal reference matrix, the radial distortion parameter and the tangential distortion parameter of the camera, and determining and acquiring a conversion relation of the image space coordinate system and an object space coordinate system according to the conversion relation of the coordinates in the image plane coordinate system and the image space coordinate system.
In one embodiment, in the pinhole camera model, the object point M, the image point M and the photographing center O are collinear, and their relationship can be expressed by the following formula
In the above formula, m ═ u v 1]T，M＝[X Y 0 1]TRespectively representing homogeneous coordinates of the image point and the object point in respective coordinate systems (the object coordinate system can set the plane target to be an X-Y plane, and the Z component of the object coordinate is 0), R and T respectively represent a rotation matrix of 3 × 3 and a translation matrix of 3 × 1, which are used for conversion between the object coordinate system and the image space coordinate system, s is a scale factor of coordinate system conversion, A is an internal reference matrix of the camera, reflecting the conversion relation between the image plane coordinate system and the image space coordinate system, and depending on the relative positions of the camera lens and the camera light-sensing plate and the relative positions of the camera lens and the camera light-sensing plate, the specific expressions are as follows,
where f is the focal length of the camera and (dx, dy) represents the physical length of the pixel in the X, Y directions on the camera plate, respectively, (u, Y)0,v0) Representing the coordinates of the image principal point in an image plane coordinate system;
in addition, the picture taken by the camera has certain distortion, the distortion model comprises radial distortion and tangential distortion,
in the above formula, (x, y) represents the image side coordinates of the image point (the origin of the coordinate system is the image principal point), (x)correct,ycorrect) Is the coordinates after the distortion is corrected,
kiand piRadial distortion parameters and tangential distortion parameters, respectively;
internal reference matrix A and distortion parameter ki、piThe intrinsic parameters of the camera are generally called, and the intrinsic parameters of the camera are determined by camera calibration; once the internal parameters of the camera are determined, the conversion relation of the coordinates in the image plane coordinate system and the image space coordinate system can be actually determined, then a linear equation of the object point, the image point and the photographing center in the image space coordinate system can be obtained, and the conversion relation of the image space coordinate system and the object space coordinate system, namely R and T, can be determined only by a series of corresponding object point coordinates and image point coordinates (not less than 3 pairs) and through space backward intersection or a PnP algorithm; in this embodiment, the object space coordinate system is established on the planar target, so that a planar equation of the planar target in the image space coordinate system can be obtained;
preferably, the method for obtaining the tire tangent plane based on the cross-line structured light further includes the steps of establishing a transformation plane checkerboard target, transforming the posture of the transformation plane checkerboard target in the space, acquiring an image of the structured light projected on the checkerboard target, extracting checkerboard angular points in the image, obtaining a selection matrix and a translation matrix according to the checkerboard angular points in the image, and obtaining plane equation parameters of the transformation plane checkerboard target in an image space coordinate system according to the selection matrix and the translation matrix.
Preferably, according to the selection matrix and the translation matrix, obtaining plane equation parameters of the planar checkerboard target in the image space coordinate system according to a formula
Acquiring plane equation parameters A, B, C, D, wherein R, T is a rotation matrix and a translation matrix respectively;
preferably, a characteristic curve corresponding to the light strip center is obtained according to the light strip center and the light plane equation, a characteristic positioning point is obtained according to the characteristic curve, and a tire tangent plane to be measured is fitted through the characteristic positioning point,
obtaining a characteristic curve of the light strip of the tire according to the light strip center and the light plane equation of the light strip, and performing space ellipse fitting on the characteristic curve of the light strip of the tire to obtain an ellipse equation; acquiring a common tangent point of an ellipse according to the ellipse equation, acquiring a characteristic positioning point according to the common tangent point of the ellipse, and fitting a tangent plane of the tire to be measured through the characteristic positioning point;
in a specific embodiment, a transformation plane checkerboard target is established, the posture of the transformation plane checkerboard target in the space is changed, and simultaneously, an image of the structured light projected on the checkerboard target is acquired, and an image acquisition schematic diagram is shown in fig. 3, three images are acquired each time, fig. 3(b) and fig. 3(c) are acquired under the same exposure, so that the imaging effect of the structured light on the checkerboard is ensured, and the accuracy of the center of the light bar is improved, the initial position of the light bar is extracted by subtracting the images of fig. 3(c) and fig. 3(b), so that the image segmentation is facilitated, fig. 3(a) is mainly used for extracting angular points to calculate the plane position of the target, and the exposure is adjusted to ensure that the black squares and the white squares are as large as each other, so that the extraction accuracy of the angular points of the checkerboard and the image exposure schematic diagram are ensured, as shown in fig., FIG. 4(a) shows underexposure, and FIG. 4(b) shows overexposure;
extracting angular points in the graph 3(a) for each group of data, calculating a rotation matrix and a translation matrix through a PnP algorithm or single image space back intersection, and then calculating a rotation matrix and a translation matrix through a formula
Calculating plane equation parameters A, B, C, D for the target plane;
extracting the center of a structured light stripe, fitting a straight line to obtain an equation of the light stripe straight line in an image coordinate system (image space coordinate system), subdividing each light stripe line segment into a plurality of small line segments, randomly selecting a point on each small line segment as a characteristic point, and obtaining the characteristic point through a formula
And calculating object point coordinates corresponding to the characteristic points, and finally fitting all the characteristic points to a light plane.
Preferably, a characteristic curve is extracted through the laser stripes, a characteristic positioning point is obtained according to the characteristic curve and the light plane, a tire cutting plane to be measured is fitted through the characteristic positioning point, and the method specifically comprises the following steps of,
and carrying out image segmentation on the laser stripes, extracting a characteristic curve, carrying out ellipse fitting on the characteristic curve of the laser stripes according to the characteristic curve and the light plane to obtain an ellipse equation, obtaining tangent points of the ellipse equation as characteristic positioning points, and fitting out the tangent plane of the tire to be measured through the characteristic positioning points.
In a specific embodiment, cross-shaped structured light is projected on the side surface of a tire to form 4 laser stripes, the schematic diagram of the cross-shaped structured light projected on the side surface of the tire is shown in fig. 5, a characteristic curve is extracted through the laser stripes, then an ellipse fitting is performed to extract a characteristic positioning point (tangent point), and finally a wheel positioning surface is fitted through 4 characteristic points;
as shown in fig. 5, the laser stripes are projected on the tire and the rim of the tire by the structured light, the tire positioning surface is determined by the extracted feature positioning points on the tire, and the laser stripes on the rim make the subsequent image processing difficult, so that the two are required to be segmented; when the structured light is not projected, the gray values of the tire and the steel ring are obviously different, so that the tire and the steel ring can be segmented by adopting a binarization method, ideally, a laser is closed to additionally collect an image without laser stripes, the image is firstly binarized to segment the steel ring and the tire, the process needs to switch the laser on and off frequently, and the service life of the tire can be greatly reduced;
in the embodiment, the fact that the tire and the steel ring are made of different materials is taken into consideration, the main material of the automobile tire is rubber, the color is black, the reflection of light is weak, and the automobile tire belongs to a diffuse reflection surface, so that the projected structured light striation has low noise and is thinner, in addition, the gray value of the striation is smaller than that of the laser striation on the steel ring, the steel ring is mainly made of metal materials, the surface of the steel ring is smooth, the reflection capability is strong, so that a large amount of noise exists on two sides of the laser striation, and the striation becomes very thick due to the problem of a visual angle in places with large curvature change;
it is noted that the laser light bars of the two laser light bars have difference not only in gray value but also in thickness, so the influence of the light bars is eliminated by adjusting the exposure of the camera, two images are collected, one image is normally exposed (the thickness of the light bar is moderate, the noise is large when the thickness of the light bar is too thick, the center of the light bar is not obvious, and the light bar is broken when the thickness of the light bar is too thin), the other image is a low-exposure image, the laser light bar on the tire in the image can not be displayed due to the excessively low exposure, the low-exposure image is used for carrying out binarization segmentation to extract the outline of the steel ring, the light bar is too thick and noisy due to the high reflectivity of the steel ring at the junction of the steel ring and the tire, therefore, the area of the steel ring needs to be slightly expanded by adopting morphological expansion treatment, and in addition, partial miscellaneous points with low gray value caused by low exposure in the steel ring can be effectively eliminated by the expansion treatment; extracting steel ring outline points by using a self-contained function findContours in OpenCV, and performing ellipse fitting on the steel ring outline points (the steel ring is a circle, and an image is an ellipse after perspective transformation); only laser stripes outside the steel ring in the normal exposure image are reserved;
because the characteristic curve of the light bar is divided by the embodiment, the thickness of the light bar formed by the structured light on the tire is uniform, the noise is low, and the light bar is in sharp contrast with the background, the center of the light bar can be extracted by the STeger algorithm;
due to the shape characteristics of the tire, a characteristic curve formed by the structured light on the side surface of the tire is an elliptic arc, so that the tire can be subjected to elliptic fitting, and further, characteristic positioning points are extracted;
the ellipse equation can be expressed as
F(a，x)＝aTx＝nx2+bxy+cy2+dx+ey+f＝0
Wherein a ═ a b c d e f]T，x＝[x2xy y2x y f]TF (a, x) is the algebraic distance from point (x, y) to the ellipse F (a, x) equal to 0, and the objective of fitting the ellipse is to minimize the average sum of the algebraic distances of all the fitted points (n total)
In order to prevent a from being 0 and any multiple of a representing the same quadratic curve, a needs to be constrained, and a is usedTCa 1 constrains a to be a,
this is for b2-4ac<0, converting the final least square solution into generalized eigenvalue decomposition calculation, introducing Lagrange factor lambda and obtaining the derivative
DrDa-λCa＝0
Wherein D ═ x1 Tx2 Tx3 T… xn T]TThe above formula can be solved by generalized eigenvalue decomposition, let (λ)i,ui) Is a set of eigenvalue-eigenvector solutions therein, then (λ)i,μui) Also satisfies the above formula, substituting it into aTCa 1, available
H2ui rCui＝1
Then
Let a be mu uiThe parameters of the ellipse equation are obtained, and then the major axis and the minor axis of the ellipse, the coordinates of the center of the ellipse and the included angle between the major axis and the x axis of the ellipse can be deduced according to a formula;
specifically, the characteristic positioning point extraction process is as follows
A schematic diagram of coordinate system conversion, as shown in fig. 6, a normal vector of a light plane is v (X, y, Z), and a projection of the light plane on an X-Z plane of an image space coordinate system is l, wherein an included angle between v and the X-Z plane is β, and an included angle between the projection l and an X axis is α;
when the coordinate system conversion is needed, the x-y plane normal vector in the new coordinate system, namely the z axis is rotated to the direction of the valve vector v, and for this purpose, the coordinate system is firstly rotated clockwise around the y axis
Degree, rotated β degrees counterclockwise about the x-axis, the rotation matrix may be represented as
As can be seen from fig. 5, the tire totally corresponds to two light planes, each light plane includes two characteristic curves, the characteristic curves of the upper left corner and the lower right corner belong to the same light plane, and the characteristic curves of the lower left corner and the upper right corner belong to another light plane perpendicular to the characteristic curves; taking data processing of one light plane as an example, acquiring corresponding space points by the image coordinates through a triangulation principle, and enabling the z coordinates of the space points to be equal through coordinate system conversion, wherein two groups of space points are shared at the moment; the data respectively belong to two characteristic curves, and ellipse fitting is respectively carried out on the two characteristic curves to obtain two ellipse equations
The light plane comprises two characteristic curves, and the characteristic positioning points are determined by solving the common tangent of the two ellipses, so that the characteristics are tighter and more accurate;
when the common tangent equation is given as y ═ kx + m, the equation is substituted into the above equation and simplified, and the product can be obtained
Tangency of a straight line to an ellipse is equivalent to the above formula discrimination being 0, i.e.
After a common tangent is calculated, substituting an elliptic equation to calculate the coordinates of the tangent points; because two ellipses are separated from each other, 4 common tangent lines exist, the intersection point of the minor axis of the ellipse and the ellipse can be calculated firstly (note that the intersection point of the ellipses corresponding to the characteristic curves of the upper left corner and the lower right corner is in the lower left of the center of the ellipse, and the other two characteristic curves are opposite), then the distance between the tangent point obtained by the common tangent lines and the intersection point is calculated, and because the position of the intersection point is very close to the actual tangent point, the common tangent point corresponding to the closest common tangent line is selected as the final characteristic positioning point; the above is an ellipse fitted on one light plane, and the feature positioning points obtained according to the ellipse, 4 feature positioning points can be obtained on two light planes, and finally the tangent plane of the tire is fitted by the 4 feature positioning points.
Example 2
The embodiment of the invention discloses a tire tangent plane acquiring system based on cross structured light, which comprises a processor and a memory, wherein a computer program is stored on the memory, and when the computer program is executed by the processor, the tire tangent plane acquiring method based on cross structured light is realized.
Example 3
The embodiment of the invention discloses a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method for acquiring the tire tangent plane based on the cross-line structured light is realized.
The invention discloses a method, a system and a storage medium for obtaining a tire tangent plane based on cross line structured light, which are characterized in that a light strip image of a cross laser line on a calibration plate is obtained, the light strip image is denoised, the denoised light strip image is segmented, an interesting region of a light strip is extracted, and the light strip center on the calibration plate is obtained according to the interesting region; establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system; acquiring a light plane according to the conversion relation between the light bar center and the object space coordinate system and the conversion relation between the image space coordinate system and the object space coordinate system; projecting the light with the cross line structure on the side face of the tire to be tested to form a light strip, extracting the light strip center of the side face of the tire to be tested through the light strip, acquiring a light plane equation of the light strip according to the light plane, acquiring a characteristic curve corresponding to the light strip center according to the light strip center and the light plane equation, acquiring a characteristic positioning point according to the characteristic curve, and fitting a tangent plane of the tire to be tested through the characteristic positioning point; the accuracy and the real-time performance of obtaining the tire tangent plane are improved.
The above-described embodiments of the present invention should not be construed as limiting the scope of the present invention. Any other corresponding changes and modifications made according to the technical idea of the present invention should be included in the protection scope of the claims of the present invention.
Claims (9)
1. A tire tangent plane obtaining method based on cross line structured light is characterized by comprising the following steps:
acquiring a light strip image of a crossed laser line on a calibration plate, denoising the light strip image, segmenting the denoised light strip image, extracting an interested region of a light strip, and acquiring the center of the light strip on the calibration plate according to the interested region;
establishing a camera imaging model according to the calibration plate, and acquiring a conversion relation between an image space coordinate system and an object space coordinate system;
acquiring a light plane according to the conversion relation between the light bar center and the object space coordinate system and the conversion relation between the image space coordinate system and the object space coordinate system;
the method comprises the steps of projecting light with a cross line structure on the side face of a tire to be tested to form a light strip, extracting the light strip center of the side face of the tire to be tested through the light strip, obtaining a light plane equation of the light strip according to the light plane, obtaining a characteristic curve corresponding to the light strip center according to the light strip center and the light plane equation, obtaining a characteristic locating point according to the characteristic curve, and fitting a tangent plane of the tire to be tested through the characteristic locating point.
2. The cross-line structured light-based tire tangent plane obtaining method as claimed in claim 1, wherein obtaining the light bar centers according to the region of interest specifically comprises performing gaussian filtering on the region of interest, calculating a first partial derivative of a point in the region of interest, obtaining a normal vector of the region of interest, obtaining a coordinate correction value of the point according to the first partial derivative and a correction coefficient, determining whether the point is a light bar center point according to the coordinate correction value, and obtaining all the light bar center points in the region of interest to form the light bar centers.
3. The method for obtaining a tire tangent plane based on cross-line structured light as claimed in claim 1, wherein the camera imaging model is established to obtain a transformation relationship between an image space coordinate system and an object space coordinate system, specifically comprising,
establishing a camera imaging model, wherein in the camera imaging model, an object point M, an image point M and a photographing center O are expressed by a formula
Representing, wherein M and M are homogeneous coordinates of the image point and the object point in respective coordinate systems, R, T are a rotation matrix and a translation matrix, s is a scale factor converted by the coordinate systems, and A is an internal reference matrix of the camera;
acquiring an internal reference matrix, a radial distortion parameter and a tangential distortion parameter of a camera, determining a conversion relation of coordinates in an image plane coordinate system and an image space coordinate system according to the internal reference matrix, the radial distortion parameter and the tangential distortion parameter of the camera, and determining and acquiring a conversion relation of the image space coordinate system and an object space coordinate system according to the conversion relation of the coordinates in the image plane coordinate system and the image space coordinate system.
4. The cross structured light tire tangent plane obtaining method as claimed in claim 3, further comprising establishing a transformation plane checkerboard target, transforming the posture of the transformation plane checkerboard target in space, collecting the image of the structured light projected on the checkerboard target, extracting the checkerboard corner points in the image, obtaining a selection matrix and a translation matrix according to the checkerboard corner points in the image, and obtaining the plane equation parameters of the plane checkerboard target in the image space coordinate system according to the selection matrix and the translation matrix.
5. The cross-structured light tire tangent plane acquisition method according to claim 4, wherein the plane equation parameters of the planar checkerboard target in the image space coordinate system are acquired according to the selection matrix and the translation matrix, and the plane equation parameters are obtained according to the formula
Plane equation parameters A, B, C, D are obtained, where R, T are the rotation matrix and the translation matrix, respectively.
6. The method for obtaining a tire tangential plane of a cross structured light according to claim 4, wherein the light plane is obtained according to the conversion relationship between the light bar center and the object coordinate system and the space coordinate system, and specifically comprises fitting the light bar center to form light bar line segments, obtaining an equation of the light bar line segments in the image space coordinate system, randomly taking a set number of points on each light bar line segment as feature points, obtaining object point coordinates corresponding to the feature points according to the conversion relationship between the image space coordinate system and the object coordinate system and the plane equation parameters of the planar checkerboard target in the image space coordinate system, and fitting the object point coordinates corresponding to the feature points into the light plane.
7. The method for obtaining a tire tangential plane of a cross structured light as claimed in claim 6, wherein a characteristic curve corresponding to the light bar center is obtained according to the light bar center and the light plane equation, a characteristic positioning point is obtained according to the characteristic curve, and a tire tangential plane to be measured is fitted through the characteristic positioning point, specifically comprising,
obtaining a characteristic curve of the light strip of the tire according to the light strip center and the light plane equation of the light strip, and performing space ellipse fitting on the characteristic curve of the light strip of the tire to obtain an ellipse equation; and acquiring a common tangent point of the ellipse according to the ellipse equation, acquiring a characteristic positioning point according to the common tangent point of the ellipse, and fitting a tangent plane of the tire to be measured through the characteristic positioning point.
8. A cross structured light based tire tangent plane acquisition system, comprising a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to implement the cross structured light based tire tangent plane acquisition method according to any one of claims 1 to 7.
9. A computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the cross-line structured light based tire tangent plane acquisition method as recited in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010523781.8A CN111784643B (en) | 2020-06-10 | 2020-06-10 | Cross line structured light-based tire tangent plane acquisition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010523781.8A CN111784643B (en) | 2020-06-10 | 2020-06-10 | Cross line structured light-based tire tangent plane acquisition method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111784643A true CN111784643A (en) | 2020-10-16 |
| CN111784643B CN111784643B (en) | 2024-01-09 |
Family
ID=72755800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010523781.8A Active CN111784643B (en) | 2020-06-10 | 2020-06-10 | Cross line structured light-based tire tangent plane acquisition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111784643B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113008158A (en) * | 2021-03-25 | 2021-06-22 | 烟台大学 | Multi-line laser tyre pattern depth measuring method |
| CN113269207A (en) * | 2021-05-28 | 2021-08-17 | 天津理工大学 | Image feature point extraction method for grid structure light vision measurement |
| CN115421310A (en) * | 2022-09-01 | 2022-12-02 | 上海交通大学 | Rectangular wide light spot shaping imaging system and centroid positioning method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059340A (en) * | 2007-05-18 | 2007-10-24 | 东南大学 | Vehicle tread measurement method based on stereo vision and laser |
| CN104848801A (en) * | 2015-06-05 | 2015-08-19 | 北京航空航天大学 | Line structure light vision sensor calibration method based on parallel bicylindrical target |
| US20180147658A1 (en) * | 2016-11-25 | 2018-05-31 | Glowforge Inc. | Engraving in a computer numerically controlled machine |
-
2020
- 2020-06-10 CN CN202010523781.8A patent/CN111784643B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059340A (en) * | 2007-05-18 | 2007-10-24 | 东南大学 | Vehicle tread measurement method based on stereo vision and laser |
| CN104848801A (en) * | 2015-06-05 | 2015-08-19 | 北京航空航天大学 | Line structure light vision sensor calibration method based on parallel bicylindrical target |
| US20180147658A1 (en) * | 2016-11-25 | 2018-05-31 | Glowforge Inc. | Engraving in a computer numerically controlled machine |
Non-Patent Citations (4)
| Title |
|---|
| CHEN PING 等: "Sub-pixel exrtaction of laser stripe and its application in laser plane calibration", 《JOURNAL OF SOUTHEAST UNIVERSITY》, vol. 31, no. 1, pages 107 - 112 * |
| JIAGAR SENJALIA 等: "Measurement of wheel ailgnment using Camera Calibration and Laser Triangulation", 《2013 NIRMA UNIVERSITY INTERNATIONAL CONFERENCE ON ENGINEERING》, pages 1 - 5 * |
| 孙长库 等: "激光视觉车轮定位系统中胎冠3D坐标的求取", 《光电子·激光》, vol. 13, no. 9, pages 959 - 961 * |
| 魏嵬 等: "车轮定位参数的线激光非接触方法研究", 《光电子·激光》, vol. 15, no. 12, pages 1479 - 1482 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113008158A (en) * | 2021-03-25 | 2021-06-22 | 烟台大学 | Multi-line laser tyre pattern depth measuring method |
| CN113269207A (en) * | 2021-05-28 | 2021-08-17 | 天津理工大学 | Image feature point extraction method for grid structure light vision measurement |
| CN115421310A (en) * | 2022-09-01 | 2022-12-02 | 上海交通大学 | Rectangular wide light spot shaping imaging system and centroid positioning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111784643B (en) | 2024-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115096206B (en) | A high-precision measurement method for part dimensions based on machine vision | |
| CN109827502B (en) | High-precision calibration method for line-structured light vision sensor for calibration point image compensation | |
| CN108225319B (en) | Monocular vision rapid relative pose estimation system and method based on target characteristics | |
| CN105300316B (en) | Optical losses rapid extracting method based on grey scale centre of gravity method | |
| CN111415376B (en) | Automobile glass subpixel contour extraction method and automobile glass detection method | |
| CN111784643A (en) | Tire tangent plane obtaining method and system based on cross line structured light | |
| CN103047943A (en) | Method for detecting vehicle door outer plate shape and size based on single-projection encoding structured light | |
| CN115880373A (en) | Calibration plate and calibration method of stereoscopic vision system based on novel coding characteristics | |
| CN111640158A (en) | End-to-end camera based on corresponding mask and laser radar external reference calibration method | |
| CN115049718B (en) | Three-dimensional deformation monitoring method, device, computer equipment and storage medium | |
| CN111932494A (en) | Tire wear degree evaluation method and device | |
| CN112629409A (en) | Method for extracting line structure light stripe center | |
| CN112233063B (en) | Circle center positioning method for large-size round object | |
| CN113436207A (en) | Method for quickly and accurately extracting line structure light stripe center of regular surface | |
| CN113310426A (en) | Thread parameter measuring method and system based on three-dimensional profile | |
| TW201913574A (en) | Camera and laser range finder data fusion method used in object detection using a geometric feature and laser ranging data fusion technique to reconstruct a 3D size of an object | |
| Jia et al. | A field measurement method for large objects based on a multi-view stereo vision system | |
| CN108510544B (en) | A method of light strip location based on feature clustering | |
| CN113155057A (en) | Line structured light plane calibration method using non-purpose-made target | |
| CN111415378B (en) | Image registration method for automobile glass detection and automobile glass detection method | |
| CN117474873A (en) | Surface treatment system before brazing of high-chromium wear-resistant castings | |
| Winkelbach et al. | Shape from single stripe pattern illumination | |
| CN111798413B (en) | Tire tangential plane positioning method and system | |
| CN120298271A (en) | A method for dimensional correction of automobile parts based on image processing | |
| CN209279912U (en) | A kind of object dimensional information collecting device |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20231211 Address after: No. 005, Area A, 5th Floor, Beacon Innovation Valley Precision Craft Building, No. 88, Post University Road, Hongshan District, Wuhan City, Hubei Province 430070 Applicant after: Wuhan Jiayue Technology Co.,Ltd. Address before: No. 589 Qianyang Road, Meicheng Town, Qianshan County, Anqing City, Anhui Province, 246300 Applicant before: Zheng Chengyu |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |