Sheet metal part three-dimensional contour measurement method fusing surface structured light and luminosity stereo

Abstract

The invention discloses a three-dimensional sheet metal part contour measurement method fusing surface structured light and luminosity, which belongs to the field of three-dimensional measurement, and adopts a mode of combining multi-view illumination and surface structured light three-dimensional measurement, and more accurate normal vector information and an illumination model are used to measure the three-dimensional profile of a sheet metal part. The profile and the wall thickness part of the thin-wall component can be distinguished more accurately; edge enhancement is performed on the image by combining depth information acquired by surface structured light, so that the image contrast is enhanced, and meanwhile, the problem of image detail loss caused by overexposure is avoided; the incident direction of the light source is estimated through the normal vector and the gray value obtained through the surface structured light technology, the direction of the light source does not need to be calibrated, and the whole image enhancement process becomes simple and efficient; and a matching optimization energy function model based on phase consistency, visual angle consistency and contour continuity is constructed to obtain an optimal parallax, so that optimal stereo matching of corresponding contour points on the left and right cameras is realized.

Claims

1. A three-dimensional profile measurement method for sheet metal parts integrating surface structured light and photometric stereo, characterized in that: include: S1, projecting a grating fringe projection onto the surface of the workpiece to be measured, and synchronously collecting an image of the workpiece to be measured by the left and right cameras to obtain phase information of the workpiece to be measured; Matching the images of the workpiece to be measured collected by the left and right cameras according to the phase information to obtain depth information of the workpiece to be measured, and obtaining the normal N of each point on the surface of the workpiece to be measured according to the depth information; S2, respectively use m light sources to illuminate the workpiece to be tested, m ≥ 3, use the left camera to collect the image of the workpiece to be tested when each light source is irradiated, and use the formula according to the light intensity of each point in the image of the workpiece to be tested and the normal direction of each point on the surface of the workpiece to be tested. Calculate the illumination direction at each point on the surface of the workpiece to be measured Among them, U, Σ, and V are the upper triangular orthogonal matrix, diagonal matrix, and lower triangular orthogonal matrix obtained by SVD decomposition of matrix A, respectively. A=N T N,N=[n k ,n k ,…,n k ] T , is the pixel grayscale value corresponding to the kth point on the image of the workpiece to be measured collected by the left camera when the workpiece to be measured is illuminated by the jth light source, j=1, 2, ..., m; S3, the illumination direction of each point on the surface of the workpiece to be measured Taking the model, obtaining the reflectivity of each point on the surface of the workpiece to be measured, thereby obtaining a two-dimensional enhanced image of the workpiece to be measured; S4, using a level set contour model to obtain a two-dimensional contour curve of the characteristic hole position in the two-dimensional enhanced image; S5, establishing a phase consistency constraint according to the phase information of the workpiece to be measured, and establishing a matching optimization energy function based on the phase consistency constraint, the viewing angle consistency constraint and the contour continuity constraint; obtaining a disparity selection path that minimizes the matching optimization energy function, searching forward from the last point in the disparity selection path, and obtaining the optimal disparity of each point in the disparity selection path in combination with a backtracking equation, thereby achieving stereo matching of the two-dimensional contour curve in the image of the workpiece to be measured that is synchronously captured by the left and right cameras.

2. The method according to claim 1, characterized in that The matching optimization energy function is: Where d is the target parallax to be calculated, d(Q li )=Q ri -Q li , N(Q li ,Q li-1 ) is the contour continuity constraint, λ s , N are all weight coefficients, i=1, 2, ..., H, H is the total number of pixels of the two-dimensional contour curve; φ(x i ,y i ) is the point (x i ,y i ) on the left camera image, w is set to the pixel Q on the left camera li The size of the local window centered at I(x i ,y i ) is the point (x i ,y i ) on the left camera image, v is set to the pixel Q on the left camera li The size of the local window centered on the 3. The method according to claim 1 or 2, characterized in that The backtracking equation is: in, E(Q li ,d ij ) represents the current contour point Q li Assign parallax d ij The matching state energy at .

4. The method according to claim 1 or 2, characterized in that: The phase consistency constraint is: Among them, (Q li ,Q ri ) are the corresponding matching points on the same contour in the left and right cameras, φ(x li ,y li ) is (x li ,y li ) on the left camera image, w is set to the pixel Q on the left camera li The size of the local window centered on the 5. The method according to claim 1, characterized in that In step S1, epipolar geometry constraints are used to simplify the two-dimensional search process during matching into a one-dimensional search process on the epipolar line; The images of the workpiece to be measured collected synchronously by the left and right cameras are stereo-corrected according to the internal and external parameters of the left and right cameras, so that they are transformed to the same aligned plane, so that all the corrected epipolar lines are aligned to the same horizontal line.

6. An electronic device, characterized in that: include: A computer readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute the method according to any one of claims 1 to 5.

8. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images