Calibration and 3D Measurement Methods for Stereo Structured Light System

To address the problem that the calibration of the stereo structured light system is difficult to find the corner points of the projection plane and the problem that there are invalid projection areas in the captured pictures, this paper proposes the method of nonlinear fitting and the statistical threshold to solve the above two problems. The absolute phase of the image can be obtained by overlaying the relative phase obtained from the four-step phase-shifting method and the corrected gray code decoding values, and then obtaining the relationship between the projector image coordinates and the camera image coordinates based on the absolute phase, and this method can ensure the accuracy and continuity of image decoding. According to the relationship between the corresponding coordinates in the two planes, a second-order polynomial is used to fit the nonlinear relationship between the camera pixel coordinates and the projector pixel coordinates, and then the coordinates of the checkerboard corner points in the camera image can be mapped into the projection plane to complete the projector calibration. In the 3D measurement process, we use the statistical threshold to determine the invalid projection areas to better complete the 3D reconstruction. Experimental results verify that the average accuracy of the system calibration was at 0.07 mm, which meets the expected requirements essentially. The invalid projection area was judged more accurately, and the algorithm was more operable.