A Calibration Method for A Linear Structured Light System with Three Collinear Points

Abstract

A calibration method to calibrate both the camera intrinsic parameters and the structured light projector based on three collinear points with any known distance is presented. First, fix one point of the target and move it in the camera field of view with unknown viewpoint at least six times. Next, use camera to obtain the image of the points so as to identify the camera intrinsic parameters. Then, move the target in the structure light plane at least twice. Using the calibrated camera obtains the images of the three points and adopts the principle of triangulation to work out the coordinates of the calibration points in the camera coordinate system. Finally, employ the random sample consensus algorithm to eliminate the outlier and adopt the genetic algorithm for system identification to find the coefficients of the structured light plane. The average absolute error and the average relative error of the segment measurement with this structured light system are 0.13 mm and 0.04%, respectively. Those show that the measuring accuracy of this system achieves the requirements of the basic error limit. Comparison results show that the measurement accuracy of our method is superior to the other method for the same targets.

Analysis of the Measuring Model of Auto-Evaluation System of Railway Coupler

“…coordinate of practical image; u σ and v σ that are respectively distortion values of Image projection point (u, v) can be compensated via distortion model [4][5][6][7]. …”