Abstract
All-purpose mathematical model of structured-light vision sensor is established. A novel calibration method for estimating the structure parameters of structured-light vision sensor with planar target is proposed. The target is freely placed at various positions of the measurement space many times, and the vision sensor projects structured-light on the surface of the target. The camera of structured-light vision sensor takes images of the target. Coordinate transformations from target coordinate frame to camera coordinate frame are solved through the feature points on the target and corresponding image points. The coordinates of the calibration points on structured-ligh at target coordinate frames are obtained according to the homography transformation between target planes and image plane, and are unified at the camera coordinate frame, based on which the equation of the structured-light is fitted. The efficiency of acquiring calibration points by using the proposed method is high. It is competent for calibrating all types, including point, line, and curved surface, of structured-light vision sensor on the spot. The experimental results show that the calibration precision of the method is high.