A Novel Global Calibration Method for Multi-vision Sensors

A novel high-precision global calibration method based on structured light and 2D target is proposed for the multiple vision sensors (MVS) measuring system. By capturing a series of feature points on the light plane at the same time, the relative position of local coordinate system and global coordinate system (GCS) can be acquired in the constraint condition of the uniqueness of a world point. The method is simple and low-cost, which is feasible for on-line measurement. The experiment is designed to illustrate the high-precision of the proposed method. The result shows that the average projection error is less than 0.5 pixel, which can meet the accuracy requirement of multiple camera.