Marking-Points Matching and Tracking Methods in Micro-table Posture Measurement System

Based on binocular visual theory, a new method is proposed for solving the marking-points matching and tracking issues in the micro-table 3D posture measurement system. The four marking-points of the micro-table are detected by Harris corner detection algorithm, an improved epipolar constraint match method is used to make the corresponding marking-points be restricted in a region, circle region search method is used to match the same marking-point, and the template-based recognition algorithm is used to track the marking-points in real-time image frames. The experiment results verify the effectiveness of this method.