Binocular vision physical coordinate positioning algorithm based on PSO-Harris operator

摘要

Aiming at the problems of low accuracy, low positioning accuracy and low efficiency of feature points extraction inexisting binocular vision positioning methods, and classical Harris corner detection algorithms depending on theselection of experiential threshold and the setting of corner points extraction, a corner detection algorithm based on PSOHarrisoperator to detect the double threshold is proposed in this paper, which calculates the optimal threshold bydynamic iterative optimization of the threshold. On this basis, the feature point parallax is calculated to match the imagefeatures, and the coordinates of the world coordinate system are calculated according to the coordinates of the targetpoint. The comparison experiments show that the algorithm effectively improves the positioning accuracy and efficiencyof feature point extraction.