A Novel Corner Point Detector for Calibration Target Images Based on Grayscale Symmetry

The research of feature corner recognition for the calibration target is a crucial area for a vision-based measurement system. For the disability of detecting the corner´s location in a vehicle test system accurately, the theoretical errors based on Harris algorithm are analyzed. Experimental results show that the method can not be used in the vision-measured field as the corner cluster phenomenon. The novel corner detection algorithm based on grayscale symmetry is presented. The Gassian weighted symmetry detector is designed to illustrate the symmetrical feature around the optimal corner. Further Experimental results show that the detector proposed can be used to test the corner locations of the calibration target precisely.