Corner Detection Based on Accumulative Chord Length of Refined Digital Curves

Corner detection based on accumulative chord length is invariant to rotation transformations,while the implicitly refining strategy of digital curves can improve computational precision and enhance the reliability of corner detection significantly.The algorithm is presented,and its feasibility is verified mathematically,with which a useful technique is developed to distinguish sharp corners from rounded corners.Experiments show that the new approach is efficient for solving practical problems.