Fast corner detection of high contrast image based on mathematical morphology and Curvature Scale Space

In this paper, we propose a fast method based on mathematical morphology and Curvature Scale Space (CSS) to detect corners of the high contrast image. Firstly, we use the grayscale iteration threshold algorithm to segment the image; secondly, mathematical morphology operation is used to extract edge of the corresponding binary image; at last, we detect the corners through the CSS corner detector. The proposed method was evaluated over the escape wheel image. We compared the proposed method to other corner detector method, and the results showed that the proposed method offers a fast and effective solution.