Gray-level-based corner detection by using wavelet transform

Corners are very attractive features for many applications in computer vision. In this paper, a novel gray-level corner detection algorithm based on the wavelet transform is presented. First, we derived several properties of the corner and the edge embedded in the wavelet domain. Next, we apply these similar properties between the corner and the edge to extract the edge image. Then we employed these different properties between the corner and the edge to locate corners. Experiments have shown that our algorithm can detect corners accurately and effectively.<<ETX>>