Wavelet transform for corner detection

A novel wavelet based corner detecting algorithm is proposed. Some intrinsic indicators implied in corners are extracted by utilizing the wavelet transform. Since these indicators are independent of the corner angle and corner curvature, they can be used to detect corners. In addition, several properties of corners in the multiscale wavelet transform are introduced. By applying these indicators and properties, corners can be detected correctly and efficiently. The experimental results show that the proposed algorithm achieves better accuracy than the conventional single-scale corner detectors. On the other hand, the algorithm is more computationally efficient and easier to implement than other multiscale corner detectors.<<ETX>>