Canny edge detection algorithm based on wavelet transform and RAMF

The traditional canny edge detection uses two-dimensional Gaussian function as a noise filter, because Gaussian filter works well for Gaussian noise but is not suitable for impulse noise. This paper based on the combination of wavelet transformation and adaptive median filter RAMF proposes a new method in place of the traditional Gaussian filter. The method makes use of the wavelet transform in time - frequency analysis features and the smooth at the same time RAMF well protected edge of the nature. Experiment results show that the proposed method performs better than the traditional method for impulse noise.