Study of Wavelet Thresholding Image De-noising Algorithm Based on Improvement Thresholding Function

In order to enhance the de-noising performance of wavelet thresholding de-noising algorithm and increase the degree of similarity between the original signal and de-noising signal, this paper proposes a new thresholding function with adjustment factors based on wavelet thresholding de-noising theory introduced by D. L. Donoho and I. M. Johnstone. First continuity and higher derivative of thresholding function can be obtained by setting adjustment factors, second the difference between the value of the original signal and the one quantized is unlimitedly decreased, finally, according to the noisy signal features and the distributional difference of noise energy, the effective thresholding function can be obtained with different adjustment factors. Objective assessment parameters of de-noising image quality are calculated from the experiment of Matlab simulation, which suggests that the impact of de-noising algorithm with the new thresholding function is better than the traditional soft and hard thresholding de-noising algorithms, furthermore, which lays a good foundation on image enhancement, feature extraction and edge detection.