Nonlinear Algorithm of Image Enhancement Based on Wavelet Transform

Traditional image enhancement methods will enhance the region of interest at the same time enhance the noise. A nonlinear algorithm of image enhancement based on wavelet transform is proposed. Firstly, the multi-scale wavelet transform is adopted to decompose the input image, which can extract the characteristic of multi-scale of the image. Secondly, wavelet threshold is used for image de-noising and low frequency coefficients are enhanced by the linear piecewise function, then a nonlinear enhancement operator is designed and applied to enhance the high frequency wavelet coefficients of different directions of each scale. Finally, the inverse wavelet transform is applied to synthesis image which can obtain the enhanced image. A group of experimental results demonstrate that the presented algorithm is effective for image de-noising and enhancement.