基于最大信息熵的小波图像去噪算法研究

去噪算法在图像处理的过程中占有极其重要的地位。为了对含有高斯白噪声和脉冲噪声的图像进行去噪，在Donoho提出的小波阈值去噪算法的基础上，提出一种基于最大信息熵的小波去噪算法，根据最大信息熵的理论确定了改进型阈值和改进型加权阈值函数中的加权因子。仿真结果表明，该算法能够同时抑制高斯白噪声和脉冲噪声，可以更好地保留图像的边缘细节，与Donoho提出的小波阈值去噪算法的去噪效果相比，具有更好的去噪性能。

Research on wavelet image denoising algorithm based on maximum information entropy

Denoising algorithm is very important in the course of image processing. In order to denoise the image with Gaussian white noise and impulse noise, on the basis of wavelet threshold denoising algorithm proposed by Donoho, a wavelet denoising method based on maximum information entropy is proposed. The modified threshold and the weighted factors in the modified weighted threshold function are ascertained by tile theory of maximum information entropy. The simulation results show that the proposed algorithm can suppress Gaussian white noise and pulse noise simultaneously and can retain the edge details of image better. The denoising effect is more effective than that of wavelet threshold denoising algorithm proposed by Donoho.