基于正交小波变换的图象去噪算法的改进

本文在BayesShrink 阈值的基础上，提出了一种计算更有效的自适应的阈值估计，并
针对传统的软、硬阈值函数在去噪中的不足，提出一种新的去噪阈值函数，最后将所提出的图象去噪方法与传统方法进行了实验比较，实验结果表明该方法在图象去噪中取得较好的效果，有效地提高了信号去噪所得信噪比增益。
关键词：正交小波变换、图象去噪、阈值
中图分类号:TN911.7 文献标识码:A
An Improved Algorithms of Image Denosing Based on Orthogonal Wavelet Transform
LEI Hui ( School of Electrical and Infomational Engineering, Changsha University of Science & Technology,Changsha,410076) Abstract：This paper proposes an efficient threshold estimation method for image denoising based on BayesShrink. So the traditional methods for noise reduction using wavelet soft threshold function and hard threshold function have some defects, an improved threshold function is proposed, At the end, experiments have been done
between this new method and the traditional methods. The experiment results indicate
that this new method achieves better effect on reducing noise and gets higher SNR
gain for the reconstructed signal.
Keywords: orthogonal wavelet transform, Image denoising, threshold