基于改进小波阈值的去噪方法

Denoising Method Based on Improved Wavelet Threshold

针对以往小波阈值图像去噪法出现的去噪不彻底、噪声残留、和噪声误判等问题,对阈值函数和阈值进行了改进,保留了传统的软阈值和硬阈值的优点,改进它们各自的缺点,提出一种新的阈值函数和阈值选取方式,使它在处理小波系数时更加灵活,以达到更好地去除噪声的目的。通过MATLAB仿真实验和对算法的精度分析表明,用改进后的阈值去噪法可以很好地去除图像噪声,使图像的对比度和峰值信噪比均得到很大的提高。

With in respect to incomplete denoising,noise residue and noise misjudgment of previous wavelet threshold image denoising method,this paper improves threshold function and threshold,retains advantages of traditional soft and hard threshold,improves their respective shortcomings,and proposes a new threshold function and threshold selection method,to make it more flexible in processing wavelet coefficients and achieve better denoising effects.Through MATLAB simulation experiment and analysis of the precision of algorithm,the improved threshold denoising method can achieve good image denoising effects and greatly improve image contrast and PSNR.