基于一种新的阈值函数的小波图像去噪

Wavelet Image Denoising Based on a New Threshold Function

文中在D.L.Donoho和I.M.Johnstone提出的小波阈值去噪基础上,提出了一种改进的阈值函数。该阈值函数采用双阈值模式,通过逐渐增大对小波系数的缩减力度来处理双阈值之间的小波系数,尽可能多地保留有用信息,直至小波系数缩减为零。在这里引入了一个控制变量来调节系数的缩减幅度。与传统的软、硬阈值方法相比,改进的阈值函数最大的优点是函数连续且减小了估计系数的误差。通过仿真实验,从视觉和客观评价标准(峰值信噪比和均方根误差)上验证了新阈值函数去噪的有效性。

An improved threshold function is presented based on the wavelet shrinkage put forward by D. L. Donoho and I. M. Johnstone. This function has two thresholds,it retains the useful information as much as possible by gradually increasing the shrink of wavelet coeffi-cients between the two thresholds,until the coefficients are reduced to zero. In this paper,introduce a control variable to adjust the shrink. Compared with the traditional soft and hard threshold methods,the greatest superiority of this new function is continuous and reducing the error of estimated coefficient. Finally,demonstrate and validate the denoising effect of this new threshold function by simulation experi-ment from vision and objective evaluation standards （ Peak Signal to Noise Ratio and Root-Mean-Square Error） .