【摘要】 根据综合剪切和递归Cycle Spinning技术,提出一种基于剪切不变的递归Contourlet变换图像去噪方法(RSICT)。为改善图像去噪由于缺少平移不变性而产生的伪吉布斯效应,使用剪切替代平移技术来提取图像中原有的几何特征,将递归Cycle Spinning方法运用在剪切技术中给出剪切不变思想,并将其用于Contourlet域图像去噪。对于被加性高斯白噪声污染的图像,实验中将RSICT方法与平移不变小波、平移不变Contourlet等方法进行了比较,结果表明在大多数情况下,RSICT的PSNR结果相比这些方法高出0.1至1.2dB,并保持良好的视觉效果。更多还原

【Abstract】 Based on Shear and Recursive Cycle Spinning,a novel contourlet transform denoising scheme was proposed.To avoid the Gibbs-like phenomena caused by transform variance of the contourlet transform,translation method was replaced by shear technique and employed to capture the geometric structure hidden in images,applied Recursive Cycle Spinning to Shear produced Shear Invariant method,and developed Recursive Shear Invariant Contourlet Transform denoising scheme(RSICT).This scheme achieves enhanced estimation results for images that are corrupted with additive Gaussian noise over a wide range of noise variance.To evaluate the performance of the proposed algorithms,experiment results were compared with those of the algorithms,such as translation invariant wavelets and translation invariant contourlet method.The simulation results indicate that the proposed method outperforms the others 0.1～1.2 dB in PSNR,and keeps better visual result as well.更多还原