论文部分内容阅读
提出一种基于第二代轮廓波变换的消噪方法,该方法在阈值选取上采用基于自适应冲激响应的多尺度多方向阈值,在硬阈值处理之后采用自适应维纳滤波。该方法充分利用了第二代轮廓波变换在频域和时域的局部化、多尺度、多方向和低冗余特征;能够利用更多较小的轮廓波变换系数以减轻轮廓波硬阈值消噪带来的纹理效应,并能更好适应人类的视觉特征。实验结果表明提出的消噪方法在峰值信噪比和视觉效果上都优于第一代轮廓波和小波的消噪效果。
A new de-noising method based on the second generation of contourlet transform is proposed. This method uses multi-scale and multi-direction threshold based on adaptive impulse response for threshold selection and adaptive Wiener filtering after hard threshold processing. The method takes full advantage of the localization, multi-scale, multi-directional and low-redundancy features of the second generation of contourlet transform in the frequency and time domains. It can utilize more smaller contourlet transform coefficients to reduce the contour wave threshold Noise brings the texture effect, and can better adapt to human visual characteristics. The experimental results show that the proposed denoising method is superior to the first generation of contourlet and wavelet denoising effect in peak signal-to-noise ratio and visual effect.