摘要
综合考虑Contourlet变换后系数尺度内和尺度间的依赖性,提出了一种基于Contourlet变换的自适应去噪方法。与几种传统的小波去噪算法比较,它能够更有效地保留图像的细节和纹理,具有更好的视觉效果和较优的SNR值。
Contourlet transform is a ＂true＂ two-dimensional image sparse representation. It is multiresolutional, localized , multidirectional and anisotropic, so it is more effectively capture high dimensional singularity. Because of all these characteristics, and considering of the intrasubband and interscale dependencies, an adaptive denoising algorithm was proposed based on the Contourlet transform. The experimental results indicate that it is better than several traditional wavelet shrinking denoising algorithm in smoothing noise and preserving texture and details, and improving the SNR.
出处
《红外技术》
CSCD
北大核心
2006年第9期552-556,共5页
Infrared Technology
基金
武器装备预研基金项目(编号:51483020105ZS9309)
