An Improved Nonsubsampled Contourlet Transform for Image Denoising

摘要：

A new method called nonsubsampled dual-tree complex contourlet transform (NSCCT) for image denoising is presented. It is composed of two parts: nonsubsampled contourlet transform (NSCT) used for direction analysis and dual-tree complex wavelet used for multi-resolution analysis. Experiment result shows that the proposed transformation has the characteristics of low redundancy and shiftinvariant, as well as more abundant directional components. So it can restrain Gibbs-like artificials around edges in the course of denoising, and preserve more image details and textures efficiently. Its performance in both PSNR value and visual quality exceeds existing techniques.

关键词： image denoising contourlet transform NSCCT nonsubsampled contourlet transform)

作者: Xiaoli Xu Zhimao Lu Gensen Zhang

作者单位: Information and Communication Engineering College Harbin Engineering University Department of Electr Information and Communication Engineering College Harbin Engineering University Harbin, China

会议类型: 国际会议

会议名称: 2010 4th International Conference on Intelligent Information Techonlogy Application(第四届智能信息技术应用国际学术研讨会 IITA 2010)

会议地点: 秦皇岛

会议语种:英文

页码: 413-416

在线出版日期: 2010-11-05（万方平台首次上网日期，不代表论文的发表时间）