科研之友
微信
新浪微博
Facebook
分享链接
摘要
Denosing method is very important for CT images because the existing noise bring troubles for diagnosis. Some existing CT image de-noising algorithm and its advantages and disadvantages are compared in the paper. And NL-means algorithm is introduced. An improved method based on wavelet is proposed in this paper to avoid inaccurate computation of region similarity caused by noise. The proposed algorithm has the same capability of noise elimination, but better in detail preserving. The experimental results prove the validity.
- 出版日期2013
- 单位哈尔滨工程大学
