Denoising for liver CT image based on texture analysis

The liver CT play an important role in the clinical diagnosis and treatment of liver disease. In this paper, we present a improved approach which introduce more information of texture image in the original NL-means algorithm. The improved algorithm keeps the excellent denioising function and preserving with edge, detail, especially the texture. The artificial artifact was also restrained effectively. The availability of the improved algorithm was validated for liver CT images by the experiment result.