Technique for image de-noising based on non-subsampled shearlet transform and improved intuitionistic fuzzy entropy

To overcome the problem of image de-noising, a novel technique based on non-subsampled shearlet transform (NSST) and intuitionistic fuzzy entropy (IFE) is proposed in this paper. Firstly, as a recently developed geometric analysis tool, NSST is utilized to decompose the noised image into multi-scale and multi-directional sub-band images. Secondly, the basic model of intuitionistic fuzzy entropy (IFE) is updated to be an improved IFE, with which each pixel in the noised image is adaptively defined by a pair of membership degree and non-membership degree, and each pixel's weight value can be figured out via improved IFE. Thirdly, the final de-noising image can be obtained by using inverse NSST. Besides, the algorithm for image de-noising is devised and concrete steps are also given. By following the above steps, the noising pixels in the image can be filtered well. Moreover, experimental results indicate that the technique is much more effective compared with other traditional filter ones.