Medical image enhancement using sliding weighted empirical mode decomposition

Since the equipment of medical imaging is influenced by the external and internal facts, the images have low contrast. The base function of wavelet and super-wavelet needs to be predefined, so that the local data characteristics are not adaptively analyzed. Meanwhile, the more noise will be induced in enhancement process which produces a distortion in details. Bi-dimensional empirical mode decomposition can analyze the local data by data driver. Because of the existence of the gray plaques or spot in the components, the traditional BEMD is not suitable for analyzing local data, and the drawbacks affect its application in image processing. In view of this, we propose the sliding weighted empirical mode decomposition (SWEMD) which is more suitable for analysis of the detail feature. To overcome the shortcomings of BEMD, the sliding weighted functions which are adaptive changed with local data are introduced. When SWEMD is introduced to the medical image enhancement and combines with the presented nonlinear enhancement rule, the images are enhanced better. The details of the enhanced image are not only more clear but also undistorted, and the contrast is suitable. The experiments have shown that the proposed algorithm is efficient in image enhancement and better than other current algorithms.