Color Medical Image Fusion using Non-subsampled Shearlet Transform

Abstract

Color medical image fusion is visually clearer than grayscale fusion, even non-professionals can quickly locate the lesions according to image. Many approaches of image fusion have not made a thorough study of the fusion of color medical images or got ideal visual effects. In view of this situation, a color medical image fusion method based on non-subsampled shearlet transform (NSST) is proposed. After NSST transformation, an improved non-negative matrix factorization (NMF) model is used to fuse the low-frequency sub-bands, while an algorithm combining visual sensitivity coefficient and energy matching degree is adopted in the high-frequency sub-band fusion. And the sensitivity of human vision to the change of grayscale is also taken into account. The results of the fusion of MRI and SPECT/PET images show that the resolution is better than the previous algorithms, and the treatment of the fine parts is more reliable.