Perform Image Fusion Based Multi Resolution Curyelet and Framework System

ABSTRACT: The image fusion is to integrate complementary information from several images to create a highly informative image which is more suitable for human visual perception or computer-processing tasks. Recent studies show that stationary wavelet transform (SWT) and non sub sampled contour let transform (NSCT) both turn out to be effective an deficient for image fusion. In order to take some complementary characteristics between the two muli resolution transformations simultaneously, we propose a hybrid multi resolution method by combining the SWT with the NSCT to perform image fusion. Two methods, serial NSCT aiding SWT (SNAS) and serial Standing NSCT (SSAN), are studied and compared with some state-of-the-art methods. Experimental results demonstrate that the SSAN method performs better than SNAS and the individual multi resolution-based methods, such as NSCT, SWT, complex wavelet (CWT), curve let (CVT) and wavelet-based contour let (WBCT)