| The development of the Wavelet Transform (WT) has enabled effective analysis of non-stationary signals. The most widely used form of the WT is the Decimated Discrete Wavelet Transform (DDWT) implemented in a multiresolution filter bank tree. The non-redundant nature and perfect reconstruction properties of the DDWT are particularly attractive. The decimated form does have some shortcomings, namely sensitivity to input signal shifts and limited directionality for multi-dimensional analysis.; This thesis presents a review of the DDWT structure, capabilities, and shortcomings. Variations of the DDWT are analyzed with focus on the recently introduced Dual Tree Discrete Wavelet Transform (DT DWT). The DT DWT overcomes the shortcomings of the DDWT with limited redundancy. The capabilities of the DDWT and DT DWT in image fusion are analyzed and compared. The DT DWT is shown to be a viable alternative to the DDWT in this application. |
