Application of Undecimated Wavelet Transform in Breast Cancer Detection: Decomposition and Reconstruction

Abstract

In this research paper, undecimated wavelet transform (UDWT) is focused on with respect to its decomposition and reconstruction and its applicability in enhancing mammographic images. It is established that UDWT eliminates decimation process, it is shift invariant and each coefficient is up sampled by a factor, hence facilitating a high degree of freedom. The proposed method, UDWT is discussed in the context of à trus algorithm and the two ways in which it can be applied to enhance mammograms –the 2-level UDWT, 3-Level UDWT and 4-Level UDWT. As such the main aim of this research paper is to demonstrate how undecimated wavelet transform differs considerably by eliminating decimation process, shift insensibility and absence of phase information.

Keywords: undecimated wavelet transform, shift invariance, wavelet transform