Warped discrete fourier transform and it's application on image compression

Abstract

Digital Transforms in Digital Signal Processing play an important role in analysis and design of discrete-time signals and systems. The popular transforms in today's use are Discrete Fourier Transform (DFT), Discrete Cosine Transform (DCT), and Karhunen Loeve Transform (KLT). We extended the DFT concept to Warped DFT (WDFT), which is nothing but the non-uniform sampling of z-transform on the unit circle. In this thesis we discussed forward WDFT, inverse WDFT and its computing methods for less reconstruction error, and developed a lossy image compression algorithm using adaptive WDFT transforms.