Abstract

Applications of the first generation of wavelets( such as wavelets B-97, Daubechies, B-spline...) in image compression and noisy filter have gained the good results comparing to those of the previous methods which are based on Fourier and DCT transforms [1],[21,[3]. Since 1996 until now, the mathematicians and the signal processing experts develop a new wavelet generation which is the so-called LIFTING transform. Because of the simple algorithm of Lifting, we can design the wavelets which adapt themselves to the signals with variable statistics models. Those signals are popular in real applications, especially such as signals of image or video. In this paper we only study the efficiency of Lifting transforms in image compression, analyse and compare them to the results of the first generation wavelets which were introduced in the previous conferences [1],[2],[3]. Contrary to the first generation wavelets designed especially for image compression, the normal Lifting transforms are used in our experiments.