An Adaptive Wavelet Transform with Translation and Scale Invariance
Xiong Hui-li, Hust Wuhan · Dianzi xuebao · 2000
In this paper,we propose an adaptive wavelet transform which possess the properties of translation and scale invariance.Firstly,the Original signal is adaptively renormalized using a scale function of an orthonomal Wavelet and the first two moments of the signal.Then, we decompose the renomalized signal according to the conventional discrete wavelet transform. As we prove this adaptive wavelet transform is translation-and scale-invariant, and an efficient algorithm for calculating these wavelet coefficients ,called adaptive Wavelet invariant moments, is proposed. Finally, we give experiment results for 2-dimension digital signals (images) to verify our conclusion.'