Improved hidden Markov models in the wavelet-domain

Wavelet-domain hidden Markov models (HMMs), in particular the hidden Markov tree (HMT) model, have recently been introduced and applied to signal and image processing, e.g., signal denoising. In this paper, we develop a simple initialization scheme for the efficient HMT model training and then propose a new four-state HMT model called HMT-2. We find that the new initialization scheme fits the HMT-2 model well. Experimental results show that the performance of signal denoising using the HMT-2 model is often improved over the two-state HMT model developed by Crouse et al.