A multi-layered image coding using multiple transforms

This paper proposes a new multi-layered image coding algorithm based on wavelet and matching pursuit transforms. The original image is divided into smooth and edge layer and texture layer, Different compression algorithms of coding are adopted to each different layer in order to achieve optimization for the entire image. Case studies have demonstrated that this coding method can capture texture and edge characteristics efficiently. Besides, the quality of reconstructed image is better than traditional single wavelet image coding algorithm at low bit rates.