Image coding using a complex dual-tree wavelet transform

An embedded wavelet-based coder is deployed to exploit the directional selectivity of a 2D complex dual-tree discrete wavelet transform. Although the dual-tree transform is redundant, a noise-shaping process increases the spar-sity of the transform coefficients, resulting in a high degree of spatially coherent regions of insignificant coefficients. The transform coefficients are coded with binary set-partitioning using k-d trees, and experimental results reveal rate-distortion results superior to the state-of-the-art JPEG2000 standard at low bitrates, particularly for images with strong directional features.