Embedded L-infinite Constrained Compression of Remote Sensing Data

A new wavelet-based L∞ -constrained embedded image coding technique is proposed in this paper. For any desired distortion bound, the embedded bit stream can be truncated at a corresponding bit-rate, for which the required upper bound on the individual elements of the reconstruction error signal is guaranteed. A lifting-based wavelet transform is employed and exact relations are established between spatial and wavelet domain distortions. The wavelet coefficients are quantized by families of embedded quantizers consisting in symmetric uniform mid-tread quantizers for the highest rate and symmetric piecewise uniform quantizers for lower rates. Finally, the quantized coefficients are losslessly encoded using a quadtree-based coding algorithm.