Wavelet image coding with adaptive dead-zone selection: application to JPEG2000

Several quantization options are provided within JPEG2000. Part I of the standard includes only uniform scalar deadzone quantization. The dead-zone is equal to two times the other quantization bin sizes. Part II includes both uniform scalar dead-zone and wavelet trellis coded quantization (WTCQ) which combines the characteristics of a scalar and vector quantization. In this work we develop an adaptive quantization technique for JPEG2000 coding scheme. Using a simple statistical model for the subband or transform coefficients, we present a procedure to select the adaptive dead-zone width. After the zeroing process, the wavelet coefficients are then quantized using WTCQ strategy. Experimental results show that the rate-distortion performance of the proposed compression algorithm compares favorably with the one provided by JPEG2000 standard.