Title :
Embedded zero-tree coding of images employing soft-thresholding
Author :
Muller, Frank ; Eigner, Klaus
Author_Institution :
Institut für Elektrische Nachrichtentechnik, Aachen University of Technology (RWTH), 52056 Aachen, Germany
Abstract :
Improvements of embedded zero-tree wavelet (EZW) coding by employment of soft-thresholding in the wavelet domain are reported. By proper adjustment of the thresholds, the attainable PSNR can be slightly improved (around 0.3 dB). Moreover, soft-thresholding can improve the visual appearance of the coded images, especially in the very low data rate case. The quality in the early stages of progressive image transmission can be improved by reduction of annoying artifacts resulting from coarse quantization of the wavelet coefficients. Adaptation of the thresholds to the "current" width of the quantization intervals preserves the embedded bit stream property of the coder.
Keywords :
Encoding; Image coding; Noise; Quantization (signal); Wavelet domain; Wavelet transforms;
Conference_Titel :
European Signal Processing Conference, 1996. EUSIPCO 1996. 8th
Conference_Location :
Trieste, Italy
Print_ISBN :
978-888-6179-83-6