Title :
The contrast pyramid based image code
Author :
Yu, Tian-Hu ; Chen, Jianhua ; Mitra, Sanjit K.
Author_Institution :
Dept. of Inf. Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
Abstract :
In this paper, a contrast pyramid based image coding method is described in which a simple nonlinear difference operation is introduced to generate residual information between two contiguous level images. The representation of the residual information can be considered as a local contrast measure. As the difference image in established pyramid coding techniques, the contrast image has low variance, and data compression is achieved by quantizing the contrast image. Based on Weber´s law and spatial masking of the human visual system, further data compression can be attained by selecting only perceptually significant information from the contrast image and coding it efficiently
Keywords :
data compression; filtering theory; image coding; prediction theory; Weber´s law; contiguous level images; contrast image; contrast pyramid based image coding; data compression; difference image; human visual system; local contrast measure; nonlinear difference operation; perceptually significant information; predictive coding; residual information; spatial masking; Data compression; Data structures; Filtering; Humans; Image coding; Image generation; Laplace equations; Low pass filters; Nonlinear filters; Visual system;
Conference_Titel :
Image Processing, 1997. Proceedings., International Conference on
Conference_Location :
Santa Barbara, CA
Print_ISBN :
0-8186-8183-7
DOI :
10.1109/ICIP.1997.631998