Title :
Higher-order statistical models of visual images
Author :
Simoncelli, Eero P.
Author_Institution :
Center for Neural Sci., New York Univ., NY, USA
Abstract :
This paper examines the empirical densities of natural photographic images, and shows that although they are highly non-Gaussian, they are quite regular and may be described using fairly simple parameterized density models. Two such models are described, and their ability to account for image content is demonstrated
Keywords :
higher order statistics; image processing; photography; wavelet transforms; computer graphics; computer vision; empirical densities; higher-order statistical models; image processing; natural photographic images; parameterized density models; visual images; wavelet domain analysis; Application software; Bandwidth; Entropy; Frequency; Histograms; Identity-based encryption; Independent component analysis; Mathematical model; Principal component analysis; Statistics;
Conference_Titel :
Higher-Order Statistics, 1999. Proceedings of the IEEE Signal Processing Workshop on
Conference_Location :
Caesarea
Print_ISBN :
0-7695-0140-0
DOI :
10.1109/HOST.1999.778691
