Title :
Cluster-based probability model and its application to image and texture processing
Author :
Popat, Kris ; Picard, Rosalind W.
Author_Institution :
Media Lab., MIT, Cambridge, MA, USA
fDate :
2/1/1997 12:00:00 AM
Abstract :
We develop, analyze, and apply a specific form of mixture modeling for density estimation within the context of image and texture processing. The technique captures much of the higher order, nonlinear statistical relationships present among vector elements by combining aspects of kernel estimation and cluster analysis. Experimental results are presented in the following applications: image restoration, image and texture compression, and texture classification
Keywords :
data compression; higher order statistics; image classification; image coding; image restoration; image texture; parameter estimation; probability; cluster analysis; cluster-based probability model; density estimation; higher order nonlinear statistical relationships; image compression; image processing; image restoration; kernel estimation; mixture modeling; texture classification; texture compression; texture processing; vector elements; Context modeling; Extraterrestrial phenomena; Image analysis; Image coding; Image restoration; Image texture analysis; Kernel; Probability distribution; Signal processing; Tail;
Journal_Title :
Image Processing, IEEE Transactions on
