Nonlinear Operators for Improving Texture Segmentation Based on Features Extracted by Spatial Filtering
M. Unser, M. Eden
IEEE Transactions on Systems, Man, and Cybernetics, vol. 20, no. 4, pp. 804-815, July-August 1990.
An unsupervised texture segmentation system using texture features obtained from a combination of spatial filters and nonlinear operators is described. Local texture features are evaluated in parallel by a succession of four basic operations: (1) a convolution for local structure detection (local linear transform); (2) a first nonlinearity of the form f(x) = |x|α; (3) an iterative smoothing operator; and (4) a second nonlinearity g(x). The Karhunen-Loève transform is used to reduce the dimensionality of the resulting feature vector, and segmentation is achieved by thresholding or clustering in feature space. The combination of nonlinearities f(x) = |x|α (in particular, α = 2) and g(x) = log x maximizes texture discrimination, and results in a description with variances approximately constant for all feature components and texture regions. This latter property improves the performance of both feature reduction and clustering algorithms significantly.
@ARTICLE(http://bigwww.epfl.ch/publications/unser9001.html, AUTHOR="Unser, M. and Eden, M.", TITLE="Nonlinear Operators for Improving Texture Segmentation Based on Features Extracted by Spatial Filtering", JOURNAL="{IEEE} Transactions on Systems, Man, and Cybernetics", YEAR="1990", volume="20", number="4", pages="804--815", month="July-August", note="")