$\:$

$\:$

Image Retrieval via Isotropic and Anisotropic Mappings

Qasim Iqbal - J. K. Aggarwal

Computer and Vision Research Center
Department of Electrical and Computer Engineering
The University of Texas at Austin
Austin, Texas 78712, USA.

{qasim,aggarwaljk}@mail.utexas.edu

Abstract:

This paper presents an approach for content-based image retrieval via isotropic and anisotropic mappings. Isotropic mappings are defined to be mappings invariant to the action of the planar Euclidean group - invariant to the translation, rotation and reflection of image data, and hence, invariant to orientation and position. Anisotropic mappings, on the other hand, are defined to be those mappings that are correspondingly variant. Structure extraction (via a perceptual grouping process) and color histogram are shown to be representations of isotropic mappings. Texture analysis using a channel energy model comprised of even-symmetric Gabor filters is considered to be a representation of anisotropic mapping. Results of retrieval of outdoor images by query and by classification using a nearest neighbor classifier are presented.