TEXTURE-BASED PATTERN RETRIEVAL FROM IMAGE DATABASES

Recent technological advances have made it possible to process and sto re large amounts of image data. Perhaps the most impressive example is the accumulation of image data in scientific applications such as med ical or satellite imagery. However, in order to realize their full pot ential, tools for efficient extraction of information and for intellig ent searches in image databases need to be developed. This paper descr ibes a new approach to image data retrieval which allows queries to be composed of local intensity patterns. The intensity pattern is conver ted into a feature representation of reduced dimensionality which can be used for searching similar-looking patterns in the database. This r epresentation is obtained by filtering the pattern with a bank of scal e and orientation selective filters modeled using Gabor functions. Exp erimental results are presented which illustrate that the proposed rep resentation preserves the perceptual similarities, and provides a powe rful tool for content-based image retrieval.