A TEXTURE THESAURUS FOR BROWSING LARGE AERIAL PHOTOGRAPHS

A texture-based image retrieval system for browsing large-scale aerial photographs is presented. The salient components of this system inclu de texture feature extraction, image segmentation and grouping, learni ng similarity measure, and a texture thesaurus model for fast search a nd indexing. The texture features are computed by filtering the image with a bank of Gabor filters. This is followed by a texture gradient c omputation to segment each large airphoto into homogeneous regions. A hybrid neural network algorithm is used to learn the visual similarity by clustering patterns in the feature space. With learning similarity , the retrieval performance improves significantly. Finally, a texture image thesaurus is created by combining the learning similarity algor ithm with a hierarchical vector quantization scheme. This thesaurus fa cilitates the indexing process while maintaining a good retrieval perf ormance. Experimental results demonstrate the robustness of the overal l system in searching over a large collection of airphotos and in sele cting a diverse collection of geographic features such as housing deve lopments, parking lots, highways, and airports.