Content based retrieval for lunar exploration image databases

Abstract

Being a novel research aspect following the recent new round of lunar explorations, content-based lunar image retrieval provides a convenient and efficient way for accessing relevant lunar remote sensing images by their visual contents. In this paper, we introduce a novel method for mining relevant images in lunar exploration databases. A novel feature descriptor derived from relationships of salient craters in lunar images and a compound feature model organizing different features are proposed. Based on the features, similarity measurement rules and a retrieval algorithm are proposed and described in detail. Both theoretical analysis and experimental results of our method are provided, verifying that our features and model are effective and the method can get a good relevant retrieval results in lunar image databases. © Springer-Verlag 2013.