June 15, 2014
June 15, 2014
June 18, 2014
Computing & Information Technology
24.1065.1 - 24.1065.18
Saliency-Based CBIR System for Exploring Lunar Surface ImageryRecent NASA missions like the Lunar Reconnaissance Orbiter (LRO) have produced vastarchives of surface imagery which must be analyzed to locate landmarks with distinctive visualfeatures, like craters, which provide important information about geologic history and potentialmineral resources. Content Based Image Retrieval allows large archives of images to beefficiently queried based on visual content by indexing multidimensional feature vectorsextracted from the images. Unlike images of particular objects or scenes traditionally retrievedusing CBIR, surface images are not focused on any particular landmark so they must be pre-processed to identify Regions of Interest (ROI) to be indexed for retrieval. Previous workidentified ROIs using manual annotation and expensive detection algorithms for specific types oflandmarks, such as Crater Detection Algorithms (CDA). In contrast, this work utilizes a general-purpose saliency-based landmark detection algorithm for identifying ROIs which are thenindexed for retrieval using feature vectors extracted from the ROI images. We evaluate theretrieval performance of several feature vectors and assess the saliency-based landmark detectionperformance in comparison to a comprehensive crater database created using manual annotationand a CDA. Experimental results demonstrate the advantages of the general-purpose saliency-based CBIR system for exploring lunar surface imagery.
Hua, K. A., & Shaykhian, G. A., & Beil, R. J., & AKPINAR, K., & Martin, K. A. (2014, June), Saliency-Based CBIR System for Exploring Lunar Surface Imagery Paper presented at 2014 ASEE Annual Conference & Exposition, Indianapolis, Indiana. https://peer.asee.org/22998
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2014 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015