DocumentCode :
3436712
Title :
Region-Based Querying of Solar Data Using Descriptor Signatures
Author :
Banda, Juan M. ; Chang Liu ; Angryk, Rafal A.
Author_Institution :
Dept. of Comput. Sci., Montana State Univ., Bozeman, MT, USA
fYear :
2013
fDate :
7-10 Dec. 2013
Firstpage :
1
Lastpage :
7
Abstract :
In this work we present our first results on the ambitious task of providing region-based querying capabilities to our existing Solar Dynamics Observatory (SDO) content-based image-retrieval (CBIR) system. By taking advantage of precomputed image descriptors, we calculate region-based histogram signatures for our training set of previously identified solar events. With these signatures we then explore the possibility of retrieving new similar solar events solely based on these scale and rotation invariant signatures. In this work we present our proposed methodology, initial retrieval results, verification and validation mechanisms, and propose future work with possible applications for other fields in astronomy.
Keywords :
Sun; astronomical image processing; content-based retrieval; image classification; image retrieval; learning (artificial intelligence); SDO content-based image-retrieval system; astronomy; descriptor signatures; precomputed image descriptors; region-based histogram signatures; region-based querying capabilities; rotation invariant signatures; scale invariant signatures; solar data; solar dynamics observatory; solar events; validation mechanisms; verification mechanisms; Accuracy; Data mining; Euclidean distance; Histograms; Image retrieval; Standards; Wavelength measurement; CBIR; image classification; image retrieval; region-based querying;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Data Mining Workshops (ICDMW), 2013 IEEE 13th International Conference on
Conference_Location :
Dallas, TX
Print_ISBN :
978-1-4799-3143-9
Type :
conf
DOI :
10.1109/ICDMW.2013.127
Filename :
6753896
Link To Document :
بازگشت