Title :
Sparse endmember extraction and demixing
Author :
Ehler, M. ; Hirn, M.
Author_Institution :
Helmholtz Zentrum Munchen, German Res. Center for Environ. Health, Neuherberg, Germany
Abstract :
A novel algorithm for endmember extraction is presented. The approach follows the linear mixture model for hyperspectral data. Endmembers are identified based on sparsity consideration. Theoretical and experimental results suggest the potential of the method.
Keywords :
feature extraction; geophysical image processing; image representation; image resolution; minerals; set theory; terrain mapping; abundance maps; chemical composition; endmember set; hyperspectral data; linear mixture model; mixture-of-materials representation; multispectral image set; multispectral measurements; sparse endmember demixing; sparse endmember extraction; sparsity consideration; Dictionaries; Hyperspectral imaging; Materials; Noise; Optimization; Signal processing algorithms; Vectors;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2012 IEEE International
Conference_Location :
Munich
Print_ISBN :
978-1-4673-1160-1
Electronic_ISBN :
2153-6996
DOI :
10.1109/IGARSS.2012.6351278
