Title :
Finding analytical solutions to abundance fully-constrained linear spectral mixture analysis
Author :
Hsiao-Chi Li ; Meiping Song ; Chein-I Chang
Author_Institution :
Dept. of Comput. Sci. & Electr. Eng., Univ. of Maryland, Baltimore, MD, USA
Abstract :
This paper revisits a well-known fully constrained least squares (FCLS) method developed by Heinz and Chang and develops an approach to finding analytical solutions to FCLS, called analytical FCLS (AFCLS) which can be solved in closed forms instead of FCLS being solved by numerically algorithms. As a result, the AFCLS-unmixed results using analytical solutions are more accurate than FCLS-unmixed results resulting from numerical solutions.
Keywords :
geophysical techniques; AFCLS-unmixed results; FCLS method; abundance fully-constrained linear spectral mixture analysis; analytical FCLS; analytical solutions; fully constrained least squares method; numerical solutions; Data models; Educational institutions; Hyperspectral imaging; Image restoration; Vectors; Analytical FCLS (AFCLS); Fully constrained least squares (FCLS); Modified fully constrained least squares (MFCLS); Non-negativity-constrained least squares (NCLS);
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International
Conference_Location :
Quebec City, QC
DOI :
10.1109/IGARSS.2014.6947282
