TY - GEN
T1 - Classification of hyperspectral images with binary fractional order darwinian pso and random forests
AU - Ghamisi, Pedram
AU - Couceiro, Micael S.
AU - Benediktsson, Jon Atli
PY - 2013
Y1 - 2013
N2 - A new binary optimization method inspired on the Fractional-Order Darwinian Particle Swarm Optimization is proposed and applied to a novel spectral-spatial classification framework.Afterwards, the new optimization algorithm is used in a novel spectral-spatial classification frameworkfor the selection of the most effective group of bands. In the proposed approach, first, the raw data set (only spectral data) along with the morphological profiles of the first effective principal components are integrated into a stacked vector. Then, the output of this step is considered as the input of the new optimization method. The Random Forest classifier is used as a fitness function for the cross-validation samples and the overall classification accuracy for the evaluation of the group of bands.Finally, the selected bands are classified by a classifier and the output provides the final classification map. Experimental results successfully confirm that the new approach works better than when considering all the raw bands, the whole morphological profile and the combination of the raw bands and morphological profile.
AB - A new binary optimization method inspired on the Fractional-Order Darwinian Particle Swarm Optimization is proposed and applied to a novel spectral-spatial classification framework.Afterwards, the new optimization algorithm is used in a novel spectral-spatial classification frameworkfor the selection of the most effective group of bands. In the proposed approach, first, the raw data set (only spectral data) along with the morphological profiles of the first effective principal components are integrated into a stacked vector. Then, the output of this step is considered as the input of the new optimization method. The Random Forest classifier is used as a fitness function for the cross-validation samples and the overall classification accuracy for the evaluation of the group of bands.Finally, the selected bands are classified by a classifier and the output provides the final classification map. Experimental results successfully confirm that the new approach works better than when considering all the raw bands, the whole morphological profile and the combination of the raw bands and morphological profile.
KW - hyperspectral image classification
KW - image processing
KW - random forest classifier
KW - swarm binary optimization
UR - http://www.scopus.com/inward/record.url?scp=84889007055&partnerID=8YFLogxK
U2 - 10.1117/12.2027641
DO - 10.1117/12.2027641
M3 - Conference contribution
AN - SCOPUS:84889007055
SN - 9780819497611
T3 - Proceedings of SPIE - The International Society for Optical Engineering
BT - Image and Signal Processing for Remote Sensing XIX
T2 - Image and Signal Processing for Remote Sensing XIX
Y2 - 23 September 2013 through 25 September 2013
ER -