A partially unsupervised classification system for a regular updating of land-cover maps is proposed. The systems is composed of different partially unsupervised cascade classifiers integrated in the framework of a multiple classifier approach. The use of cascade classifiers allows one to exploit temporal correlation between multitemporal remote-sensing images to increase the accuracy and robustness of the partially unsupervised classification approach. Experimental results confirm the effectiveness of the proposed classification system.
A non-parametric classification system for a partially unsupervised updating of land-cover maps
Bruzzone, Lorenzo;
2001-01-01
Abstract
A partially unsupervised classification system for a regular updating of land-cover maps is proposed. The systems is composed of different partially unsupervised cascade classifiers integrated in the framework of a multiple classifier approach. The use of cascade classifiers allows one to exploit temporal correlation between multitemporal remote-sensing images to increase the accuracy and robustness of the partially unsupervised classification approach. Experimental results confirm the effectiveness of the proposed classification system.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione
