Title :
Robust classification of multispectral data using multiple neural networks and fuzzy integral
Author :
Kumar, A. Senthil ; Basu, S.K. ; Majumdar, K.L.
Author_Institution :
Image Process. & Products Div., Space Applications Centre, Ahmedabad, India
fDate :
5/1/1997 12:00:00 AM
Abstract :
The use of fuzzy integral (FI) for combining an ensemble of neural networks is proposed for multispectral data classification. The FI is further employed for integrating original data with its smoothed version. In both cases, the FI combine is found to be superior to that of their individual classification performance
Keywords :
fuzzy neural nets; geophysical signal processing; geophysical techniques; geophysics computing; image classification; remote sensing; data classification; fuzzy integral; geophysical measurement technique; image classification; land surface; multiple neural net; multiple neural network; multispectral method; optical imaging; remote sensing; robust classification; terrain mapping; Backpropagation; Convergence; Feeds; Fuzzy neural networks; Layout; Neural networks; Remote sensing; Robustness; Testing; Training data;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
