Combination of neural and statistical algorithms for supervised classification of remote-sensing images

摘要: Abstract Various experimental comparisons of algorithms for supervised classification remote-sensing images have been reported in the literature. Among others, a comparison neural and statistical classifiers has previously made by authors (Serpico, S.B., Bruzzone, L., Roli, F., 1996. Pattern Recognition Letters 17, 1331–1341). Results experiments clearly shown that superiority one algorithm over another cannot be claimed. In addition, they pointed out often require expensive design phases to attain high accuracy. this paper, combination is proposed as method obtain accuracy values after much shorter improve accuracy–rejection tradeoff those allowed single algorithms.