Optimum Learning Rate in Back-Propagation Neural Network for Classification of Satellite Images (IRS-ID)

Remote sensing data are essentially used for land cover and vegetation classification, however,classes of interest are often imperfectly separable in the feature space provided by the spectraldata, application of neural networks (nn) to the classification of satellite images is increasinglyemerging, without any assumption about the probabilistic model to be made, the networks arecapable of forming highly non-linear decision boundaries in the feature space, training has animportant role in the nn, there are several algorithms for training and the variable learningrate (vlr) is one of the fastest, in this paper, a network that focuses on the determination ofan optimum learning rate is proposed for the classification of satellite images, different networkswith the same conditions are used for this and the results showed that a network with one hiddenlayer with 20 neurons is suitable for the classification of irs-1d satellite images, an optimumlearning rate between the ranges of 0,001-0,006 was determined for training the vlr algorithm,this range can be used for training algorithms in which the learning rate is constant