Artificial neural network with different learning parameters for crop classification using multispectral datasets

Present study evaluated the performance of artificial neural network (ANN) algorithm using different learning parameters for various crop classification in Varanasi, India. Satellite images such as Linear Imaging Self Scanning (LISS) IV and Landsat 8 Operational Land Imager (OLI) were used for crop classification and comparative analysis study. The following crop such as barley, wheat, lentil, mustard, pigeon pea, linseed, corn, pea, sugarcane and other crops and non-crop such as water, sand, built up, fallow land, sparse vegetation and dense vegetation were identified in the area and classified. Results indicated a better classification accuracy of ANN algorithm for crop classification study when used with LISS-IV data in the comparison to Landsat 8-OLI multispectral satellite data. The larger values of the learning rates resulted high fluctuations and less classification accuracy using LISS-IV data, while less but nearly uniform results were found using the Landsat 8-OLI data.