Crop classification using feature extraction from satellite imagery

Plants are a major source of food stuff, medicine and industry. However it is an important and difficult task to recognize species of crop on earth. Therefore it is pertinent to design an appropriate recognition system of crops. As the frontier of space technology is progressed rapidly, remote sensing provides very convenient resource for development in agriculture. In this paper, we proposed a new method for crop classification from satellite imagery. In this method, first of all, we assembled satellite imagery database of different crops and then extract Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) features from each crop and then classify them by different individual classifiers. At the end we described analysis of different classifiers that which classifier performed best. Various techniques have been compared against our proposed model for determining the supremacy of our model and results have shown significant improvement in accuracy and reliability.