Back-Propagation Neural Network (BP-NN) model for the detection of borer pest attack

Abstract

Abstract The multi-disciplinary agri-technologies domain have paved a way to the big data technologies, through Machine learning. Pest management is one of the most important problems facing farmers. A normal human monitoring cannot accurately predict the amount and intense of pests attacked. The issue of plant pests and diseases detection of agriculture has been tackled using the various available Neural Network (NN) techniques to process spectral data. In this manuscript, authors have presented a Back propagation Neural Network ( BP-NN) model, which was developed on data of the reflectance spectra (in range of 400 to 900 nm) cashew trees leaves infested with a borer pest attack as well asgood leaves spectra. With the help of BP-NN model the classification accuracy was foundto be 85% which is quite good. However, the accuracy of the model needs to be improved with better trainingalgorithmandlargerdataset.