Applications of ANN for Agriculture Using Remote Sensed Data

Introduction

In India more than 60% of population is depending on agriculture. In agriculture, crop yield estimation before harvesting is a challenging task. Many models have been developed in Asia, USA, Europe and elsewhere in the country, but due to the complexity of agriculture ecosystem, yield prediction is still based on the traditional methods or the statistical methods. Artificial Neural Network (ANN) is one of the most powerful and self-adaptive model for crop yield estimation using remote sensing. This method employs a nonlinear response function that iterates many times in a special network structure in order to learn the complex functional relationship between input and output training data. Once trained, an ANN model can remember a functional relationship and be used for further calculation. For these reasons, the ANN concept has been widely used to develop models, especially in strongly nonlinear, complicated systems. Since Remote sensing provides the availability of large data in time with respect to the crop season would be combined with ANN to develop an efficient model for predicting the yield before harvesting. ANN and satellite remote sensing has got an unlimited scope in the sector of agriculture these days as it is being used for land resource mapping, weed detection, pesticide management, soil health mapping, crop yield estimation, and for assessment of natural calamities. In India the technology is being promoted by ministry of agriculture through MGNREGA scheme for rural area to assist farmers remotely.

The chapter is presented as follows. Section II covers overview of ANN, followed by detailed study of remote sensing in section III. Applications of ANN in agriculture using remote sensed data is briefed in section IV, followed by contribution to chapter, future scope for research and conclusions in the remaining sections.