Feature Selection and Optimization in IoT based Smart Agriculture

Abstract

Agriculture is the main backbone of any developing country in which it is primary newlineresource. It leads to the economic growth of such countries. For improvement in the agricultural product, climate change is the most important factor. When it comes to agriculture, any country needs both excellent quality and large quantities of farm products. That s why it s so important to pinpoint and manage plant quality-reducing factors. Diseases, stresses, and pathogens affecting plant development and quality will be the main points of discussion, as well as the methods and algorithms utilized to learn more about them. Hybrid intelligent system uses techniques like genetic algorithms, Support Vector Machines, neural network to classify different disease types. So, my research will include investigation of efficient feature extraction and optimization techniques. Here role of IoT will be monitoring of different environmental parameters like temperature, pressure, humidity, pH etc. of the agricultural field. Presently, it is the matter of great concern to control pest by using reduced amount newlineof chemical fertilizer. Beneficial insects that eat dangerous insects and mites are known as predators. But both pest and predators cause huge loss in production. So, we have studied the tritrophic model which consists of tea plant, pest and predator. We have studied stability of positive fixed points and boundary (boundedness). For positive equilibrium point, the global stability properties are also examined. The dynamical behavior of the system has been shown analytically as well as numerically by computer simulation. Two step Adams-Bashforth and Runge-Kutta method of fourth order has been employed in the solution and simulation of the system of non-linear ordinary differential equation of tritrophic model. newline