Rice Nitrogen Status Estimation Of Western Tract Of Odisha Using SVM Based On Color Feature: A Comparative Analysis With LCC

Abstract

Rice is the most important human food crop in the world, directly feeding more people than any other crop. Rice crop contains different types of nutrients. Sometime due to deficiency of nutrition, rice crops undergo in a hindrance effects on its growth and production. Nitrogen is the main component among all nutrients for rice crop growth and production. The leaf nitrogen concentration (LNC) is highly correlated with chlorophyll content. There are many devices like Leaf Color Chart (LCC), SPAD, at LEAF+ for measurement of chlorophyll &/ or nitrogen. As these devices are cost effective and unavailable with all farmers, a digitize image acquisition and interpretation system is required. This paper proposed a site-specific approach for nitrogen status estimation based on image captured by smart phone, matching with LCC and train the support vector machine (SVM) using radial basis function (RBF). The proposed methodology have accuracy of 98%, which implies it is quite successful to classify the rice leaf according to their nitrogen status.