Machine Learning Technique Approaches Versus Statistical Methods in Classification of Multispectral Remote Sensing Data using Maximum Likelihood Classification: Koluru Hobli, Bellary Taluk, District, Karnataka, India

Abstract

Classification is challenging task on complex features of Remote sensing satellite imageries color pixels variability of patterns. Machine learning techniques have delivered the improved in accuracy of classification of patterns of features. Remote sensing color based imageries having hard to cluster color pixels with variability in intensity of colors. Challenges in estimation of various features viz, crop fields, fallow land, buildings, roads, rivers, water bodies, forest, and other trivial items. Urge in estimation of crop yield predictions through satellite imageries. We are attempted to converging accuracy of estimation of vegetation crop yield of fields. Kappa coefficient to achieve high degree accuracy estimation of crop wise with suitable thresh hold to ground truth data.