A Framework for Satellite Imaginary using Deep Sat-4 and Deep Sat-6 Datasets

Abstract

Satellite image classification is important for many real-time applications like environmental monitoring, disaster response and law enforcement.  These real time applications require the manual identification of objects. Satellite imaginary sorting is an vital task for remote sensing, machine learning and computer vision applications. The high variability of data most of the latest categorization approaches not appropriate for handle the imaginary  datasets. In this proposed system proposes the SAT-4 and SAT-6 deep sat datasets imaginary classifications using deep learning algorithms. In this proposed produces the classifications accuracy 98.3 %. Its more accuracy compared to the previous systems. Also in this paper we describe the dataset SAT-6 sand SAT-6.The proposed approach better representations for satellite imaginary.