DefinePK

Abstract

This paper studies the performance of RIM support vector machine in the analysis of SAR based images. Analysis of complex Synthetic Aperture Radar images remains an inspiring and unsolved problem in the field of research. This paper proposes an optimal methodology to classify SAR based images with the help of support vector machines embedded RIM. The methodology which is proposed in this study is to classify the SAR based images is based on machine learning algorithms. The proposed implicit SAR image classification methodology has got many application areas such as filtering, routing relevant images to suitable databases and search engines. Proposed methodology is described by high dimensional data in which every pixel of SAR image is treated as an attribute. All SAR images used under this study are collected from publicly available AXA EORC database. This paper describes a mathematical model for automatic SAR image classification which is implemented in R programming language. Many algorithms were proposed to classifying SAR images but one of the most promising methodology is RIM support vector machine. The results shown in this paper to classify SAR images are highly effective with accuracy of 94% without heuristic and greedy concepts.