Microwave Based Classification Of Material : A Statistical Approach

Abstract

In this paper we present a method for classification of material based on their electromagnetic property. Microwave radar has emerged as useful tool in many remote sensing applications. When microwave propagates from one medium to other, it causes reflections depending on permittivity and permeability of medium. Microwave in X-band range is used for scanning the sheets of various materials like Copper, Metal, Plywood and Acrylic in free space. Depending on their electromagnetic property, reflections from each are measured over the region of interest and finally radar image of reflections from each sheet is obtained using Digital Image Processing tools in MATLAB.Further various statistical features such as Entropy, Standard deviation, Mean and Energy etc are extracted for classification as Metal or Non metal. Results show good performance.