Feature analysis for indoor radar target classification

Travis D. Bufler; Ram M. Narayanan

doi:10.1117/12.2224041

12 May 2016 Feature analysis for indoor radar target classification

Travis D. Bufler, Ram M. Narayanan

Proceedings Volume 9829, Radar Sensor Technology XX; 98290Y (2016) https://doi.org/10.1117/12.2224041
Event: SPIE Defense + Security, 2016, Baltimore, MD, United States

Abstract

This paper analyzes the spectral features from human beings and indoor clutter for building and tuning Support Vector Machines (SVMs) classifiers for the purpose of classifying stationary human targets. The spectral characteristics were obtained through simulations using Finite Difference Time Domain (FDTD) techniques where the radar cross section (RCS) of humans and indoor clutter objects were captured over a wide range of frequencies, polarizations, aspect angles, and materials. Additionally, experimental data was obtained using a vector network analyzer. Two different feature sets for class discrimination are used from the acquired target and clutter RCS spectral data sets. The first feature vectors consist of the raw spectral characteristics, while the second set of feature vectors are statistical features extracted over a set frequency interval. Utilizing variables of frequency and polarization, a SVM classifier can be trained to classify unknown targets as a human or clutter. Classification accuracy over 80% can be effectively achieved given appropriate features.

Citation Download Citation

Travis D. Bufler and Ram M. Narayanan "Feature analysis for indoor radar target classification", Proc. SPIE 9829, Radar Sensor Technology XX, 98290Y (12 May 2016); https://doi.org/10.1117/12.2224041

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available