Document Type
Conference Proceeding
Publication Date
2019
Publisher
IEEE
Source Publication
2019 53rd Asilomar Conference on Signals, Systems, and Computers
Source ISSN
978-1-7281-4301-9
Abstract
A recently developed improved spectrogram that uses the discrete fractional Fourier transform (DFRFT) is used to retrieve the vibration signature that represents targets in synthetic aperture radar (SAR) data. The retrieved signature is used as input to a feature extraction process, which characterizes the vibration waveform using the DFRFT as well as histograms and statistics. The study of the performance of two classifiers, one trained with features extracted from vibration measurements and the other trained with feature extracted from simulated SAR data generated from the same vibration measurements, validates the suitability of the DFRFT-based spectrogram for retrieving and characterizing the dynamics of vibrating objects in SAR images.
Recommended Citation
Pérez, Francisco; Santhanam, Balasubramaniam; Shrestha, Bipesh; Gerstle, Walter; and Hayat, Majeed M., "Fractional Spectrogram for Characterizing and Classifying Vibrating Objects in SAR Images" (2019). Electrical and Computer Engineering Faculty Research and Publications. 641.
https://epublications.marquette.edu/electric_fac/641
Comments
Accepted version. Published in the proceedings of the 2019 53rd Asilomar Conference on Signals, Systems, and Computers. DOI. © 2019 The Institute of Electrical and Electronics Engineer. Used with permission.