A new spectral estimation-based feature extraction method for vehicle classification in distributed sensor networks

Abstract

Ground vehicle detection and classification with distributed sensor networks is of growing interest for bordersecurity. Different sensing modalities including electro-optical, seismic, and acoustic were evaluated individually and incombination to develop a more efficient system. Despite previous works that mostly studied frequency-domain featuresand acoustic sensors, in this work we analyzed the classification performance for both frequency and time-domain featuresand seismic and acoustic modalities. Despite their infrequent use, we show that when fused with frequency-domainfeatures, time-domain features improve the classification performance and reduce the false positive rate, especially forseismic signals. We investigated the performance of seismic sensors and showed that the classification performance varieswith the type of road due to the distinct spectral characteristics of the medium. Our proposed classifier fuses time andfrequency-domain features and acoustic and seismic modalities to achieve the highest classification rate of 98.6% usinga relatively small number of features.