Unsupervised vehicle recognition using incremental reseeding of acoustic signatures

2018/02/17 by Justin Sunu, Sunu, Justin, Blake Hunter +3
Computer Science · Mathematics · Physics and Astronomy · #Music and Audio Processing #Speech Recognition and Synthesis #Speech and Audio Processing #cs.LG #physics.data-an #stat.ML

paper · pdf · doi:10.48550/arxiv.1802.06287

arxiv created 2018/02/17 · arxiv updated 2018/02/20

Abstract

Vehicle recognition and classification have broad applications, ranging from traffic flow management to military target identification. We demonstrate an unsupervised method for automated identification of moving vehicles from roadside audio sensors. Using a short-time Fourier transform to decompose audio signals, we treat the frequency signature in each time window as an individual data point. We then use a spectral embedding for dimensionality reduction. Based on the leading eigenvectors, we relate the performance of an incremental reseeding algorithm to that of spectral clustering. We find that incremental reseeding accurately identifies individual vehicles using their acoustic signatures.

Related

• Dimensionality reduction for acoustic vehicle classification with spectral embedding
• Intelligent Acoustic Module for Autonomous Vehicles using Fast Gated Recurrent approach
• MVD:A Novel Methodology and Dataset for Acoustic Vehicle Type Classification
• Improved Vehicle Sub-type Classification for Acoustic Traffic Monitoring
• Intelligent Traffic Monitoring with Distributed Acoustic Sensing