Unsupervised vehicle recognition using incremental reseeding of acoustic signatures
This addresses vehicle classification for applications like traffic management or military targeting, but appears incremental as it builds on spectral clustering methods.
The paper tackled unsupervised vehicle recognition from roadside audio sensors by using incremental reseeding on spectral embeddings of acoustic signatures, achieving accurate identification of individual vehicles.
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.