Classification of Bryde's whale individuals using high-resolution time-frequency transform and support vector machines

Tary, Jean-Baptiste; Peirce, Christine; Hobbs, Richard

Abstract

Whales generate vocalizations which may, deliberately or not, encode caller identity cues. In this study, we analyze calls produced by Bryde’s whales and recorded by ocean-bottom arrays of hydrophones deployed close to the Costa Rica Rift in the Panama basin. These repetitive calls, consisting of two main frequency components at ~20 and ~36 Hz, have been shown to follow five coherent spatio-temporal tracks. Here, we use a high-resolution time-frequency transform, the 4th-order Fourier synchrosqueezing transform (FSST4), to extract time-frequency characteristics (ridges) from each call to appraise their suitability for identifying individuals from each other. Focusing on high-quality calls recorded less than 5 km from their source, we then cluster these ridges using a Support Vector Machine (SVM) model resulting in an average cross-validation error of ~11% and balanced accuracy of ~86 ±5%. Comparing these results with those obtained using the standard short-time Fourier transform, k-means clustering, and lower-quality signals, the FSST4 approach, coupled with SVM, substantially improves classification. Consequently, the Bryde’s whale calls potentially contain individual-specific information, implying that individuals can be studied using ocean-bottom data.

Citation

Tary, J.-B., Peirce, C., & Hobbs, R. (in press). Classification of Bryde's whale individuals using high-resolution time-frequency transform and support vector machines. The Journal of the Acoustical Society of America, 157(3), 2091–2101. https://doi.org/10.1121/10.0036223

• Journal Article Type: Article
• Acceptance Date: Mar 5, 2025
• Deposit Date: Mar 6, 2025
• Journal: The Journal of the Acoustical Society of America
• Print ISSN: 0001-4966
• Publisher: Acoustical Society of America
• Peer Reviewed: Peer Reviewed
• Volume: 157
• Issue: 3
• Pages: 2091–2101
• DOI: https://doi.org/10.1121/10.0036223
• Public URL: https://durham-repository.worktribe.com/output/3680785
• Publisher URL: https://pubs.aip.org/asa/jasa