Advances in deep learning-driven photo identification and meta analysis of cetaceans in large data repositories

Photo-identification of cetaceans remains a labor-intensive task, requiring expert annotation of long-tailed image datasets in which most individuals are rarely encountered. We present a scalable, end-to-end framework that automates this process using lightweight deep learning models optimized for r...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Ecological informatics Jg. 91; S. 103396
Hauptverfasser: Barnhill, Alexander, Towers, Jared R., Shaw, Tasli J.H., Arias, Magdalena, Bécares, Adrián, Doniol-Valcroze, Thomas, von Fersen, Lorenzo, Genoves, Rodrigo, Rörup, Tim, Sutton, Gary J., Thornton, Sheila, Weiss, Michael, Maier, Andreas, Nöth, Elmar, Bergler, Christian
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.11.2025
Elsevier
Schlagworte:
Data curation
Deep learning
Marine conservation
Photo identification
Resource efficient machine learning
Deep learning
Data curation
Photo identification
Resource efficient machine learning
Marine conservation
ISSN:1574-9541
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Photo-identification of cetaceans remains a labor-intensive task, requiring expert annotation of long-tailed image datasets in which most individuals are rarely encountered. We present a scalable, end-to-end framework that automates this process using lightweight deep learning models optimized for resource-constrained environments. Our modular pipeline integrates state-of-the-art detection (YOLOv8-small), individual identification via metric learning (EfficientNet-B0 with a contrastive head), and auxiliary modules for image quality scoring, side classification, and identifiability prediction. Unlike previous approaches limited to single-species applications or high-resource settings, our framework generalizes across five cetacean populations with diverse visual characteristics. We achieve top-1 identification accuracies of 0.92 for Bigg's killer whales (Orcinus orca rectipinnus), 0.96 for Southern resident killer whales (Orcinus orca ater), 0.96 for Lahille's bottlenose dolphins (Tursiops truncatus gephyreus), 0.82 for common minke whales (Balaenoptera acutorostrata scammoni), and 0.85 for humpback whales (Megaptera novaeangliae), yielding a cross-species accuracy of 0.90. To support image triage in large datasets, we include a quality scoring module that predicts image utility using learned embedding features. This module achieves an R2 of 0.799, enabling intelligent prioritization of data. Runtime evaluations show processing speeds of 1.6–3.2 images/s on CPU and 9.6–23.3 FPS with GPU acceleration, making it suitable for archival and real-time applications. We also evaluate the impact of demographic metadata (age, sex) on identification performance and provide practical recommendations for future dataset design. The system is available via a web interface designed to support real-world conservation workflows with minimal computational overhead. [Display omitted] •Modular AI pipeline filters, detects, scores, and identifies cetaceans step-by-step•Quality and orientation classifiers remove poor or unusable detections early and allow triaging of unlabelled datasets•Each step is lightweight and runs efficiently on modest hardware setups•Combines YOLOv8, EfficientNet, and contrastive learning for identification•Enables scalable, automated ID across large, heterogeneous image corpora
ISSN:1574-9541
DOI:10.1016/j.ecoinf.2025.103396