Acoustic Classification of Bird Species Using an Early Fusion of Deep Features
Birds · 01.03.2023 · 4 (1) · 138-147
DOI: 10.3390/birds4010011 ↗Kurz gefasst
Fachartikel in der Zeitschrift Birds, online erschienen am 01.03.2023. Verfasst von Jie Xie und 1 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province, Jiangsu Province Post Doctoral Fund.
Abstract (Originalsprache)
Bird sound classification plays an important role in large-scale temporal and spatial environmental monitoring. In this paper, we investigate both transfer learning and training from scratch for bird sound classification, where pre-trained models are used as feature extractors. Specifically, deep cascade features are extracted from various layers of different pre-trained models, which are then fused to classify bird sounds. A multi-view spectrogram is constructed to characterize bird sounds by simply repeating the spectrogram to make it suitable for pre-trained models. Furthermore, both mixup and pitch shift are applied for augmenting bird sounds to improve the classification performance. Experimental classification on 43 bird species using linear SVM indicates that deep cascade features can achieve the highest balanced accuracy of 90.94% ± 1.53%. To further improve the classification performance, an early fusion method is used by combining deep cascaded features extracted from different pre-trained models. The final best classification balanced accuracy is 94.89% ± 1.35%.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Jiangsu Province
- Jiangsu Province Post Doctoral Fund
Zitieren
APA
Xie, J., & Zhu, M. (2023). Acoustic Classification of Bird Species Using an Early Fusion of Deep Features. Birds, 4(1), 138-147. https://doi.org/10.3390/birds4010011
Vancouver
Xie J, Zhu M. Acoustic Classification of Bird Species Using an Early Fusion of Deep Features. Birds. 2023;4(1):138-147. doi:10.3390/birds4010011
BibTeX
@article{Xie2023,
author = {Xie, Jie and Zhu, Mingying},
title = {Acoustic Classification of Bird Species Using an Early Fusion of Deep Features},
journal = {Birds},
year = {2023},
volume = {4},
number = {1},
pages = {138--147},
doi = {10.3390/birds4010011}
}
EUVIMED