Taxonomy‐informed hierarchical deep learning for otolith image identification in coral reef fishes
Journal of Fish Biology · 29.10.2026
DOI: 10.1111/jfb.70717 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Fish Biology, online erschienen am 29.10.2026. Verfasst von Zhipeng Kang und 7 weiteren Autor:innen. Beteiligte Einrichtung: South China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences/Key Laboratory of South China Sea Fishery Resources Exploitation & …. Die Arbeit nennt 56 Literaturangaben. Gefördert von: Science & Technology Fundamental Resources Investigation Program, National Natural Science Foundation of China, Ministry of Agriculture and Rural Affairs of the People's Republic of China.
Abstract (Originalsprache)
Existing studies on automatic fish otolith identification have predominantly adopted classification frameworks that use species as the sole output level, limiting the effective use of hierarchical information embedded in fish taxonomy. In this study, using otolith images of coral reef fishes collected from the Qilianyu Islands, Xisha Islands, China, during 2020–2025, we developed a hierarchical deep learning model incorporating taxonomic constraints and compared it with a non‐hierarchical baseline model. Both models were trained, validated and tested on otolith images from 25 species, and the generalization performance of the hierarchical model was further evaluated on 33 species not included in model training. The results showed that the hierarchical model achieved higher recognition accuracy than the non‐hierarchical model at the family, genus and species levels, with species‐level accuracy improving from 94.80% to 98.00%. Both micro‐averaged and macro‐averaged metrics also improved, and misclassifications across families and genera were markedly reduced. UMAP and Grad‐CAM visualizations showed that the hierarchical model learned discriminative embedding features with clear species‐level separation, and that its high‐response regions were mainly concentrated along the otolith contour and local structural features. External validation further showed that, under unseen‐species conditions, the hierarchical model achieved strong species recognition with only a small number of support samples. Overall, these results show that taxonomic hierarchy constraints can improve otolith image identification and support the development of a biologically informed framework that is extensible to unseen species.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- South China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences/Key Laboratory of South China Sea Fishery Resources Exploitation & Utilization, Ministry of Agriculture and Rural Affairs/Observation and…
- Scientific Observation and Research Station of Xisha Island Reef Fishery Ecosystem of Hainan Province/Sanya Tropical Fisheries Research Institute, Sanya, China.
- College of Fisheries, Huazhong Agricultural University, Wuhan, China.
- College of Animal Science, Guizhou University, Guiyang, China.
ORCID-Profile: Teng Wang, Xianghong Dong
Förderung
- Science & Technology Fundamental Resources Investigation Program
- National Natural Science Foundation of China
- Ministry of Agriculture and Rural Affairs of the People's Republic of China
- Central Public-interest Scientific Institution Basal Research Fund, South China Sea Fisheries Research Institute, CAFS
- Central Public-interest Scientific Institution Basal Research Fund, Chinese Academy of Fishery Sciences
Zitieren
APA
Kang, Z., Wang, T., Liu, Y., Shen, J., Xie, H., Ao, X., Dong, X., & Liu, Y. (2026). Taxonomy‐informed hierarchical deep learning for otolith image identification in coral reef fishes. Journal of Fish Biology. https://doi.org/10.1111/jfb.70717
Vancouver
Kang Z, Wang T, Liu Y, Shen J, Xie H, Ao X, et al. Taxonomy‐informed hierarchical deep learning for otolith image identification in coral reef fishes. Journal of Fish Biology. 2026. doi:10.1111/jfb.70717
BibTeX
@article{Kang2026,
author = {Kang, Zhipeng and Wang, Teng and Liu, Yu and Shen, Jianzhong and Xie, Hongyu and Ao, Xuefu and Dong, Xianghong and Liu, Yong},
title = {Taxonomy‐informed hierarchical deep learning for otolith image identification in coral reef fishes},
journal = {Journal of Fish Biology},
year = {2026},
doi = {10.1111/jfb.70717}
}
EUVIMED