EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

YOLOv8-TF: Transformer-Enhanced YOLOv8 for Underwater Fish Species Recognition with Class Imbalance Handling

Chiranjibi Shah iD, M M Nabi iD, Simegnew Yihunie Alaba iD, Iffat Ara Ebu, Jack Prior iD, Matthew D Campbell, Ryan Caillouet, Matthew D Grossi iD, Timothy Rowell, Farron Wallace, John E Ball iD, Robert Moorhead iD

Sensors · 16.03.2025 · 25 (6) · 1846

DOI: 10.3390/s25061846 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 16.03.2025. Verfasst von Chiranjibi Shah und 11 weiteren Autor:innen. Beteiligte Einrichtung: Northern Gulf Institute, Mississippi State University, Starkville, MS 39759, USA. Hauptthemen laut MEDLINE: Fishes. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 72 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Northern Gulf Institute at Mississippi State University from NOAA's Office of Oceanic and Atmospheric Research, U.S. Department of Commerce., Northern Gulf Institute at Mississippi State University from NOAA’s Office of Oceanic and Atmospheric Research, U.S. Department of Commerce..

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In video-based fish surveys, species recognition plays a vital role in stock assessments, ecosystem analysis, production management, and protection of endangered species. However, implementing fish species detection algorithms in underwater environments presents significant challenges due to factors such as varying lighting conditions, water turbidity, and the diverse appearances of fish species. In this work, a transformer-enhanced YOLOv8 (YOLOv8-TF) is proposed for underwater fish species recognition. The YOLOv8-TF enhances the performance of YOLOv8 by adjusting depth scales, incorporating a transformer block into the backbone and neck, and introducing a class-aware loss function to address class imbalance in the dataset. The class-aware loss considers the count of instances within each species and assigns a higher weight to species with fewer instances. This approach enables fish species recognition through object detection, encompassing the classification of each fish species and localization to estimate their position and size within an image. Experiments were conducted using the 2021 Southeast Area Monitoring and Assessment Program (SEAMAPD21) dataset, a detailed and extensive reef fish dataset from the Gulf of Mexico. The experimental results on SEAMAPD21 demonstrate that the YOLOv8-TF model, with a mean Average Precision (mAP)0.5 of 87.9% and mAP0.5–0.95 of 61.2%, achieves better detection results for underwater fish species recognition compared to state-of-the-art YOLO models. Additionally, experimental results on the publicly available datasets, such as Pascal VOC and MS COCO datasets demonstrate that the model outperforms existing approaches.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

FishesAnimalsAlgorithmsEcosystem

Stichwörter der Autor:innen

SEAMAPD21YOLOv8-TFfish species recognitionobject detection

Autor:innen und Institutionen

  • Northern Gulf Institute, Mississippi State University, Starkville, MS 39759, USA.
  • The School of Engineering and Applied Sciences, Western Kentucky University, Bowling Green, KY 42101, USA.
  • Department of Electrical and Computer Engineering, James Worth Bagley College of Engineering, Mississippi State University, Starkville, MS 39762, USA.
  • National Marine Fisheries Services, Southeast Fisheries Science Center, 3209 Frederic Street, Pascagoula, MS 39567, USA.
  • NOAA Fisheries, 4700 Avenue U, Galveston, TX 77551, USA.

ORCID-Profile: Chiranjibi Shah, M M Nabi, Simegnew Yihunie Alaba, Jack Prior, Matthew D Grossi, John E Ball, Robert Moorhead

Förderung

  • Northern Gulf Institute at Mississippi State University from NOAA's Office of Oceanic and Atmospheric Research, U.S. Department of Commerce.
  • Northern Gulf Institute at Mississippi State University from NOAA’s Office of Oceanic and Atmospheric Research, U.S. Department of Commerce.

Zitieren

APA

Shah, C., Nabi, M. M., Alaba, S. Y., Ebu, I. A., Prior, J., Campbell, M. D., Caillouet, R., Grossi, M. D., Rowell, T., Wallace, F., Ball, J. E., & Moorhead, R. (2025). YOLOv8-TF: Transformer-Enhanced YOLOv8 for Underwater Fish Species Recognition with Class Imbalance Handling. Sensors, 25(6), 1846. https://doi.org/10.3390/s25061846

Vancouver

Shah C, Nabi MM, Alaba SY, Ebu IA, Prior J, Campbell MD, et al. YOLOv8-TF: Transformer-Enhanced YOLOv8 for Underwater Fish Species Recognition with Class Imbalance Handling. Sensors. 2025;25(6):1846. doi:10.3390/s25061846

BibTeX
@article{Shah2025,
  author = {Shah, Chiranjibi and Nabi, M M and Alaba, Simegnew Yihunie and Ebu, Iffat Ara and Prior, Jack and Campbell, Matthew D and Caillouet, Ryan and Grossi, Matthew D and Rowell, Timothy and Wallace, Farron and Ball, John E and Moorhead, Robert},
  title = {YOLOv8-TF: Transformer-Enhanced YOLOv8 for Underwater Fish Species Recognition with Class Imbalance Handling},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {6},
  pages = {1846},
  doi = {10.3390/s25061846}
}
Kostenlos starten →