Freshwater Fish Detection Based on the Yolov8 Model
DOI:
https://doi.org/10.62051/ijcsit.v5n3.13Keywords:
Freshwater fish, Yolov8, Target detection, Deep learning, Water environment protectionAbstract
Fish is a class of important components of river and lake ecosystems, and the judgment of their population migration and the statistics of their number is an important task in water environment regulation. Current intelligent identification methods based on deep learning models provide efficient and accurate means. In this paper, based on the Yolov8 model, we realize the recognition of freshwater fish species under the underwater complex environment. We add the attention mechanism of random masks to the model to reduce the computational complexity. A dataset containing ten freshwater fish species is also constructed from integrating web open datasets. Experimental results show that the proposed method achieves good recognition results, and the evaluation index Precision\Recall\ mAP50\ mAP50-95 is achieved.
Downloads
References
[1] Grabowska, Joanna, J. Kotusz, and A. Witkowski. "Alien invasive fish species in Polish waters: an overview." Folia Zoologica 59(2010):73-85.
[2] Shuai, Xinhui Zhao, Tian. "Biological invasions undermine the functional diversity of fish community in a large subtropical river." Biological invasions 20.10(2018).
[3] Redmon, Joseph, et al. "You Only Look Once: Unified, Real-Time Object Detection." Computer Vision & Pattern Recognition IEEE, 2016.
[4] Redmon, Joseph, and A. Farhadi. "YOLOv3: An Incremental Improvement." arXiv e-prints (2018).
[5] Li, Chuyi, et al. "YOLOv6: A single-stage object detection framework for industrial applications." arXiv preprint arXiv:2209.02976 (2022).
[6] Yaseen, M. "What is YOLOv8: An in-depth exploration of the internal features of the next-generation object detector. arXiv 2024." arXiv preprint arXiv:2408.15857.
[7] Li, Xiang, et al. "Generalized focal loss: Learning qualified and distributed bounding boxes for dense object detection." Advances in neural information processing systems 33 (2020): 21002-21012.
[8] Han, Yanling, et al. "Underwater Holothurian target-detection algorithm based on improved CenterNet and scene feature fusion." Sensors 22.19 (2022): 7204.
[9] Wang, Jing, et al. "YOLOv6-ESG: A lightweight seafood detection method." Journal of Marine Science and Engineering 11.8 (2023): 1623.
[10] Wang, Jun, et al. "B-YOLOX-S: a lightweight method for underwater object detection based on data augmentation and multiscale feature fusion." Journal of Marine Science and Engineering 10.11 (2022): 1764.
[11] Ren, Liqiu, et al. "An Underwater Target Detection Algorithm Based on Attention Mechanism and Improved YOLOv7." Computers, Materials & Continua 78.2 (2024).
[12] Wu, Fei, et al. "Fish target detection in underwater blurred scenes based on improved YOLOv5." Ieee Access 11 (2023): 122911-122925.
[13] Vaswani, Ashish, et al. "Attention is all you need." Advances in neural information processing systems 30 (2017).
[14] He, Kaiming, et al. "Masked autoencoders are scalable vision learners." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2022.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Computer Science and Information Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
