Automated Fish Species Identification Using Deep Learning Models: A Comprehensive Study
Keywords:
Automated classification, Convolutional neural networks, Computer vision, Deep learning, Fish species identification, Marine conservationAbstract
Automated identification of fish species is essential for the conservation of marine biodiversity, effective fisheries management, and ecological research. Conventional methods of identification are often labor-intensive, necessitate specialized expertise, and are susceptible to human error. In this study, we explore deep learning-based approaches for accurate and efficient fish species classification. Multiple state-of-the-art Convolutional Neural Network (CNN) architectures, including ResNet, EfficientNet, and Vision Transformers, are evaluated on benchmark fish datasets. The experiments assess model performance based on accuracy, precision, recall, and inference speed. Additionally, the study addresses challenges such as occlusion, underwater lighting variations, and species with similar morphological features. The deep learning models significantly outperform traditional methods, with the best-performing model achieving an accuracy of over 95%. The study also discusses dataset augmentation, transfer learning, and real-time deployment considerations. Our findings highlight the potential of deep learning for automated fish identification, paving the way for scalable, real-world applications in marine research and conservation.
References
J. C. Ovalle, C. Vilas, and L. T. Antelo, "On the use of deep learning for fish species recognition and quantification on board fishing vessels," Marine Policy, vol. 139, p. 105015, May 1, 2022. https://doi.org/10.1016/j.marpol.2022.105015
A. Saleh, M. Sheaves, D. Jerry, and M. R. Azghadi, "Applications of deep learning in fish habitat monitoring: A tutorial and survey," Expert Syst. Appl., vol. 238, p. 121841, Mar. 15, 2024. https://doi.org/10.1016/j.eswa.2023.121841
D. Siri, G. Vellaturi, S. H. Ibrahim, S. Molugu, V. S. Desanamukula, R. Kocherla, and R. Vatambeti, "Enhanced deep learning models for automatic fish species identification in underwater imagery," Heliyon, vol. 10, no. 15, Aug. 15, 2024. https://doi.org/10.1016/j.heliyon.2024.e35217
J. S. Murugaiyan, P. Mirunalini, D. Thenmozhi, and V. Muthukumar, “Fish species recognition using transfer learning techniques,” International Journal of Advances in Intelligent Informatics, vol. 7, no. 2, pp. 188–188, Jul. 2021, doi: https://doi.org/10.26555/ijain.v7i2.610.
M. J. Islam, C. Edge, Y. Xiao, P. Luo, M. Mehtaz, C. Morse, S. S. Enan, and J. Sattar, "Semantic segmentation of underwater imagery: Dataset and benchmark," in Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS), Oct. 24, 2020, pp. 1769–1776. https://doi.org/10.1109/IROS45743.2020.9340821
V. Pagire, A. Phadke, and H. J., "A deep learning approach for underwater fish detection," J. Integr. Sci. Technol., vol. 12, no. 3, pp. 1–10, 2024. https://doi.org/10.62110/sciencein.jist.2024.v12.765
M. N. Ayyad and S. S. Abu-Naser, "Fish classification using deep learning," Int. J. Acad. Inf. Syst. Res., vol. 8, no. 4, pp. 51–58, 2024. https://philpapers.org/rec/AYYFCU
A. Mohanty, G. Goldsztein, and R. Pellegrin, "Fish species image classification using convolutional neural networks," J. Student Res., vol. 11, no. 3, pp. 1–12, Aug. 2022. https://doi.org/10.47611/jsrhs.v11i3.3058
M. Hamzaoui, M. O. Aoueileyine, L. Romdhani, and R. Bouallegue, "An improved deep learning model for underwater species recognition in aquaculture," Fishes, vol. 8, no. 10, p. 514, Oct. 2023. https://doi.org/10.3390/fishes8100514
D. Marrable, K. Barker, S. Tippaya, M. Wyatt, S. Bainbridge, M. Stowar, and J. Larke, "Accelerating species recognition and labelling of fish from underwater video with machine-assisted deep learning," Front. Mar. Sci., vol. 9, p. 944582, Aug. 2, 2022. https://doi.org/10.3389/fmars.2022.944582
T. Hasegawa, K. Kondo, and H. Senou, "Transferable deep learning model for the identification of fish species for various fishing grounds," J. Mar. Sci. Eng., vol. 12, no. 3, p. 415, Feb. 2024. https://doi.org/10.3390/jmse12030415
V. Allken, N. O. Handegard, S. Rosen, T. Schreyeck, T. Mahiout, and K. Malde, "Fish species identification using a convolutional neural network," ICES J. Mar. Sci., vol. 76, no. 1, pp. 342–349, 2019. https://doi.org/10.1093/icesjms/fsy147
F. F. Khan, X. Li, A. J. Temple, and M. Elhoseiny, "Fishnet: A large-scale dataset and benchmark for fish recognition, detection, and functional trait prediction," in Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV), 2023, pp. 20496–20506. https://doi.org/10.1109/ICCV51070.2023.01874
J. J. Mol and S. A. Jose, "Fish species classification using optimized deep learning model," Int. J. Adv. Comput. Sci. Appl., vol. 13, no. 9, pp. 825–837, 2022. https://doi.org/10.14569/IJACSA.2022.0130996
P. Prasenan and C. D. Suriyakala, "Fish species classification using a collaborative technique of firefly algorithm and neural network," EURASIP J. Adv. Signal Process., vol. 2022, no. 1, p. 116, Dec. 2, 2022. https://doi.org/10.1186/s13634-022-00950-8
J. Deka, S. Laskar, and B. Baklial, “Automated Freshwater Fish Species Classification using Deep CNN,” Journal of The Institution of Engineers , vol. 104, no. 3, pp. 603–621, Apr. 2023, doi: https://doi.org/10.1007/s40031-023-00883-2.
N. Garcia-d’Urso, A. Galan-Cuenca, P. Pérez-Sánchez, P. Climent-Pérez, A. Fuster-Guillo, J. Azorin-Lopez, M. Saval-Calvo, J. E. Guillén-Nieto, and G. Soler-Capdepón, "The DeepFish computer vision dataset for fish instance segmentation, classification, and size estimation," Sci. Data, vol. 9, no. 1, p. 287, Jun. 9, 2022. https://doi.org/10.1038/s41597-022-01416-0
S. Y. Alaba,M .M. Nabi, C. Shah, J. Prior "Class-aware fish species recognition using deep learning for an imbalanced dataset," Sensors, vol. 22, no. 8268, pp. 1–18, 2022. https://doi.org/10.1016/j.marpol.2022.105015
