Enhancing Aquarium Fish Classification through YOLO: A Deep Learning Approach

Abstract

In recent years, advancements in deep learning have revolutionized the field of image classification and object detection. This study presents a novel application of the YOLO (You Only Look Once) model for the classification of aquarium fish species. The model is designed to accurately identify and classify five distinct species: Percula Clownfish, Moorish Idol, Yellow Tang, Queen Angel Fish, and Blue Tang. Our approach leverages the YOLO v8 architecture due to its balance between speed and accuracy, making it suitable for real-time applications. The model was trained on a comprehensive dataset of annotated images collected manually from the internet, capturing various poses, lighting conditions, and background variations to enhance robustness. Experimental results demonstrate that the YOLO v8 model achieves high precision and recall rates, with an overall accuracy exceeding 95.78 %. The model's performance was evaluated using accuracy, precision, recall and F1-Score metrics. This research showcases the potential of advanced neural networks in automated aquatic biodiversity monitoring and contributes to the growing body of work in marine biology and conservation technology.