Welcome to the repository for the Underwater Animal Detection using YOLOv8 project. This project aims to detect and classify underwater animals using the YOLOv8 object detection algorithm, achieving an impressive accuracy rate of 97.12%(100 epochs). The model is trained to recognize seven distinct classes of underwater creatures: fish, starfish, jellyfish, penguin, puffin, shark, and stingray.
Object detection in underwater environments poses unique challenges due to varying lighting conditions, distortions, and the diversity of marine life. This project tackles these challenges using the state-of-the-art YOLOv8 model, which enables accurate and efficient detection of multiple animal species simultaneously.
1.Utilizes the YOLOv8 architecture for accurate and real-time object detection.
2.Trained on a diverse dataset of underwater animal images to ensure robustness.
3.Achieves an impressive accuracy of 97.12% across seven different animal classes.
4.The model can be easily extended to include additional classes or adapt to new underwater environments.
1.Clone this repository to your local machine.
2.Install the necessary dependencies listed in the data.yaml file.
3.Download the pre-trained YOLOv8 weights compatible with the number of classes used in this project.
4.Run the provided script to perform object detection on your own images or videos.