A Deep Learning Framework for Detecting Underwater Trash

The increasing environmental concerns related to underwater pollution and marine debris necessitate effective methods for detecting and managing underwater trash. This paper presents a novel approach for underwater trash detection utilizing the YOLOv8 (You Only Look Once version 8) object detection framework. YOLOv8 is known for its efficiency and accuracy in objectdetection tasks, making it a suitable choice for detecting submerged waste in aquatic environments. Innovative data augmentation techniques and pre-processing steps are employed to enhance the model's robustness and adaptability to underwater environments. Our tests' outcomes show how well the suggested method works to precisely recognize and classify underwater trash. This technology has the potential to aid in marine conservation efforts, enabling the timely removal of waste materials from aquatic ecosystems and reducing the harm caused to marine life.