Augmenting Underwater Plastic Detection: A Study with YOLO-V8m on Enhanced Datasets

Abstract

Ocean animals and plants are in danger because there is a lot of trash in the water. This old way of keeping an eye on trash made of plastic in the ocean is hard to do, takes a long time, and can’t be used to check the whole ocean. In this plan, the new deep learning algorithms, called You Only Look Once version 8 (YOLOv8), are used to find and sort plastics in the ocean right away. There’s a new way to handle this subject. Everyone knows YOLOv8 for how fast and well it can find things. It has been changed in this study to deal with the problems that only happen when taking pictures underwater, like changing light, cloudy water, and different kinds of plastic. A unique set of labeled underwater pictures are used to help it find different kinds of plastic very accurately and quickly. They are different sizes and shapes. Together, YOLOv8 and underwater robotic platforms make a useful and adaptable way to always keep an eye on the ocean. This collects useful information that can be used to protect the environment and lower pollution. Better tools for keeping an eye on the environment are made possible by the study. It also shows how cutting edge AI can be used to solve big environmental problems.