Date of Publication :25th July 2024

Abstract:"Predicting Underwater Waste using YOLOv5" In response to the urgent need to solve environmental pollution, our research center developed a vision-based autonomous underwater wastewater cleaning robot. This paper presents a new litter detection method based on YOLOv5, a state-of-the-art object detection algorithm known for its efficiency and accuracy. YOLOv5 algorithm was chosen as the main neural network for task detection. We use new technology designed for YOLOv5 architecture to increase verification accuracy. An optimization method is also used to demonstrate speed while maintaining accuracy. Through extensive testing, our wireless drivers have demonstrated superior performance in detecting and collecting underwater debris. The results demonstrate the effectiveness of the YOLOv5-based approach in solving environmental problems.

Reference :