Intelligent Robotic Detection of Cigarette Butts: A Neural Network Approach for Visual Pollution Monitoring

The research offers a method for detecting cigarette butts in outdoor settings using YOLOv8, enabling intelligent robotic vision scanning. Effective waste management strategies are essential as cigarette butt litter continues to be a problem for the environment. Our work suggests an intelligent system that can identify cigarette butt waste by using cutting-edge robotics and computer vision technology. After a significant amount of experiments, the YOLOv8 model performs well, with a mean Average Precision (mAP) of 88.8%. The model's ability to precisely detect and localize cigarette butt objects is highlighted by its high mAP. Moreover, the model is flexible enough to handle situations with different levels of background and object density complexity. Our research offers a feasible approach to mitigate the damaging effects of cigarette butt litter on ecosystems by highlighting potential uses for automated waste management and environmental conservation initiatives.