YOLOv8-S in Plastic Waste Segregation

Plastic pollution poses a significant threat in Sustainable Development Goal (SDG) 11, which aimed to make cities and human settlements inclusive, safe, resilient, and sustainable [1]. Despite mitigation of the said problem, traditional manual sorting methods remain inefficient and prone to errors [2]. To address this issue, the researchers utilized You Only Look Once version 8-Small (YOLOv8-S), a deep learning model optimized for real-time object detection, to classify six major plastic types: Polyethylene Terephthalate (PET), High-Density Polyethylene (HDPE), Low-Density Polyethylene (LDPE), Polyvinyl Chloride (PVC), Polypropylene (PP), and Polystyrene (PS) under various conditions such as deformed, dirty, and overlapping. The trained model achieved an mAP@0.5 of 0.702 and an mAP@0.5:0.95 of 0.792, indicating strong detection performance. Macro average precision, recall, and F1-score were 0.797, 0.431, and 0.529, respectively, suggesting high accuracy but moderate sensitivity. These findings demonstrate the potential of YOLOv8-S to enhance the accuracy and reliability of plastic waste segregation and support more efficient and sustainable waste management systems.