<![CDATA[The rapid increase in waste generation due to urbanization and population growth has necessitated more efficient waste management solutions. Traditional waste sorting methods rely on manual labor, which is time-consuming, error-prone, and inefficient at large scales. This paper proposes an automated waste classification system using YOLOv11, the latest iteration of the YOLO family, which is known for its high speed and accuracy in object detection. By leveraging a custom dataset containing 10,464 labeled waste images from various categories—such as biodegradable, plastic, metal, paper, and glass—this study trains and evaluates a deep learning model capable of real-time waste identification and categorization. Experimental results demonstrate that YOLOv11 achieves high detection accuracy, with an overall classification accuracy of 94% and a mean average precision (mAP) exceeding previous methods. The model effectively differentiates between various waste types, though some misclassifications occur, particularly between visually similar materials like transparent plastic and glass. Performance metrics, including precision and recall, indicate the robustness of the proposed system in real-world applications. This research highlights the potential of YOLOv11 for integration into smart waste management systems, such as automated sorting machines and AI-powered recycling bins, to enhance efficiency and reduce environmental impact. Future work will focus on optimizing model performance by incorporating additional training data, applying advanced image augmentation techniques, and exploring hybrid approaches such as texture analysis and spectral imaging to improve classification accuracy. The implementation of this technology is expected to streamline waste recycling processes, minimize contamination in recyclable materials, and contribute to sustainable waste management practices.]]>
