<![CDATA[Waste management in the school environment poses a significant challenge due to the high volume of mixed organic and inorganic waste, which hinders the recycling process. The utilization of object detection technology can offer a solution. However, previous studies employed older YOLO architectures, which still have room for improvement. This research aims to implement a detection model to differentiate between organic and inorganic waste within the school environment, with a focus on the implementation of the YOLOv11 architecture. The method used is a Convolutional Neural Network (CNN) featuring the YOLOv11 architecture, utilizing a public dataset from Kaggle that is divided into 7 waste classes. The research stages include image preprocessing, image augmentation, and dataset partitioning using Stratified K-Fold Cross Validation. The model’s performance will be evaluated using mean Average Precision (mAP), precision, recall, and F1-score metrics. Subsequently, the model will be developed into a desktop-based system application. The result of this study are expected to provide an accurate and efficient waste detection model to assist in recognizing the types of waste present in the school environment.]]>
Copyrights © 2026
