<![CDATA[Effective waste management is a critical challenge in urban environments, necessitating the development of automated systems for efficient waste classification. This study aims to develop an intelligent system for detecting and classifying waste into organic and inorganic categories using the You Only Look Once (YOLO)v11 object detection algorithm. The proposed system identifies the presence and location of waste objects through bounding boxes and classifies them based on their visual characteristics to facilitate real-time detection. Experimental results demonstrate the model's effectiveness, achieving an overall accuracy of 0.717. Performance analysis indicates consistent reliability, with "Anorganik" waste achieving an accuracy of 0.708 and "Organik" waste reaching an accuracy of 0.725. Notably, despite the moderate accuracy, the model demonstrates significant robustness when deployed in real-world conditions; it is highly capable of identifying multiple objects within a single frame and maintains consistent detection performance even when objects are overlapping or clustered, confirming its viability for practical, real-time sorting applications. Despite these promising results, the study identifies several influencing factors, including object clustering, visual material similarity, lighting conditions, and camera-to-object distance. The current prototype faces limitations due to dataset size, particularly in detecting uncommon waste objects. Future development efforts will focus on expanding the training dataset to include a wider variety of waste items and enhancing robustness for detecting small or occluded objects in real-world scenarios.]]>
Copyrights © 2026
