<![CDATA[Waste sorting remains a significant challenge in environmental management, mainly because it is still performed manually, which may lead to misclassification of waste types. To address this issue, this study develops a waste classification system based on deep learning by utilizing Computer vision technology. The system is designed to classify five types of waste, namely plastic, glass, metal, paper, and organic waste, based on images uploaded by users. The proposed method includes dataset collection from Kaggle and Google Images, data Preprocessing, model design using the EfficientNet-B0 architecture with a transfer learning approach, and model training until the final model is saved in the .keras format. The system is then implemented into a web-based application using the Flask framework and deployed through Railway cloud computing services to enable online accessibility. The experimental results show that the system is able to classify waste images automatically with good performance and can be accessed flexibly across different devices. This system is expected to improve the efficiency and effectiveness of waste sorting in real-world applications.]]>
Copyrights © 2026
