TEM Journal | 2026
Authors: Hung D.M.; Anh V.T.L.
DOI: 10.18421/TEM151-79
Journal: TEM Journal
Year: 2026
Publisher: UIKTEN - Association for Information Communication Technology Education and Science
Document Type: Article
Open Access: All Open Access; Gold Open Access
Cited by: 0
In Vietnam, individuals and households must classify domestic solid waste into 3 groups before transferring it to collection and transportation units, as stipulated in the Law on Environmental Protection (2020). In order to take advantage of the superior capabilities of deep learning, this paper builds a custom lightweight model that is light enough but has high accuracy to be embedded in edge devices with the aim of supporting people in classifying household waste at source in compliance with the Environmental Protection Law of Vietnam. To develop a system to help people classify household waste, a new research method based on deep learning algorithms is proposed in this study to build a lightweight model that can be easily embedded in edge devices. The experimental results show that the proposed model is highly effective in implementing waste classification support systems, proven with accuracy value and model file size of 0.915 and 6.2 megabytes, respectively. © 2026 Do Manh Hung & Vuong Thi Lan Anh; published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 License.
Classifying; convolutional neural network; deep learning; household waste; R-CNN; YOLOv8