Research Article
A Trial of Recognition of Electronic Parts by Deep-Learning for Efficient Recycling
@INPROCEEDINGS{10.4108/eai.18-12-2025.2365288, author={Takuto SHIRAISHI and Yihong TANG and Qi LI and Tomonori IZUMI}, title={A Trial of Recognition of Electronic Parts by Deep-Learning for Efficient Recycling}, proceedings={Proceedings of the 13th International Conference on Identification, Information and Knowledge in the Internet of Things, IIKI 2025, 18-21 December 2025, Chengdu, China}, publisher={EAI}, proceedings_a={IIKI}, year={2026}, month={6}, keywords={electronic wastes recycling deep learning CNN resource paradox SDGs}, doi={10.4108/eai.18-12-2025.2365288} }- Takuto SHIRAISHI
Yihong TANG
Qi LI
Tomonori IZUMI
Year: 2026
A Trial of Recognition of Electronic Parts by Deep-Learning for Efficient Recycling
IIKI
EAI
DOI: 10.4108/eai.18-12-2025.2365288
Abstract
Waste electronic appliances contain a large amount of recyclable materials, but identifying and separating them still requires significant manual labor. To improve material recycling efficiency, we develop a system that utilizes deep learning to recognize and analyze electronic components automatically. Since the inference speed of deep learning models can become a bottleneck in real-time recycling systems, we propose a lightweight neural network specifically designed for the classification of electronic components on wasted electronic boards. Experimental results show that the proposed model achieves approximately 95% accuracy while requiring only about 120 µsec to classify a single component image.
Copyright © 2025–2026 EAI
