Year: 2025
Type: article
Abstract: Fluorescence sensing technology has been extensively used in the field of explosives monitoring for highly sensitive trace detection and early warning systems. However, these advancements have been hindered by a significant reliance on professional technicians and a low degree of automation. Here, we present a cloud-based website that encapsulates a deep-learning-assisted technique based on the PaddlePaddle You Only Look Once (PPYOLO) model and integrates a highly sensitive external fluorescence sensor. This approach, combined with a deep learning model, streamlines the detection of nitro explosives, enhancing automation levels and efficiency. The cloud-based interface and self-developed portable sampling software offer a user-friendly, interactive, and portable system. Moreover, the model boasts a target recognition accuracy (bounding box mean average precision [bbox_map]) of up to 99%, making it capable of automatically extracting six independent color signals from images and providing detection results. These results suggest that the proposed method could provide promising perspectives for the future development of efficient and portable explosives detection systems.
Source: Cell Reports Physical Science
Authors: Jingyi Xiong, Jiawei Zhang, Xufan Zheng, Tianyu Hu, Huimin Xiang, Yuda Li, Biao Xiao, Xunchang Wang, Renqiang Yang
Institutions: Jianghan University, Wuhan Institute of Technology
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $5,620
OpenAlex ID: W4412087212