| Title |
Detection of Illicit Drugs with Multi-sensor E-nose using Multi-branch Convolutional Neural Networks |
| Authors |
이호중(Hojung Lee) ; 박흰돌(Hwin Dol Park) ; 김도현(Do Hyeun Kim) ; 최재훈(Jae Hun Choi) ; 이종석(Jong-Seok Lee) |
| DOI |
https://doi.org/10.5573/ieie.2024.61.10.159 |
| Keywords |
Detection; Illicit drugs; Convolutional neural network; Multi-branch; E-nose system |
| Abstract |
Detecting drug trafficking is a significant challenge, and its resolution through deep learning has been limited due to the lack of available drug datasets. In this paper, we present a new deep learning method to classify drug gas collected by a multi-sensor e-nose system. We propose a multi-branch convolutional neural network, where each branch processes the data from each flow rate of gas, and the results from the branches are combined to produce the final classification output. Our approach demonstrates a great improvement in accuracy when compared to traditional machine learning algorithms. |