Rapid recognition of different sources of methamphetamine drugs based on hand-held near infrared spectroscopy and multi-layer-extreme learning machine algorithms

Abstract

The rapid recognition of the sources of the drugs can provide valuable clues and provide the basis for determining the nature of a drug case. Here, a novel recognition method was put forward to identify the source of methamphetamine drugs rapidly and non-destructively by using a hand-held near infrared (NIR) spectrometer and a multi-layer-extreme learning machine (ML-ELM) algorithm. The accuracy, precision, sensitivity, and F-score were higher with the proposed ML-ELM algorithm than in traditional linear discriminant analysis (LDA), extreme learning machine (ELM) classification, and partial least squares (PLS) regression algorithms. The prediction accuracy of ML-ELM algorithm is 25.0%, 15.3% and 18.1% higher than that of LDA, ELM and PLS regression, respectively. The ML-ELM models for recognizing the different sources of methamphetamine drugs had the best generalization ability and prediction results. The experimental results indicated that the combination of hand-held NIR technology and ML-ELM algorithm can recognize the different sources of methamphetamine drugs rapidly, accurately, and non-destructively.