Adulteration quantification of cheap honey in high-quality Manuka honey by two-dimensional correlation spectroscopy combined with deep learning

Abstract

The adulteration of cheap honey in high value honey has a negative impact on consumer health, and its detection is also a challenging task. A fast, non-invasive and cost-effective spectroscopic strategy was proposed to quantify the honey fraud based on the two-dimensional correlation spectroscopy (2D-COS) of Raman spectra combined with multiple deep learning techniques. 700 Raman spectra of Manuka, acacia and multi-floral honeys were collected, and the corresponding synchronous, asynchronous and integrative correlation spectra were obtained. The t-distributed stochastic neighbor embedding (t-SNE) and partial least square regression (PLSR) were used to analyze one-dimensional spectra and 2D-COS image datasets of the same sample, demonstrating that 2D-COS can highlight the complex fingerprint features of samples and thus contribute to the spectral characterization. The combination of synchronous 2D-COS and deep residual shrinkage networks (DRSN) achieved the best performance compared to the other models, with the root mean square errors of prediction (RMSEP) of 3.1166 for Manuka honey and 2.3188 for acacia honey, respectively. The impressive performance of the proposed spectral method is sufficient to quantify the adulteration of cheap honey. This study provides a novel alternative for relevant detection tasks and facilitates to ascertain the authenticity of high-value honey products.