Geographical origin authentication of instant coffee from southern Bahia using MIR and NIR spectroscopy coupled with DD-SIMCA

Abstract

This study presents an innovative approach for authenticating the geographical origin of instant coffee from Southern Bahia using vibrational spectral data coupled with Data-Driven Soft Independent Modeling of Class Analogy (DD-SIMCA). Combining data from a benchtop mid-infrared (MIR) spectrometer and a portable near-infrared (NIR) spectrometer, the methodology achieved 100 % efficiency with Multiplicative Scatter Correction (MSC) and Standard Normal Variate (SNV) pre-processing, alongside offset correction for MIR data. MIR spectroscopy enabled detailed chemical composition analysis, while the portable NIR spectrometer provided cost-effective in situ capabilities. These rapid, non-destructive techniques align with Green Chemistry principles, supporting environmentally sustainable practices. By ensuring authenticity, safeguarding product value, and enhancing consumer trust, this strategy addresses potential fraud related to geographical origin. With its exceptional performance and practical advantages, the proposed methodology offers a robust solution for quality control in the coffee industry and holds potential for broader applications in related fields.

Keywords: Coffea arabica; Coffea canephora; Food authenticity; Food fraud; One-class classification; Pattern recognition.