Detection of spent turmeric adulteration in powdered Curcuma longa using Vis-NIR spectroscopy and machine learning

Abstract

Adulteration of turmeric powders with spent turmeric is, unfortunately, a common practice in India, and there is an increased demand for in-field analysis during authentication testing for quick decision making. We propose the use of a miniaturized spectrometer and reflectance spectroscopy in the Vis-NIR range along with machine learning for detecting the level of such adulteration. Six different regression methods were used, and linear regression was found to be the best with less than 15% error in prediction. Demonstration of this application paves the way for use of portable Vis-NIR reflectance spectroscopy to inspect the quality of turmeric powder.