Identification and Classification of Sudan Dye I Adulterants in Turmeric Powder by Nir Spectroscopy and Support Vector Machine

摘要

The main objective of this study is to estimate the potential of near infrared (NIR) spectroscopy as a rapid and non-destructive tool for the detection as well as quantification of sudan dye I adulterated turmeric powder samples. Pure turmeric powder was prepared at the laboratory and spiked with Sudan dye I and investigated accordingly. The concentrations of the adulterants were 0.05%, 0.1%, 0.2%, 0.5%, 1%, 1.5%, 2%, 5%, 10%, 15%, 20%, 25% and 30% (w/w), respectively. Exploratory data analysis was done for the visualization of the adulterant classes by principal component analysis (PCA). In the classification approach, principal components (PCs) extracted by PCA were fed as the inputs of the support vector machine (SVM) classifier. The average accuracy of the adulterants classes noted was greater than 90%.