Detection of sugar adulteration in black tea using multispectral imaging

Abstract

Black tea, valued globally for its flavor, aroma, and nutritional benefits, is a major export commodity for countries such as China, India, and Sri Lanka. Rising demand has led to sugar adulteration to enhance color, twist, and weight, compromising quality and posing health risks, highlighting the need for rapid and reliable verification methods. This study presents a multispectral imaging (MSI) based approach for detecting and quantifying sugar adulteration in black tea. A custom-built system with thirteen narrow band LEDs (365 nm to 940 nm) sequentially illuminated powdered and brewed samples, capturing 26 spectral images in reflectance and transmittance modes, respectively. Spectral features corresponding to sugar induced color changes were extracted independently of natural tea variability. Preprocessing steps, including dark current subtraction, cropping, histogram equalization, and dimensionality reduction via linear discriminant analysis (LDA), ensured high quality data for analysis. Classification was performed using linear discriminant analysis (LDA), K nearest neighbors (K-NN), support vector machine (SVM), feed forward neural network (FFNN), and convolutional neural network (CNN), achieving accuracies above 93% across the tested models. These methods showed high sensitivity in detecting adulteration at levels as low as 5% (w/w) and strong specificity in distinguishing pure from adulterated brewed samples. Polynomial regression was applied to quantify sugar content, yielding R² values above 0.97 for polynomial orders from the first to the fifth. A third order polynomial was selected as it provided a slightly improved fit (R² = 0.9739) while maintaining low model complexity. These results demonstrate that multispectral imaging combined with machine learning enables reliable detection of sugar adulteration and continuous estimation of adulteration levels between 5% and 25%, supporting rapid and non-destructive monitoring of black tea authenticity