<![CDATA[An important factor in choosing a food product is its quality and safety. Meanwhile, visual aspects are a benchmark for product acceptance. Dragon fruit peel powder (DFP) has excellent potential as a natural food coloring. This study aims to detect adulteration in dragon fruit peel powder using two spectroscopy techniques: Visible-Near Infrared (Vis-NIR) and Shortwave-Near Infrared (SW-NIR) spectroscopy. The adulterants include purple sweet potato flour (PP), erythrosine dye powder (ER), and remazol textile dye powder (TX) with varying concentrations of 0%, 0.5%, 1%, 5%, 10%, 20%, 30%, 40%, 50%, and 100%. Partial least squares regression (PLSR) with ten spectral preprocessing methods was used to analyze data and assess model performance. The results show that combining of spectroscopy with the PLSR model significantly improves accuracy, achieving R²P values above 0.92 for all adulterants. These findings highlight Vis-NIR and SW-NIR spectroscopy combined with PLSR modeling, as rapid, non-destructive tools. Vis-NIR spectroscopy proved superior to SW-NIR spectroscopy in detecting food colorant adulteration because of its sensitivity to color pigments.]]>
Copyrights © 2026
