Non-destructive quality prediction of domestic, commercial red pepper powder using hyperspectral imaging

We analyzed the major quality characteristics of red pepper powders from various regions and predicted these nondestructively using shortwave infrared hyperspectral imaging (HSI) technology. conducted partial least squares regression analysis on 70% (n=71) acquired data to examine characteristics. Rc2 values >0.8 were obtained for ASTA color value (0.9263) capsaicinoid content (0.8310). The developed prediction model was validated remaining 30% (n=35) data; highest accuracy achieved (Rp2=0.8488), similar validity levels moisture contents. To increase model, we spectrum preprocessing SNV, MSC, SG-1, SG-2, model’s verified. results indicated that most significantly improved by MSC method, all methods. Our findings suggest powders, even do not conform specific variables such as particle size content, can be via HSI.