Prediction of chemical compositions for on-line quality measurement of red pepper powder using near infrared reflectance Spectroscopy (NIRS)

Applicability of near infrared reflectance spectroscopy (NIRS) was examined for quality control of red pepper powder in milling factories. Prediction of chemical composition was performed using modified partial least square (MPLS) techniques. Analysis of total 51 and 21 red pepper powder samples by conventional method,, for calibration and validation, respectively, revealed standard error of prediction (SEP) and correlation coefficient (R-2) of moisture content. ASTA color value, capsaicinoid content, and total sugar content were 0.55 and 0.90, 8.58 and 0.96. 31.60 and 0,65, and 1,82 and 0.86, respectively, SEP and R-2 were low and high. respectively. except for capsaicinoid content, The results indicate with slight improvement, on-line quality measurement of red pepper powder with NIRS Could be applied in red pepper milling factories.