Rapid determination of the texture properties of cooked cereals using near-infrared reflectance spectroscopy

Abstract Rapid determination of the texture properties of various cooked cereals, including the hardness, springiness, stickiness, chewiness, gumminess and cohesiveness, was performed by near-infrared (NIR) diffuse reflectance spectroscopy. Calibration models were established by partial least squares (PLS) or principal component regression (PCR) after derivative and smoothing pretreatments. The coefficients of determination ( R 2 c ) were over 0.94 in PCR models and over 0.91 in PLS models. High performance can also be achieved at selected wavebands 1,000–2,500, 1,122–2,294, 1,243–1,428, 1,728–2,452, 1,125–1,760 and 1,107–2,466 nm, for hardness, springiness, stickiness, chewiness, gumminess and cohesiveness, respectively. Good prediction results were obtained for the texture parameters, which demonstrated that NIR reflectance spectroscopy with common models of PLS or PCR can be used as a rapid tool to determine the texture properties of cooked cereals.