NIR Spektroskopisi Kullanılarak Bütün Haldeki Newton Domates Çeşidinde C Vitamini ve Likopen İçeriğinin Tahribatsız Tespiti

Abstract

Lycopene and vitamin C are two vital compositions of tomatoes. Analytical quantification of these components using common destructive methods is expensive and time consuming. In this study, the feasibility of using near-infrared (NIR) spectroscopy at the range of 930-1650nm was assessed to measure the lycopene and vitamin C contents of intact cv. Newton tomatoes. To develop prediction models based on spectral data and analytical measurements achieved using common destructive methods, partial least squares (PLS) regression was utilized. Before modeling, some pre-processing techniques were also used to reduce the irrelevant data from the spectra. The prediction accuracy of the models was evaluated using root mean square error of cross validation (RMSECV), correlation coefficient (rcv) and residual predictive deviation (RPD). The best prediction models had RMSECV of 2.256 µg g-1 and 1.087 mg 100g-1 , r cv of 0.840 and 0.818, and RPD of 1.835 and 1.701 for lycopene and vitamin C contents, respectively. Results indicated that NIR spectroscopy can predict the lycopene and vitamin C contents of tomato non-destructively with good and fair accuracy, respectively.