BBP Mektan BBP Mektan - Mekanisasi Pertanian

(Application of Visible and Near Infrared Spectroscopy for Nondestructive Analysis of Tomato Quality)

Suparlan 1) and Kazuhiko ITOH 2)
1) Indonesian Center for Agricultural Engineering Research and Development (ICAERD),
2) Graduate School of Agricultural Science, Hokkaido University, Kita 9, Nishi 9, Kita-ku, Sapporo 060-8589, Japan.


ABSTRACT

Near infrared spectroscopy has been used as a rapid and nondestructive method for determining the quality of fruit and vegetable. Calibration models of Visible Near Infrared (Vis-NIR) spectroscopy for color and soluble solids content (SSC) of fresh intact tomatoes was developed by partial least squares (PLS) regression analysis with a full cross validation method using 150 samples. The calibration model of color was developed using the wavelength ranged from 600 to 730 nm, and that of SSC was developed using the wavelength ranged from 700 to 1098 nm. Correlation coefficient of calibration (Rc2) and standard error of calibration (SEC) for color were 0.97 and 0.15, respectively, and those for SSC were 0.96 and 0.42 oBrix, respectively. The models were then used to predict color and SSC of an independent data set consist of 104 samples. The correlation coefficient of prediction (Rp2), the standard error of prediction (SEP), and the bias for prediction of color were 0.92, 0.19, and 0.09, respectively, and those for prediction of SSC were 0.91, 0.49 oBrix, and –0.04 oBrix, respectively. It was concluded that Vis-NIR spectroscopy interfaced with fiber optic is suitable method to determine color and SSC of fresh intact tomatoes. Thus, Vis-NIR technology could be used for classifying fresh intact tomatoes into qualitative groups based on both external and internal quality.

Keywords: nondestructive analysis; near-infrared spectroscopy; partial least squares regression; tomatoes quality;