Nondestructive Firmness Estimation of Tomato Fruit Using Near-Infrared Spectroscopy

Today, nondestructive methods are widely used to determine the quality of agricultural products. Meanwhile, visible and near-infrared (Vis/NIR) spectroscopy is regarded as one of the most widely used methods in the field of quality assessment of agricultural products. In this study, a system was dev...

Full description

Bibliographic Details
Main Authors: M. Hagh-shenas Adarmonabadi, S. A. Mireei, M. Sadeghi, M. Nazeri
Format: Article
Language:fas
Published: Isfahan University of Technology 2019-11-01
Series:Tulīd va Farāvarī-i Maḥṣūlāt-i Zirā̒ī va Bāghī
Subjects:
Online Access:http://jcpp.iut.ac.ir/article-1-2745-en.html
Description
Summary:Today, nondestructive methods are widely used to determine the quality of agricultural products. Meanwhile, visible and near-infrared (Vis/NIR) spectroscopy is regarded as one of the most widely used methods in the field of quality assessment of agricultural products. In this study, a system was developed to measure the Vis/NIR spectra of tomato fruit samples in the half-transmittance mode of measurement. Halogen lamps were used as the sources of irradiation and a portable fiber optic spectrometer was utilized to collect the spectra in the range of 400-100 nm. Immediately after collecting the spectra of 170 tomato samples, harvested at different stages of ripening, the firmness of the samples was measured using the standard Magness-Taylor penetration test. In order to eliminate the effects of irregular factors, different preprocessing methods were applied to the collected spectra. Then, partial least squares (PLS) regression was implemented to develop the firmness predictive models. Preliminary results showed two absorption peaks around 670 and 990 nm in the acquired spectra that were related to the chlorophyll and water contents of tomato, respectively. The spectroscopy method could nondestructively predict the tomato fruit firmness with a correlation coefficient (rp) of 0.920 and the root mean squares error of prediction (RMSEP) of 2.5 nm.
ISSN:2251-8517