Nondestructive estimation of moisture content, pH and soluble solid contents in intact tomatoes using hyperspectral imaging

Authors: RAHMAN, ANISUR - Chungnam National University; KANDPAL, LALIT - Chungnam National University; LOHOMI, SANTOSH - Chungnam National University; Kim, Moon; LEE, HOONSOO - Forest Service (FS); MO, CHANG-YEUN - Rural Development Administration - Korea; CHO, BYOUNG-KWAN - Chungnam National University

Submitted to: Applied Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/17/2017
Publication Date: 1/17/2017
Citation: Rahman, A., Kandpal, L.M., Lohomi, S., Kim, M.S., Lee, H., Mo, C., Cho, B. 2017. Nondestructive estimation of moisture content, pH and soluble solid contents in in intact tomatoes using hyperspectral imaging. Applied Sciences. 7(1):109. https://doi:10.3390/app7010109.
DOI: https://doi.org/10.3390/app7010109

Interpretive Summary: In this investigation, a near-infrared hyperspectral imaging technique was used to nondestructively evaluate moisture content (MC), pH, and soluble solids content (SSC) for intact tomatoes. Using special processing methods, the hyperspectral data were used to create an image of the tomato. The hyperspectral image data were generally successful in describing the chemical compostion of tomatoes; correlation coefficients of 0.81, 0.69, and 0.74 for MC, pH, and SSC, respectively were obtained. The results revealed that when used with the appropriate analysis, hyperspectral imaging is a promising technology for nondestructive evaluation of chemical components in intact tomatoes. This research provides useful methods of rapidly and nondestructively determining important fruit quality parameters to produce growers and processors.

Technical Abstract: The objective of this study was to develop a nondestructive method to evaluate chemical components such as moisture content (MC), pH, and soluble solids content (SSC) for intact tomatoes by using hyperspectral reflectance imaging in the range of 1000 – 1550 nm. The mean spectra of 95 samples of mature tomatoes were extracted from the hyperspectral images, and multivariate calibration models were built by using partial least squares (PLS) regression with different spectral preprocessing methods. The results showed that the regression model developed by PLS regression for spectra preprocessed using Savitzky–Golay (S-G) 1st derivatives resulted in the better performance for MC and pH prediction, and models developed using spectra preprocessed by smoothing resulted in better prediction of SSC, with correlation coefficients (rpred) of 0.81, 0.69, and 0.74 and root mean square error of prediction (RMSEP) of 0.63%, 0.06, and 0.33% Bx, respectively, compared to models using spectra subjected to other preprocessing methods. The hyperspectral image data in the full wavelength range were used to create chemical images by applying regression coefficients from the best PLS regression model. These results showed that hyperspectral imaging together with an appropriate suitable analysis model is a promising method for the nondestructive prediction of chemical components in intact tomatoes.