Nondestructive Estimation of Moisture Content, pH and Soluble Solid Contents in Intact Tomatoes Using Hyperspectral Imaging

Abstract

Abstract:The objective of this study was to develop a nondestructive method to evaluate chemical components such as moisture content (MC), pH, and soluble solid content (SSC) in intact tomatoes by using hyperspectral imaging in the range of 1000-1550 nm. The mean spectra of the 95 matured tomato samples were extracted from the hyperspectral images, and multivariate calibration models were built by using partial least squares (PLS) regression with different preprocessing spectra. The results showed that the regression model developed by PLS regression based on Savitzky-Golay (S-G) first-derivative preprocessed spectra resulted in better performance for MC, pH, and the smoothing preprocessed spectra-based model resulted in better performance for SSC in intact tomatoes compared to models developed by other preprocessing methods, with correlation coefficients (r pred ) of 0.81, 0.69, and 0.74 with root mean square error of prediction (RMSEP) of 0.63%, 0.06, and 0.33% Brix respectively. The full wavelengths were used to create chemical images by applying regression coefficients resulting from the best PLS regression model. These results obtained from this study clearly revealed that hyperspectral imaging, together with suitable analysis model, is a promising technology for the nondestructive prediction of chemical components in intact tomatoes.

Hyperspectral imaging for predicting the allicin and soluble solid content of garlic with variable selection algorithms and chemometric models

“…By contrast, the images were binned during acquisition in the spatial direction to provide images with spatial dimensions of 384 × 500 pixels with 275 spectral bands from 1000 to 1700 nm. To minimize the effects of the camera dark current, the acquired images ( R 0 ) were corrected using white and dark reference images according to the equation: where R is the corrected image, D is the dark current image (< 0% reflectance) obtained with the light source turned off and the camera lens completely covered with its own non‐reflective opaque black cap to remove the thermal activities of the CCD detector, and W is the white reference image obtained using a standard white Teflon tile (>99% reflectance).…”

“…By contrast, the images were binned during acquisition in the spatial direction to provide images with spatial dimensions of 384 × 500 pixels with 275 spectral bands from 1000 to 1700 nm. To minimize the effects of the camera dark current, the acquired images (R 0 ) were corrected using white and dark reference images according to the equation 8 :…”

“…36,37 The derivatives (particularly the firstand second-order derivatives) were intensively used to remove baseline shifts and to improve resolution. 8,34,38,39 After preprocessing, the samples were divided into a calibration set (two-thirds of all samples) and a prediction set (one-third of all samples) using the Kennard-Stone sampling algorithm. 40 In this study, a k-fold cross-validation was performed on the calibration set to avoid overfitting with the k-fold value set to 20.…”

“…Recently, hyperspectral imaging (HSI) has gained prominence as a promising and emerging tool that can rapidly and non-destructively sense the spectral and spatial information of fruits and vegetables and determine their quality. [6][7][8][9] A hyperspectral image, known as a hypercube, contains a three-dimensional dataset with two spatial dimensions and one wavelength dimension. 7 The most common spectral ranges are 400-1000 nm [visible-near-infrared (Vis-NIR)] and 900-2500 nm [shortwave-infrared (SW-IR)], which comprise the HSI system.…”

“…[25][26][27] At the same time, full spectral regions may include wavelengths that contribute more collinearity, redundancy and noise than relevant information. 27 Recently, extensive studies have been conducted using different variable selection methods, such as regression coefficients, 8,16,21,28,29 variable importance in projection (VIP) [28][29][30][31] and the successive projections algorithm (SPA) 14,16,27,32,33 in pursuit of a robust calibration model for the quantitative or qualitative analysis of fruits and vegetables. However, to the best of our knowledge, no reports address the prediction of allicin and SSC in the quality assessment of garlic using the HSI technique in combination with different variable selection methods and robust calibration models.…”