Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: TECHNOLOGIES FOR ASSESSING AND GRADING QUALITY AND CONDITION OF CUCUMBERS AND TREE FRUITS

Location: Sugarbeet and Bean Research

Title: Prediction of olive quality using FT-NIR spectroscopy in reflectance and transmittance modes

Authors
item Kavdir, Ismail - CANAKKALE OM UNIV-TURKEY
item Burukcan, M Burak - CANAKKALE OM UNIV-TURKEY
item Lu, Renfu
item Kocabiyik, Habib - CANAKKALE OM UNIV-TURKEY
item Seker, Murat - CANAKKALE OM UNIV-TURKEY

Submitted to: Biosystems Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 16, 2009
Publication Date: July 1, 2009
Citation: Kavdir, I., Burukcan, M., Lu, R., Kocabiyik, H., Seker, M. 2009. Prediction of Olive Quality Using FT-NIR Spectroscopy in Reflectance and Transmittance Modes. Biosystems Engineering. 103(3):304-312.

Interpretive Summary: Freshly harvested olives that are delivered to processing plants in the bins tend to have large variations in maturity, oil content, firmness and color. Unripe olives would cause problems in the fermentation process as these olives have poor skin permeability and low flesh to stone ratio. Overripe olives, on the other hand, become soft quickly. When the same processing procedure is applied to the olives with different properties, it tends to produce inferior final products such as low quality-grade olive oil and table olives. Hence, sorting olives for inner and surface quality attributes prior to processing is needed to assure high quality final products. In this research, Fourier-transform near-infrared (FT-NIR) spectroscopy was used to acquire spectral data from freshly harvest olives of 'Ayvalik' and 'Gemlik' varieties for the near-infrared (longer than visible light) region between 780 and 2500 nm in reflectance mode and between 800-1725 nm in transmittance mode. After the spectral measurements, individual olives were evaluated for firmness, oil content and color using standard methods. Mathematical models were developed and validated for predicting olive firmness, oil content and fruit color. Good firmness predictions were obtained for both olive varieties in transmittance with the coefficient of determination (R**2) of 0.77. FT-NIR spectroscopy also gave relatively good prediction of the oil content for the pooled data of the two varieties (R**2=0.64 for reflectance and 0.61 for transmittance). Superior color predictions were obtained with both reflectance and transmittance modes (R**2 values were between 0.82 and 0.92). This research demonstrated that FT-NIR spectroscopy is useful for assessing surface and internal quality attributes of olives. The technique has potential for sorting and grading olives for these quality attributes. This could, in turn, greatly improve the quality of final olive products, thus benefitting both growers/processors and consumers.

Technical Abstract: The objective of this research was to use FT-NIR spectroscopy to predict the firmness, oil content and color of two olive (Olea europaea L) varieties (‘Ayvalik’ and ‘Gemlik’). Spectral measurements were performed on the intact olives for the wavelengths of 780-2500 nm in reflectance and for 800-1725 nm in transmittance. After the spectral measurements, olive firmness, oil content and color were evaluated using standard methods. Calibration models for prediction of the olive quality attributes were developed using the partial least squares method, and they were validated by leave-one-out cross validation. Better firmness prediction results for Magness-Taylor (MT) maximum force were obtained for both varieties in transmission mode, with the coefficient of determination (R**2) of 0.77 and the root mean squared error for cross validation (RMSECV) of 1.36 N for ‘Ayvalik’. Reflectance mode, on the other hand, had a lower R**2 value of 0.65 (RMSCV=1.82 N) for ‘Ayvalik’. Similar firmness prediction results were obtained for ‘Gemlik’ olives. Oil content prediction for each olive variety was poor, due to the relatively homogenous samples. However, better oil content prediction was achieved for the pooled data with the R**2 value of 0.64 (RMSECV=0.05%) in reflectance and 0.61 (RMSECV=0.05%) in transmittance. Both FT-NIR reflectance and transmittance measurements gave good prediction of olive color, with the R**2 values for chroma ranging between 0.83 and 0.88 in reflectance and between 0.85 and 0.92 in transmittance. Similar results for hue prediction were also obtained. These research results demonstrated that FT-NIR spectroscopy is potentially useful for assessing internal and external quality attribute of olives.

Last Modified: 4/18/2014
Footer Content Back to Top of Page