|Ariana, Diwan - MICHIGAN STATE UNIVERSITY|
Submitted to: Applied Engineering in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 19, 2002
Publication Date: September 1, 2002
Citation: LU, R., ARIANA, D. A NEAR-INFRARED SENSING TECHNIQUE FOR MEASURING INTERNAL QUALITY OF APPLE FRUIT. APPLIED ENGINEERING IN AGRICULTURE. v. 18(5). p. 585-590. Interpretive Summary: Poor, inconsistent fruit quality is a major concern to the apple industry. One way to solve this problem is to sort and/or grade fruit for internal quality before they are delivered to the market. Firmness and sugar content are two important attributes in determining the overall quality of apple fruit and technologies that are capable of sorting fruit for these quality attributes will be of great value to the fruit industry. The objective of this research was to investigate a near-infrared sensing (NIR) technique for fast, nondestructive determination of the sugar content and firmness of apple fruit and to study the effect of sensor design on measurement results. Results showed that the sensing technique developed in this study was capable for rapid determination of the sugar content of apple fruit with prediction errors between 0.5% and 0.7% for Empire and Red Delicious apples. Sensor position relative to the light source affected measurement results and better sugar content predictions were obtained wit the sensor placed closer to the light source. NIR measurements were related to fruit firmness, but prediction results were not consistent and good enough for practical applications. This research provides useful information for developing a NIR sensing system for rapid assessment of the sugar content of apple fruit. Such a system should help the fruit industry in providing superior, consistent fruit to the consumer and improve its competitiveness and profitability of producing apples.
Technical Abstract: Near-infrared (NIR) spectroscopy is a promising technique for nondestructive sensing of apple fruit for multiple quality attributes. The objective of this research was to investigate a NIR sensing technique in interactance mode for rapid acquisition of spectral information to predict the sugar content and possibly firmness of apple fruit. A NIR sensing system was developed with a specially designed probe assembly for simultaneous acquisition of spectral data at two different distances from the light source in the spectral region between 900 nm and 1500 nm. Experiments were performed to collect spectral data from two apple cultivars: 'Empire' and 'Red Delicious'. Chemometric models were developed using partial least squares regression. Better predictions of the sugar content of apples were obtained at the sensing position 3.5 mm from the light source than that 5.5 mm away from the light source. The correlations of prediction for sugar content were equal to or greater than 0.81 and prediction errors between 0.5% and 0.7% for the two cultivars. Firmness predictions were not as good as those for sugar content with a prediction error of 10.6 N for Empire. The ratio of spectra from the two sensing positions did not give as good predictions of the fruit sugar content and firmness as those by relative reflectance. This research shows that the proposed NIR sensing technique can be used for rapid determination of fruit sugar content. However, it is still difficult to accurately predict fruit firmness with NIR spectroscopy.