Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Publications at this Location » Publication #316179
Title: Predict compositions and mechanical properties of sugar beet using hyperspectral scattering

Author
item PAN, LEIQING - Nanjing Agricultural University
item Lu, Renfu
item ZHU, QIBING - Jiangnan University
item TU, KANG - Nanjing Agricultural University
item CEN, HAIYAN - Michigan State University

Submitted to: Food and Bioprocess Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/28/2016
Publication Date: 6/16/2016
Citation: Pan, L., Lu, R., Zhu, Q., Tu, K., Cen, H. 2016. Predict compositions and mechanical properties of sugar beet using hyperspectral scattering. Food and Bioprocess Technology. 9(7):1177-1186.

Interpretive Summary: The sucrose yield and processing quality of sugar beet is largely determined by sucrose, soluble solids and moisture contents as well as the tissue mechanical properties. Rapid measurement of these quality parameters is valuable in breeding, production and postharvest processing of sugar beet. Hyperspectral scattering technique, which was recently developed by the USDA postharvest engineering laboratory at East Lansing, Michigan, provides a new means for quality assessment of food and agricultural products through the measurement of light scattering and absorption properties. This research was aimed at predicting the sucrose, soluble solids and moisture contents and compressive mechanical properties of beets, using hyperspectral scattering imaging technique. Hyperspectral scattering images for the wavelengths of 500-1,000 nm were collected from slices of 398 beet samples, followed with standard laboratory tests to measure the multiple quality parameters of the beets. Mathematical prediction models for the quality parameters were developed from the scattering data using two processing methods, i.e., full spectrum partial least squares and uninformative variable elimination partial least squares. Good predictions for moisture, sucrose and soluble solids contents were obtained by using both methods, with the correlation coefficients of 0.75-0.88. However, much lower predictions were obtained for the mechanical properties with the correlation coefficients of 0.46-0.63. Hyperspectral scattering is a promising technique for rapid quality evaluation of sugar beet. However, improvements in the sample measurement procedure and data processing are needed for more accurate prediction of quality for sugar beet.

Technical Abstract: Sucrose, soluble solids, and moisture content and mechanical properties are important quality/property attributes of sugar beet. In this study, hyperspectral scattering images for the spectral region of 500-1,000 nm were acquired from 398 beet slices, from which relative mean spectra were calculated. Prediction models for these quality attributes were developed using partial least squares regression for both full spectra and selected wavelengths. The results showed that using relative mean spectra gave good predictions for the moisture, soluble solids and sucrose content of beet slices with the correlations of 0.75-0.88 and the standard errors of prediction of 0.95-1.08 based on full-spectrum partial least squares regression (PLSR) models. PLSR models using wavelengths selection with the uninformative variable elimination (UVE) method produced similar prediction accuracy. However, both modeling approaches gave poor predictions for the mechanical properties of beets with the correlation values of 0.46-0.63. The research demonstrated the potential of hyperspectral scattering imaging for measuring quality attributes of sugar beet.