Click on the underlined title to access the document or go back to the Search Results screen to download the PDF version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Onion quality assessment using diffuse reflectance hyperspectral images with a shape correction algorithm
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.orgCitation: 2011 Louisville, Kentucky, August 7-10, 2011 1110708.
Authors: Haihua Wang, Changying Li
Keywords: hyperspectral imaging, onion, postharvest, internal quality, shape correction, diffuse reflectance
Onion internal quality attributes such as dry matter content, soluble solid content, and firmness are important for onion processors and consumers. Hyperspectral imaging (HSI) combines conventional imaging with spectroscopy, and thus could be a powerful tool for non-destruction evaluation of the internal quality of onions. However, the elliptical shape of the onion can result in uneven reflection at different positions on onion surface. In this study, diffuse reflectance images were acquired by a line scan hyperspectal imaging system for onion internal quality evaluation after incorporation an elliptical shape correction algorithm. Four steps were followed: 1) Color images were acquired to retrieve geometric information of the onion; 2) Shape correction algorithm was performed on the diffuse reflectance hyperspectral image; 3) Spectral characteristics were extracted from the Region of interests (ROIs) in the corrected images; 4) The partial least square (PLS) was used to build the correlation model between the internal quality attributes and spectral data. Validation data showed that the coefficients of determination of the PLS models were 0.50, 0.79, and 0.80 for firmness, soluable solid content, and dry matter content, respectively. This shape correction algorithm could be used for other fruits and vegetables with the elliptical shape.