Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Environmental Microbial & Food Safety Laboratory » Research » Publications at this Location » Publication #326338
Title: Non-destructive evaluation of bacteria-infected watermelon seeds using Vis/NIR hyperspectral imaging

Author
item LEE, HOONSOO - Forest Service (FS)
item Kim, Moon
item SONG, YU-RIM - Chungnam National University
item OH, CHANGSIK - Kyung Hee University
item LIM, HYUN-SUB - Chungnam National University
item KANG, SUN-SOON - Chungnam National University
item CHO, BYUNG-KWAN - Chungnam National University

Submitted to: Journal of the Science of Food and Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/28/2016
Publication Date: 6/6/2016
Citation: Lee, H., Kim, M.S., Song, Y., Oh, C., Lim, H., Kang, S., Cho, B. 2016. Non-destructive evaluation of bacteria-infected watermelon seeds using Vis/NIR hyperspectral imaging. Journal of the Science of Food and Agriculture. doi: 10.1002/jsfa 7832.

Interpretive Summary: The objective of this study was to evaluate the potential of hyperspectral imaging for detecting bacterium-infected watermelon seeds. A hyperspectral reflectance imaging system (spectral region of 400–1000 nm) was constructed to obtain hyperspectral reflectance images for 336 bacteria-infected watermelon seeds which consequently subjected to multivariate analysis algorithms to classify bacteria-infected watermelon seeds from healthy watermelon seeds. The developed algorithms detected bacteria-infected watermelon seeds with accuracy above 91%, suggesting that Vis/NIR hyperspectral imaging system is effective for quarantining bacteria-infected watermelon seeds. This research provides nondestructive methods to rapidly assess produce seed viabilities and the information is beneficial to produce growers and seed industries.

Technical Abstract: It is needed to minimize the economic loss by sorting infected seeds from healthy seeds before seeding. However, current methods of detecting infection seeds such as seedling grow-out, enzyme-linked immunosorbent assay (ELISA), polymerase chain reaction (PCR), and real-time polymerase chain reaction (RT-PCR) have a critical drawback that are time-consuming, labor-intensive, and destructive procedures. The objective of this study was to evaluate the potential of Vis/NIR hyperspectral imaging system for detecting bacteria-infected watermelon seeds. A hyperspectral Vis/NIR reflectance imaging system (spectral region of 400–1000 nm) was constructed to obtain hyperspectral reflectance images for 336 bacteria-infected watermelon seeds which consequently subjected to partial least square discriminant analysis (PLS-DA) and least-squares support vector machine (LS-SVM) to classify bacteria-infected watermelon seeds from healthy watermelon seeds. The developed system detected bacteria-infected watermelon seeds with accuracy above 90% (PLS-DA: 91.7%, LS-SVM: 90.5%), suggesting that Vis/NIR hyperspectral imaging system is effective for quarantining bacteria-infected watermelon seeds.