Estimating percentages of fusarium-damaged kernels in hard wheat by near-infrared hyperspectral imaging

Authors: Delwiche, Stephen - Steve; TORRES RODRIGUEZ, IRINA - UNIVERSITY OF CORDOBA; Rausch, Steven - Steve; GRAYBOSCH, ROBERT

Submitted to: Journal of Cereal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/22/2019
Publication Date: 2/28/2019
Citation: Delwiche, S.R., Torres Rodriguez, I., Rausch, S.R., Graybosch, R.A. 2019. Estimating percentages of fusarium-damaged kernels in hard wheat by near-infrared hyperspectral imaging. Journal of Cereal Science. 87:18-24.

Interpretive Summary: Wheat scab, formally known as fusarium head blight (FHB), is a fungal disease that is prevalent around the world. Not only does the disease cause a reduction in yield, it also can result in the production of the toxin deoxynivalenol or vomitoxin, which is harmful when consumed by humans or livestock. Plant breeders are continually evaluating the hardiness of new lines against FHB as do official grain inspectors when assigning grades to wheat lots. Visual analysis is the traditional method for estimating percentages of fusarium-damaged kernels. This procedure is tedious and subjective as it relies on the consistent performance of trained inspectors. As a substitute, an objective instrument system was developed that utilizes the combination of digital imaging and near-infrared spectroscopy. Three or four wavelengths of radiation, slightly out of range of human eye sensitivity, were identified as best in differentiating sound and scabby kernels. The accuracy is equivalent to visual analysis. Breeders and inspectors will benefit from this study. While not the focus of the study, a commercial sorting instrument utilizing the identified wavelengths could be developed for use in grain elevators, terminals, and mills.

Technical Abstract: Fusarium head blight (FHB) is among the most common fungal diseases affecting wheat. This preharvest disease causes decreased yield, low-density kernels and, most concerning, the potential for occurrence of the mycotoxin deoxynivalenol, a compound toxic to humans and livestock. Human visual analysis of representative wheat samples has been the traditional method for FHB assessment in both official inspection and plant breeding operations. While not requiring specialized equipment, visual analysis is dependent on a trained and consistent workforce, such that in the absence of these aspects, biases may arise among inspectors and evaluation dates. This research was intended to avoid this level of fine tuning by using longer wavelength radiation than the visible using hyperspectral imaging (HSI) on individual kernels. Linear discriminant analysis (LDA) models to differentiate between sound and scab-damaged kernels were developed based on mean of reflectance values of the interior pixels of each kernel at four wavelengths (1100, 1197, 1308, and 1394 nm). Other input variables for LDA were examined, including kernel morphological properties and histogram features from the pixel responses of selected wavelengths of each kernel, with the mean response at the selected wavelengths deemed superior. The benefit of this finding is that future spectral imaging hardware may be based on multi-spectral instead of hyperspectral design, which translates into lower cost. The results of the classification modeling indicate the strong potential of HSI in determining fusarium-damaged kernels. At the same time, however, improvement in aligning this procedure to visual analysis is hampered by the inherent level of subjectivity in visual analysis.