|Maghirang, Elizabeth - KSU, MANHATTAN, KS|
Submitted to: Cereal Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: January 3, 2003
Publication Date: May 1, 2003
Citation: Maghirang, E.B., Dowell, F.E. 2003. Hardness measurement of bulk wheat by single-kernel visible and near-infrared reflectance spectroscopy . Cereal Chemistry. 80(3): 316-322. Interpretive Summary: Wheat hardness is a primary quality trait that is related to its milling properties and end-use quality. The current standard measurement techniques for wheat hardness are destructive, i.e., they require grinding or crushing of wheat samples. There is a need for a measurement technique, such as in breeding programs, that is non-destructive, rapid, accurate and that requires small sample sizes. A commercially available single kernel visible and near-infrared reflectance (VisNIR) spectrometer was used to develop a bulk hardness measurement and wheat classification technique. This technique predicted hardness values with 83% accuracy and correctly classified wheat as soft, hard or mixed wheat with 100% accuracy. This technique of using whole kernels has already proven effective for measuring numerous grain attributes such as protein, moisture content, vitreousness, color class, internal insects, and bunt. Thus, instruments being used to measure these attributes can be used for grain hardness measurement and hardness classification. This technology will prove essential as the wheat industry shifts to an end-use oriented market. Likewise, wheat breeding programs are expected to benefit from this technique considering its on-destructive feature, small sample size requirement, accuracy, and speed.
Technical Abstract: Reflectance spectra (400 to 1700 nm) of single wheat kernels collected using the Single Kernel Characterization System (SKCS) 4170 were analyzed for wheat grain hardness using partial least squares (PLS) regression. Average hardness of pure hard, pure soft, and mixed wheat validation samples was predicted (R²=0.85) based on mass-averaged spectra of 30 individual kernels. Additionally, slightly better prediction results were observed when the 550-1690 nm region was used compared to 950-1690 nm region across all sample sizes. For the 30-kernel mass-averaged model, the 550-1690 nm spectra resulted in R²=0.91, SECV=7.70, and RPD= 3.3 while the 950-1690 nm had R²=0.88, SECV=8.67, and RPD=2.9. Thus, including the visible wavelength range improved measurements. The visible and NIR prediction model correctly classified samples as mixed, hard, and soft wheat with 100% accuracy when using the modified classification criteria. Results confirmed the potential of using visible and near-infrared reflectance spectroscopy of whole single kernels of wheat as a rapid and non-destructive measurement of bulk wheat grain hardness and for differentiating between mixed and pure wheat samples.