Skip to main content
ARS Home » Research » Publications at this Location » Publication #166390


item Dowell, Floyd
item Armstrong, Paul

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 9/19/2004
Publication Date: 9/19/2004
Citation: Dowell, F.E., Pearson, T.C., Armstrong, P.R. 2004. High throughput grain quality analysis. Meeting Abstract.

Interpretive Summary:

Technical Abstract: Many grain quality attributes are not uniformly distributed within the sample or bulk lot. For example, aflatoxin or fumonisin in grain may be present in only a small percentage of kernels. Our research program concentrates on developing technology or procedures to rapidly detect specific grain quality characteristics, and then sort the sample or bulk lot based on that measurement. We are working with automated single kernel systems that utilize acoustics, and color and near-infrared (NIR) sensors that can detect specific attributes and sort kernels at rates of 1 to 1000 kernels/s. The acoustical system is being investigated for detecting insects inside single wheat kernels. The color and NIR sensors are being used to detect and remove: aflatoxin and fumonisin from corn; fumonisin from wheat; red from white wheat, red from white millet, low from high protein wheat, and damaged from undamaged corn for developing new cultivars; soft from hard wheat for studying the affects of hardness on bread quality; Karnal bunt from wheat for routine inspection; defective soybeans with off-flavors from good beans; and insect infested wheat kernels from undamaged kernels. The lower-speed detection and sorting is based on technology developed within our research unit and commercialized by Perten Instruments, Springfield, Illinois. The higher-speed detection and sorting is based on electronic sorting technology developed by Satake USA Inc, Houston, Texas. This technology can help regulatory agencies rapidly screen samples; breeders develop cultivars with specific end-use traits; and researchers study specific intrinsic kernel characteristics.