|Haley, Scott - COLORADO STATE UNIV|
Submitted to: Sensing and Instrumentation for Food Quality and Safety
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: September 12, 2008
Publication Date: October 3, 2008
Repository URL: http://www.ars.usda.gov/SP2UserFiles/Place/54300520/406Imagesorter-redwhitewheat.pdf
Citation: Pearson, T.C., Brabec, D.L., Haley, S. 2008. Color image based sorter for separating red and white wheat. Sensing and Instrumentation for Food Quality and Safety. 2:280-288. Interpretive Summary: A low cost sorting device for wheat was built using a standard personal computer and color camera. Programming techniques used in computer video games were used so that the throughput of the sorter would be high while keeping the sorter cost low. The sorting system was tested on its ability to separate red wheat from white wheat for wheat breeding programs. At a wheat throughput of 30 kernels per second, or 3.5 Kg per hour, the sorter is able to correctly separate 95% to 99% of the wheat. The accuracy is 15 to 20% higher than what can be achieved with traditional sorters. This sorter will help breeding programs isolate desirable kernels so that they can be propagated, which will results in faster releases of new and improved varieties of grain.
Technical Abstract: A simple imaging system was developed to inspect and sort wheat samples and other grains at moderate feed-rates (30 kernels/s or 3.5 kg/h). A single camera captured color images of three sides of each kernel by using mirrors, and the images were processed using a personal computer (PC). The camera transferred the images to the PC in real time via an IEEE 1394 cable, using DirectX application software and use of a dual-core computer processor. Image acquisition and transfer to the PC required approximately 17 ms per kernel, and an additional 1.5 ms was required for image processing. After classification, the computer could output a signal from the parallel port to activate an air valve to divert (sort) kernels into a secondary container. Hard red and hard white wheat kernels were used in this study. Simple image statistics and histograms were used as features. Discriminant analysis was performed with one, two, or three features to demonstrate classification improvements with increased numbers of features. The sorter was able to separate hard red kernels from hard white kernels with 95% to 99% accuracy, depending on the wheat varieties, feed-rate, and number of classification features. The system is an economical and useful instrument for sorting wheat and other grains with high accuracy.