High-Speed Sorting of Grains by Color and Surface Texture

Abstract

A high-speed, low-cost, image-based sorting device was developed to detect and separate grains with different colors/textures. The device directly combines a complementary metal-oxide-semiconductor (CMOS) color image sensor with a field-programmable gate array (FPGA) that was programmed to execute image processing in real-time without the need for an external computer. The spatial resolution of the imaging system is approximately 16 pixels/mm. The system utilizes the inherent parallel processing capabilities of FPGA's to inspect three separate streams of grain with a single camera/FPGA combination. Kernels are imaged immediately after dropping off the end of a chute and are diverted by activating an air valve. The system has a throughput rate of approximately 225 kernels/s overall, which is much higher than previously developed image inspection systems. This throughput rate corresponds to an inspection rate of approximately 25 kg/h of wheat. Testing of the system resulted in accuracies of 96% for separating red wheat from white wheat, 93% accuracy for separating barley from durum, and 92% for separating brown flax from yellow flax. The sorter should find use in removing other defects found in grain, such as scab-damaged and bunted wheat. Parts for the system cost less than $2,000, so it may be economical to run several systems in parallel to keep up with processing plant rates.