Low-cost bi-chromatic image sorting device for grains

摘要

A low-cost linescan imaging system was developed to inspect and sort grains and other products at high speeds (40 kernels/s). The device captures bi-chromatic images from opposite sides of each kernel and processes the images in real time using high speed microcontrollers. Detection of scab-damaged wheat kernels was used in this study to establish system feasibility and limits. Simple image statistics and intensity histograms were used as features and were able to distinguish good kernels from scab-damaged kernels with 95% accuracy. For each kernel, image acquisition required approximately 15 ms, while 5 ms were required for image processing and classification. The controller can output a signal to divert (sort) kernels or save the images on a compact flash card for transfer to a personal computer for off-line analysis. All parts for the system cost less than 2000 dollar.