谷物视觉检测系统的硬件环境

Hardware Environment of Vision System for Grain Inspection

探讨了光源波长、放大率、背景对谷物视觉检测效果的影响。通过测定不同质量谷物表面的反射光谱和对不同CCD的响应分析,结果表明:对于黑白图像识别系统,若采用815nm的滤光片,则有助于充分发挥CCD的响应性能,并为检测时区分正常与霉变谷物提供最大的对比;而彩色图像识别系统应采用在可见光波段具有足够辐射强度的全色光源,利用机器视觉试验系统分别在使用不同延长管或背景时采集谷物样品图像,并用开发的算法软件进行识别。放大率试验结果表明,当采用20～25mm的镜头延长管时,采集的样品组图像中正常和微裂谷粒的分类精度均超过97%,是最优的放大率选择。背景试验表明:在检测霉变和芽谷时,采用白背景比黑背景的识别精度高,进而提示以往传统认为采用暗背景有助于提供与观察对象最大的对比,只适用于无霉变的谷物。可见,经过优化的硬件环境有助于视觉检测精度的提高。

Obtaining clear images advantaged of improving the classification accuracy involves many factors. Wavelength of light source, magnification and background were studied in this paper. The analysis of rice seed reflectance spectroseopic curves showed that the wavelength of light source for discrimination of the moulded seeds from normal rice seeds in the monochromic image recognition mode was about 815 nm for jinyou 402 and shanyoul0. To determine optimizing conditions for acquiring digital images of rice seed using a computer vision system, an adjustable color machine vision system was developed. The machine vision system with 20 mm to 25 mm lens extender produce close-up images which made it easy to recognize incompletely closed glumes in grain. Their classification accuracies were 〉98%, which was an optimum selection for magnification. White background was proved to be better than black background for inspecting grains infected by mould using the algorithms based on color and inspecting germinated grains using the algorithms based on shape. Experimental results indicated better classification in optimizing condition for quality inspection of grain. Specifically, the image proeessing can correct for details such as fine fissure with the machine vision system.