基于机器视觉的大米垩白米的检测方法研究

本文提出了一种检测大米垩白米的检测方法。由于光线反射的缘故,摄像机拍摄的正常米粒有部分的光线反射区,该反射区与垩白米的亮度非常近似,用最大类间方差法无法区分开垩白米和正常米粒的光线反射区。本文提出通过分析两种米粒的不同的直方图分布表现来进行区分的方法,并进而分析出大米垩白率,垩白度等信息。本文提出的方法相较于以往的方式,能快速,高效地检测出大米垩白米及相关信息,实验结果表明了该方法的可行性。

Research on Detection Method of Chalk Rice Based on Machine Vision

The paper proposed a detection algorithm of chalk rice. Because of the light reflection, the image of normal rice has an area of light reflection, which is similar with the character of chalk rice. It is difficult to distinguish the chalk rice and light reflection area of normal rice by OTSU. In this paper, a new detection algorithm by analyzing the histogram character of the two kinds of rice to get the distinguish information is proposed. Furthermore we can then get the information of chalky grain rate and chalk degree. The algorithm can detect chalk rice quickly and efficiently. Experiment result indicates that our algorithm achieves significant improvement.