基于色调的黄粒米检测方法

A Hue Based Detecting Approach to Yellow Rice Kernel

提出了一种基于色调的黄粒米检测算法。通过对标准米和黄粒米图像色调直方图的分析和研究,提出了以黄色区域的频度均值作为特征参数,区分标准米和黄粒米。研究结果表明,该算法的识别精度大于90%,可以有效地检测黄粒米。

A new algorithm based on hue was developed, which could be used for detecting and classifying yellow rice kernels. The mean of pixel numbers in rice particle area, which represents yellow color, was selected as a color criterion to distinguish yellow rice kernels from the standard ones by comparison of these two kinds of rice images with a hue histograms. The experimental results showed that the detection accuracy is higher than 90% by using this method.