基于机器视觉的枸杞分级方法

针对目前传统的枸杞分级主要采用人工方法，费时费力且效率不高的缺点，提出了一种基于机器视觉技术对枸杞进行自动分类的方法。采用数字图像处理技术对枸杞图像进行了预处理、分割，从而提取枸杞的色泽、大小及形状等特征参数；用K-means算法对特征进行聚类，得到枸杞相应等级的基准；根据聚类分析得到的基准采用最小距离分类器对枸杞进行分级。实验结果表明，该方法能够准确快速地对不同色泽和大小的枸杞进行分类。

Wolfberry classification method based on machine vision

Traditiona1 wo1tberry sorting primari1y uses artificia1 method. It has timeconsuming and inefficient shortcomings. An automatic wo1fberry c1assification method based on machine vision is proposed. This paper uses digita1 image processing tech no1ogy for wo1fberry image preprocessing, segmentation and extraction of characteristic parameters of co1or, size and shape; it uses the Kmeans c1ustering feature to get the base1ine of wo1fberry appropriate 1eve1; it grades wo1fberry by minimum distance c1assifier based on the trained benchmark. The experimenta1 resu1ts show that this method can c1assify different co1ors and sizes ofwo1fberry more accurate1y and quick1y.