基于图像处理的茶叶识别方法研究

Research on Tea Identification Method Based on Image Processing

为了剔除茶叶中的黄叶、茶梗,将茶叶分选为多个等级,提高茶叶的附加价值。以贵州卷曲状机采茶为样本,采用色选算法、色选+形选综合算法对茶叶、黄叶、茶梗进行识别,采用工业线阵CCD相机获取茶叶实时动态RGB数据,转换为HSV彩色图像,通过滤波预处理、颜色和形状特征提取、阈值分割、形态学处理等程序,根据茶叶的有效特征参数将黄叶、茶梗剔除。结果表明:采用色选+形选综合算法时茶叶、黄叶、茶梗正确识别率分别提高至89%、88%、93%。相较于色选算法,采用色选+形选综合算法,茶叶、黄叶、茶梗的识别率显著提高。

In order to eliminate the yellow leaves and tea stalks in the finished tea,the tea was classified into multiple grades to improve the added value of tea.In this paper,taking guizhou curly tea as a sample,the correct recognition rate of tea,yellow leaves and tea stalks was studied when color selection algorithm and color selection+shape selection algorithm were adopted.Real-time dynamic RGB data of tea were obtained by industrial linear CCD camera and converted into HSV color images.Yellow leaves and tea stalks were removed according to effective characteristic parameters of tea leaves through filtering pretreatment,extraction of color and shape features,threshold analysis and morphological processing.The results showed that the correct recognition rate of tea,yellow leaves and tea stalks was improved to 89%,88%and 93%respectively under color selection+shape selection algorithm.Compared with the color selection algorithm,the recognition rate of yellow leaves and tea stalks is significantly improved by using the color selection+shape selection algorithm.