基于神经网络图像分析的智能鲜茶叶分选机  

Smart fresh tea sorter with neural network image analysis

作　　者：常春[1] 陈怡群[1] 肖宏儒[1] 任守纲[2] 宋卫东[1] 

机构地区：[1]农业部南京农业机械化研究所,南京市210014 [2]南京农业大学信息科技学院,南京市210095

出　　处：《中国农机化学报》2013年第1期137-141,共5页Journal of Chinese Agricultural Mechanization

基　　金：中国农业科学院中央级公益性科研院所基本科研业务费专项(项目编号:0032010049)

摘　　要：介绍一种用人工神经网络技术对鲜茶叶进行分类的方法和采用这种方法对鲜茶叶进行分类的智能鲜茶叶分选机。利用茶叶图像的面积、周长、等二阶距椭圆长、短轴长度等几何特征和图像的对比度、平滑度等纹理特征进行BP网络的训练可以将茶叶分类,并获得较好的茶叶分类效果。智能鲜茶叶分选机的分选工作在一个旋转水平圆盘上完成,摄像机采集到圆盘上的茶叶图像,计算机的神经网络图像分析程序对茶叶进行种类识别,并将计算结果通知分选机的执行装置,完成分类。试验表明,这种鲜茶叶分选机的分类正确率可达到90%。Introducing a technique to classify fresh tea leaves using artificial neural network,and the machine conducting the technique we hereby name it Smart Fresh Tea Sorter.With BP network,Smart Fresh Tea Sorter can analyze the image of a tea leaf in its geometric features such as perimeter,major axis length,minor axis length of the same second-order moment ellipse,and in its texture parameters such as image contrast and smoothness,and the classification is satisfactory.The sorting job is done on a rotating flat disc.The camera captures the images of tea leaves on the disc,and the neural network program will identify the images,classify them,and then transfer the results of calculation to the operating device to complete the classification.The tests showed that Smart Fresh Tea Sorter's accuracy is 90%.

关 键 词：人工神经网络 图像识别 茶叶分选 特征值 茶叶机械 

分 类 号：TS272[农业科学—茶叶生产加工]