基于图像处理技术和神经网络实现机采茶分级

Grading of Machine Picked Tea Based on Image Processing Technology and Neural Network

为解决名优绿茶采摘环节的瓶颈问题,提出对机采大宗绿茶进行分级的思路。现有绿茶机采设备采摘的鲜叶一般只能制作普通的大宗绿茶,鲜叶存在混杂、破碎率高和老梗叶等问题,本文基于Labview vision、图像处理技术和神经网络算法分析机采绿茶成品的凸包面积、凸包周长、长轴长度、短轴长度等形态特征并对样本进行分类,实现从机采大宗绿茶中分选出名优绿茶。其中样本的形态特征采用工业CCD摄像头获取;用户界面用Labview自定义开发设计,数据交互方便,开发周期短。茶叶样本试验结果表明:该方案机采绿茶成品的分级正确率可以稳定在90%以上。本研究为进一步研究机采茶分级设备提供了良好的理论基础。

To solve the picking problem of famous green tea,a new technology was proposed to classify machinepicked tea in this paper.Fresh tea leaves plucked by machine are often mixed,with tea stalks and have a high brokenrate and thereby only suitable for making general green tea.The convex hull area,convex hull perimeter,long axiallength,short axial length and other morphological features of machine plucked tea leaves were analyzed by Labviewvision,image processing technology and neural network to screen high quality tea.Industrial CCD camera withappropriate optical system was used to collect object classification features.User interface was developed byLabview,which can realize the data interaction,convenient operation,short development cycle and meet differentusers’requirements.Finally,sample test showed that the correct rate of tea classification could reach about90%,which provides a good theoretical basis for further research of tea grading equipment.