欢迎访问《中国农学通报》,

中国农学通报 ›› 2015, Vol. 31 ›› Issue (26): 261-265.doi: 10.11924/j.issn.1000-6850.casb15050120

所属专题: 园艺

• 农业科技信息 • 上一篇    下一篇

基于移动溯源与图像分析的茶叶品级鉴定方法研究

葛 迪,李绍稳,魏 同,徐 静   

  1. (安徽农业大学信息与计算机学院,合肥 230036)
  • 收稿日期:2015-05-22 修回日期:2015-06-20 接受日期:2015-07-13 出版日期:2015-09-23 发布日期:2015-09-23
  • 通讯作者: 李绍稳

Research on the Identification Method of Tea Grade Based on Mobile Traceability and Image Analysis

Ge Di, Li Shaowen, Wei Tong, Xu Jing   

  1. (School of Information and Computer, Anhui Agriculture University, Hefei 230036)
  • Received:2015-05-22 Revised:2015-06-20 Accepted:2015-07-13 Online:2015-09-23 Published:2015-09-23

摘要: 为解决茶叶在质量追溯时防伪和品质鉴定问题, 以六安瓜片为研究对象, 提出一种结合移动计算和图像分析的茶叶质量追溯方法。该方法的实现以 Android操作系统为平台, 通过扫描茶叶质量追溯二维码查询溯源信息。在此基础上, 利用智能终端拍照获取完整的单片茶叶正面照片并上传至后台服务器。后台编写的 MATLAB图像处理程序对采集的图像进行预处理和特征参数提取, 将提取的颜色和形状共 7项特征参数作为参数集合, 运用 SVM分类器对茶叶进行分类, 最终将结果反馈到溯源客户端。测试结果证明, 该方法能快速帮助消费者对茶叶品级进行初步鉴定。

关键词: 耕地, 耕地, 质量评价, GIS

Abstract: In order to solve the problem of anti-counterfeiting and quality identification in the process of tea quality traceability, we take Lu’ an Guapian as the research subject and propose a method of tea quality tracing combining mobile computing with image analysis. Based on Android operating system, the query traceability information was realized through scanning the quality of the tea two- dimensional code. On these basic, complete single photo of tea was obtained by the mobile terminal and then was uploaded to the server. Later, the MATLAB image processing program written by backstage preprocessed the collected images and extracted the feature parameter. Taking a total of seven parameters of the extracted colors and shapes as parameters sets, we utilized SVM classifier to classify the tea, and sent feedback results to the source client. Test results show that this method can help consumers to identify the tea varieties rapidly.