• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Modeling, Analysis, Simulation, Technologies and Applications >Intelligent Identification of Traditional Chinese Medicine Materials Based on Multi-feature Extraction and Pattern Recognition
【24h】

Intelligent Identification of Traditional Chinese Medicine Materials Based on Multi-feature Extraction and Pattern Recognition

机译：基于多特征提取和模式识别的中药材智能鉴定

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A discussion about image pattern recognition for Tradition Chinese Medicine (TCM) materials was explained in this paper. 150 images of each category of TCM materials were gathered, in total of five categories. 80% of the images were distributed as training samples randomly and the other 20% were used to test the pattern recognition algorithms. A multi-feature vector for each image was proposed including textual features, shape features and category labels to train pattern recognition methods K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) and test the recognition rates. Statistics of average recognition rates were made and indicated that the methods could classified the chosen five categories of TCM materials significantly with the accuracy of around 70% in average, providing a new solution for TCM materials intelligent identification.
机译：本文解释了关于传统中医（TCM）材料的图像模式识别的探讨。每类中医材料的150个图像共聚集，共有五类。 80％的图像作为随机训练样本分配，其他20％用于测试模式识别算法。提出了每个图像的多个特征向量，包括培训模式识别方法K-COMBERD邻（KNN）和支持向量机（SVM）并测试识别率的文本特征，形状特征和类别标签。进行了平均识别率的统计数据，并表明该方法可以在大约70％的精度下显着分类为选定的五类中药材，为TCM材料智能识别提供了新的解决方案。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号