|
|||||
|
|
| 小麦腥黑穗病害的模式识别诊断方法研究 | |
| 引用本文： | 邓继忠,李敏,袁之报,金济,黄华盛,王张.小麦腥黑穗病害的模式识别诊断方法研究[J].农机化研究,2012,34(5):38-41. |
| 作者姓名： | 邓继忠 李敏 袁之报 金济 黄华盛 王张 |
| 作者单位： | 1. 华南农业大学工程学院,广州,510642 2. 海南出入境检验检疫局热带植物隔离检疫中心,海口,570311 3. 海南出入境检验检疫局检验检疫技术中心,海口,570311 |
| 基金项目： | 国家质检公益性行业科研专项项目 |
| 摘 要： | 针对小麦腥黑穗病害的形状和纹理特征选取了最有利于病害分类的6个特征参数,并分别利用最小距离法、BP神经网络和支持向量机3种模式识别方法实现了对小麦腥黑穗病害的诊断研究。对包括小麦网腥黑穗病、小麦印度腥黑穗病以及小麦矮腥黑穗病共48个孢子病害图像进行了分类诊断实验。实验表明,支持向量机法对小麦腥黑穗病的分类识别能力优于最小距离法和BP神经网络,当核函数是Sigmoid时,支持向量机法性能最优,总体识别率达到93.9%。 |
| 关 键 词： | 小麦腥黑穗病 模式识别 分类诊断 孢子 支持向量机 |
Research on Pattern Recognition Diagnosis of Tilletia Diseases |
|
| Deng Jizhong , Li Min , Yuan Zhibao , Jin Ji , Huang Huasheng , Wang Zhang.Research on Pattern Recognition Diagnosis of Tilletia Diseases[J].Journal of Agricultural Mechanization Research,2012,34(5):38-41. | |
| Authors: | Deng Jizhong Li Min Yuan Zhibao Jin Ji Huang Huasheng Wang Zhang |
| Institution: | 1.College of Engineering,South China Agricultural University,Guangzhou 510642,China;2.a.Tropcs Plants Segregate Quarantime Centre;b.Inspection and Quarantine Technology Center, Hainan Entry-Exit Inspection and Quarantine Bureau,Haikou 570311,China) |
| Abstract: | 6 typical features are chosen from the shape and texture features of tilletia diseases,which parameters are beneficial to classify the diseases.Then three pattern recognition methods,minimum distance classifier,BP neural network and support vector machine,are used for the classification and diagnosis of tilletia diseases.48 spore images of tilletia diseases contain tilletia caries(DC.)Tul.,tilletia indica mitra,tilletia controversa Kühn are tested for the classification.The result shows that support vector machine is superior to minimum distance classifier and BP neural network.SVM responses the optimal performance with sigmoid kernel function,and the total accuracy rate reaches up to 93.9%. |
| Keywords: | tilletia diseases pattern recognition classification and diagnosis spores support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |