首页 | 本学科首页   官方微博 | 高级检索  
     

基于核K—均值聚类算法的植物叶部病害识
引用本文:王守志,何东健,李文,王艳春. 基于核K—均值聚类算法的植物叶部病害识[J]. 农业机械学报, 2009, 40(3): 152-155
作者姓名:王守志  何东健  李文  王艳春
作者单位:1. 西北农林科技大学机械与电子工程学院,陕西杨凌,712100;威海职业学院机电工程系,威海,264210
2. 西北农林科技大学信息工程学院,陕西杨凌,712100
3. 威海职业学院机电工程系,威海,264210
4. 西北农林科技大学机械与电子工程学院,陕西杨凌,712100
摘    要:针对植物叶部病害图像的特点,首先对采集到的玉米病害彩色图像采用矢量中值滤波法去除噪声,然后提取玉米病叶彩色图像的纹理特征和颜色特征作为特征向量,利用Mercer核,把输入空间的样本映射到高维特征空间进行K-均值聚类以及植物病害识别.试验涉及的4种玉米病害识别正确率达82.5%,核K-均值聚类方法适合玉米叶部病害分类.

关 键 词:植物病害  病害识别  核K-均值聚类

Leaf Disease Recognition Based on Kernel K-means Clustering Algorithm
Wang Shouzhi,He Dongjian,Li Wen,Wang Yanchun. Leaf Disease Recognition Based on Kernel K-means Clustering Algorithm[J]. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(3): 152-155
Authors:Wang Shouzhi  He Dongjian  Li Wen  Wang Yanchun
Affiliation:1.College of Mechanical and Electronic Engineering;Northwest A&F University;Yangling;Shaanxi 712100;China2.Mechanical and Electronic Engineering Department;Weihai Vocational College;Weihai 264210;China3.College of Information Engineering;China
Abstract:Based on the features of plant disease image,vector median filter was firstly applied to remove noise of the acquied color images of grape leaf with disease.Then texture features and color features of color image of leaf with disease were extraccted as feature vector.And by using Mercer kernel functions,the data in the original space was maped to a high-dimensional feature space in which the data has been clustered efficiently.The precision of four kinds of experimental maize diseases recognition is 82.5%,a...
Keywords:Plant disease  Disease recognition  Kernel K-means clustering algorithm  
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《农业机械学报》浏览原始摘要信息
点击此处可从《农业机械学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号