| 面向对象影像分类中基于最大化互信息的特征选择 |
| |
| 引用本文： | 吴波,朱勤东,高海燕,周小成.面向对象影像分类中基于最大化互信息的特征选择[J].国土资源遥感,2009(3):30-34. |
| |
| 作者姓名： | 吴波 朱勤东 高海燕 周小成 |
| |
| 作者单位： | 福州大学空间数据挖掘与信息共享教育部重点实验室,福州,350002 |
| |
| 基金项目： | 国家自然科学基金，福建省科技厅重点项目，教育部重大项目培育资金项目 |
| |
| 摘 要： | 高分辨率影像面向对象分割后产生了大量的光谱、形状以及纹理特征,如何抽取出最佳特征子集是遥感影像识别的重要问题.本文利用最大化互信息统计独立准则抽取最优特征子集,提高了面向对象遥感影像分类精度.基本过程包含以下3个方面:首先,利用eCoginition软件对高分辨遥感影像进行对象分割;然后,基于互信息最大关联、最小冗余准则(mRMR)获取优选的特征子集;最后,基于支持向量机分类器完成影像分类.以福建省漳州市QuickBird数据为例的实验表明,该方法能够有效提高遥感影像的分类精度,平均误分率降低了约4%.
|
| 关 键 词： | 互信息 特征选择 面向对象分类 高分辨 |
FEATURE SELECTION BASED ON MAXIMAL MUTUAL INFORMATION CRITERION IN OBJECT-ORIENTED CLASSIFICATION |
| |
| WU Bo,ZHU Qin-dong,GAO Hai-yan,ZHOU Xiao-cheng.FEATURE SELECTION BASED ON MAXIMAL MUTUAL INFORMATION CRITERION IN OBJECT-ORIENTED CLASSIFICATION[J].Remote Sensing for Land & Resources,2009(3):30-34. |
| |
| Authors: | WU Bo ZHU Qin-dong GAO Hai-yan ZHOU Xiao-cheng |
| |
| Institution: | Key Lab of Spatial Data Mining and Information Sharing of Ministry of Education;Fuzhou University;Fuzhou 350002;China |
| |
| Abstract: | It is a key problem to select optimal features from the total set where spectral,geometric,shape,texture features and some other features are extracted by the process of image segmentation in object-oriented classification.In this paper,the authors present a method for selecting good features from object-oriented image segmentation according to the maximal statistical mutual information dependency criterion so as to improve the classification accuracy of high spatial resolution image.The proposed method is ... |
| |
| Keywords: | Mutual information Feature selection Object-oriented classification High spatial resolution image |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
