利用决策级融合进行遥感影像分类  

Classification for Remote Sensing Data with Decision Level Fusion

作　　者：许凯[1] 秦昆[1] 杜鹚 

机构地区：[1]武汉大学遥感信息工程学院,武汉市430079 [2]通信网络技术管理中心,北京市100840

出　　处：《武汉大学学报（信息科学版）》2009年第7期826-829,共4页Geomatics and Information Science of Wuhan University

基　　金：国家973计划资助项目(2007CB311003);国家自然科学基金资助项目(60875007)

摘　　要：提出了一种基于决策级融合的遥感影像分类方法。该方法对遥感影像特征以最大似然分类器进行预分类,应用Adaboost算法将分类的结果进行决策级融合,实现影像的分类。实验结果表明,该方法的分类精度较传统分类方法有明显的提高。With the development of remote sensing technology, dealing with high-dimension features with traditional classification methods is difficult. Multiple classifiers fusion technology not only deals with high-dimension features but also improves the classification accuracies. We focuses on classifier fusion in decision level, and proposes a new classification method for remote sensing data based on Adaboost. Experiments show that this method is more effective than traditional classification algorithms.

关 键 词：决策级融合 ADABOOST 遥感影像分类 纹理 

分 类 号：P237.4[天文地球—摄影测量与遥感]