Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Science of Surveying and Mapping》 2013-05
Add to Favorite Get Latest Update

Objects detection in high-resolution remote sensing images using multiple kernel learning

LI Xiang-juan;SUN Xian;WANG Hong-qi;Key Laboratory of Technology in Geo-spatial Information Processing and Application System/Institute of Electronics,Chinese Academy of Sciences;Graduate University of Chinese Academy of Sciences;  
In order to improve the performance of objects detection within complex environment in remote sensing application,a multiple kernel learning-based detection method was proposed in the paper. The proposed method includes two main stages: feature extraction and model training. In the feature extraction stage,point feature and appearance feature in multi-scales are both used to describe multi-classes of objects. In the model training stage,linear combination of kernels and product of kernels is used to combine the extracted basis kernels.Then,the basis kernel weights are learnt under SVM framework. The proposed classifier acts as a sliding-window detector to search objects in remote sensing images. Experimental results demonstrated that this method could outperform the traditional single-kernel SVM.
【Fund】: 国家自然科学基金资助项目(41001285)
【CateGory Index】: TP751;P237
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved