MULTIPLE KERNEL RELEVANCE VECTOR MACHINE FOR GEOSPATIAL OBJECTS DETECTION IN HIGH-RESOLUTION REMOTE SENSING IMAGES1

Geospatial objects detection within complex environment is a challenging problem in re- mote sensing area. In this paper, we derive an extension of the Relevance Vector Machine （RVM） technique to multiple kernel version. The proposed method learns an optimal kernel combination and the associated cla...

Full description

Saved in:
Bibliographic Details
Published in电子科学学刊：英文版 Vol. 29; no. 5; pp. 353 - 360
Main Author Li Xiangjuan Sun Xian Wang Hongqi LiYu Sun Hao
Format Journal Article
LanguageEnglish
Published 2012
Online AccessGet full text

Cover

More Information
Summary:Geospatial objects detection within complex environment is a challenging problem in re- mote sensing area. In this paper, we derive an extension of the Relevance Vector Machine （RVM） technique to multiple kernel version. The proposed method learns an optimal kernel combination and the associated classifier simultaneously. Two feature types are extracted from images, forming basis kernels. Then these basis kernels are weighted combined and resulted the composite kernel exploits interesting points and appearance information of objects simultaneously. Weights and the detection model are finally learnt by a new algorithm. Experimental results show that the proposed method improve detection accuracy to above 88%, yields good interpretation for the selected subset of features and appears sparser than traditional single-kernel RVMs.
Bibliography:11-2003/TN
Object detection; Feature extraction; Relevance Vector Machine （RVM）; Support VectorMachine （SVM）; Sliding-window
Geospatial objects detection within complex environment is a challenging problem in re- mote sensing area. In this paper, we derive an extension of the Relevance Vector Machine （RVM） technique to multiple kernel version. The proposed method learns an optimal kernel combination and the associated classifier simultaneously. Two feature types are extracted from images, forming basis kernels. Then these basis kernels are weighted combined and resulted the composite kernel exploits interesting points and appearance information of objects simultaneously. Weights and the detection model are finally learnt by a new algorithm. Experimental results show that the proposed method improve detection accuracy to above 88%, yields good interpretation for the selected subset of features and appears sparser than traditional single-kernel RVMs.
ISSN:0217-9822
1993-0615