IGSNRR OpenIR
MPS-based information extraction method for remotely sensed imagery: a comparison of fusion methods
Ge, Yong1,2; Bai, Hexiang1,3
2010-12-01
Source PublicationCANADIAN JOURNAL OF REMOTE SENSING
Volume36Issue:6Pages:763-779
AbstractRecently, multiple-point simulation (MPS) was introduced to increase the accuracy of information extraction from remotely sensed imagery by incorporating structural information through a training image. An important procedure in the MPS-based information extraction method is the fusion of two probability fields from two different classifiers, extracting spectral and spatial structure information. In previous studies the fusion process was accomplished using the theory of evidence and the theory of consensus. It has been shown that these fusion methods each have their own capabilities and characteristics for different data under different circumstances. This paper investigates primarily the advantages and disadvantages of three different types of fusion methods: evidence-based, consensus-based, and probability-based, and then compares the fusion results through an accuracy assessment. For validation purposes, we selected two remotely sensed images taken in different areas and with distinct structural characteristics of roads. Both images, from Satellite Pour l'Observation de la Terre 5 (SPOT5) with spatial resolution of 10 m, were used to investigate the performance of the three fusion methods in extracting road information with distinct structural characteristics from the images. A comparison of the different fusion methods can assist users in selecting the appropriate fusion method for the given data characteristics. Based on the results of two experiments, the relationships between these fusion methods are further investigated.
SubtypeArticle
WOS HeadingsScience & Technology ; Technology
WOS Subject ExtendedRemote Sensing
WOS KeywordCLASSIFICATION
Indexed BySCI
Language英语
WOS SubjectRemote Sensing
WOS IDWOS:000208356600011
PublisherCANADIAN AERONAUTICS SPACE INST
Citation statistics
Cited Times:5[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.igsnrr.ac.cn/handle/311030/67978
Collection中国科学院地理科学与资源研究所
Corresponding AuthorGe, Yong
Affiliation1.Chinese Acad Sci, State Key Lab Resources & Environm Informat Syst, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China
2.York Univ, Dept Earth & Space Sci & Engn, Toronto, ON M3J 1P3, Canada
3.Chinese Acad Sci, Grad Univ, Beijing 100049, Peoples R China
Recommended Citation
GB/T 7714
Ge, Yong,Bai, Hexiang. MPS-based information extraction method for remotely sensed imagery: a comparison of fusion methods[J]. CANADIAN JOURNAL OF REMOTE SENSING,2010,36(6):763-779.
APA Ge, Yong,&Bai, Hexiang.(2010).MPS-based information extraction method for remotely sensed imagery: a comparison of fusion methods.CANADIAN JOURNAL OF REMOTE SENSING,36(6),763-779.
MLA Ge, Yong,et al."MPS-based information extraction method for remotely sensed imagery: a comparison of fusion methods".CANADIAN JOURNAL OF REMOTE SENSING 36.6(2010):763-779.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Ge, Yong]'s Articles
[Bai, Hexiang]'s Articles
Baidu academic
Similar articles in Baidu academic
[Ge, Yong]'s Articles
[Bai, Hexiang]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Ge, Yong]'s Articles
[Bai, Hexiang]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.