Presentation 2004/5/21
AdaBoost-Based Classification of Multifrequency Polarimetric Speckled SAR Images
Shohei NAKAMURA, Seisuke FUKUDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Boosting is a method of pattern recognition that executes ensemble learning by combining simple rules. Even when performing a complicated classification, Boosting dose not suffer increase in computation time and degradation of generalization by updating weights of each learner. In the classification of synthetic aperture radar (SAR) images, fluctuation of features due to speckle noise sometimes reduces classification accuracy. In this paper, we have performed land cover classification of multifrequency polarimetric SAR images robust against speckle using the AdaBoost algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Boosting / Speckle / SAR / Multilook / Image Classification / Generalization
Paper # SANE2004-17
Date of Issue

Conference Information
Committee SANE
Conference Date 2004/5/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Space, Aeronautical and Navigational Electronics (SANE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) AdaBoost-Based Classification of Multifrequency Polarimetric Speckled SAR Images
Sub Title (in English)
Keyword(1) Boosting
Keyword(2) Speckle
Keyword(3) SAR
Keyword(4) Multilook
Keyword(5) Image Classification
Keyword(6) Generalization
1st Author's Name Shohei NAKAMURA
1st Author's Affiliation Musashi Institute of Technology()
2nd Author's Name Seisuke FUKUDA
2nd Author's Affiliation Japan Aerospace Exploration Agency Institute of Space and Astronautical Science
Date 2004/5/21
Paper # SANE2004-17
Volume (vol) vol.104
Number (no) 97
Page pp.pp.-
#Pages 5
Date of Issue