Presentation 2003/3/6
Automatic Satellite Image Segmentation Using Topology of Data
Yukimasa OKADA, Yuji WAIZUMI, Nei KATOt, Yoshiaki NEMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a method for an automatic satellite image segmentation, which utilizes data distribution patterns in multi-dimensional feature space, cloud pixels in particular. The algorithm of the proposed method mainly consists of k-means method and Competitive Hebbian Learning to preserve a topology of data in the feature space. To demonstrate a validity of the proposed method, we apply several satellite image data to the method. The result clearly shows that extraction of cloud pixels is possible with high accuracy. In addition, our proposed method identifies high and low cloud pixels, and snow pixels in satellite image data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Segmentation / Remote Sensing / Competitive Hebbian Learning / Topology Preserving Map
Paper # PRMU2002-224
Date of Issue

Conference Information
Committee PRMU
Conference Date 2003/3/6(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Automatic Satellite Image Segmentation Using Topology of Data
Sub Title (in English)
Keyword(1) Segmentation
Keyword(2) Remote Sensing
Keyword(3) Competitive Hebbian Learning
Keyword(4) Topology Preserving Map
1st Author's Name Yukimasa OKADA
1st Author's Affiliation Graduate School of Information Sciences, TOHOKU University()
2nd Author's Name Yuji WAIZUMI
2nd Author's Affiliation Graduate School of Information Sciences, TOHOKU University
3rd Author's Name Nei KATOt
3rd Author's Affiliation Graduate School of Information Sciences, TOHOKU University
4th Author's Name Yoshiaki NEMOTO
4th Author's Affiliation Graduate School of Information Sciences, TOHOKU University
Date 2003/3/6
Paper # PRMU2002-224
Volume (vol) vol.102
Number (no) 707
Page pp.pp.-
#Pages 6
Date of Issue