Presentation | 2017-11-09 Flexible Unsupervised PolSAR Land Classification System Based on Quaternion Neural Networks Hyunsoo Kim, Akira Hirose, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | We propose a flexible unsupervised PolSAR land classification system based on quaternion neural networks. The existing PolSAR land classification systems use the feature information necessary for the land classification based on a few scattering models which human beings designed beforehand. However, such methods have limitations in the near future when we expect classification into a large number of land categories. By using quaternion auto-encoder and quaternion SOM for feature extraction and classification, respectively, our proposed system realizes unsupervised land classification that does not require predefining or learning by human beings. As a result, we can discover even new and more detailed land categories that can not be classified by the existing systems. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Polarimetric synthetic aperture radar (PolSAR) / unsupervised land classification / quaternion neural network / Poincare parameter / auto-encoder / self-organizing map (SOM) |
Paper # | EMT2017-48 |
Date of Issue | 2017-11-02 (EMT) |
Conference Information | |
Committee | EMT / IEE-EMT |
---|---|
Conference Date | 2017/11/9(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Tendo Hotel (Tendo, Yamagata) |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Electromagnetic Theory, etc. |
Chair | Akira Hirose(Univ. of Tokyo) / Keiji Goto(National Defense Academy) |
Vice Chair | Koichi Hirayama(Kitami Inst. of Tech.) |
Secretary | Koichi Hirayama(Univ. of Hyogo) / (Tokyo Metro. Coll. of Tech) |
Assistant | Tsuyoshi Matsuoka(Kyushu Sangyo Univ.) / Yoshihiro Naka(Kyushu University of Health and Welfare) |
Paper Information | |
Registration To | Technical Committee on Electromagnetic Theory / Technical Meeting on Electromagnetic Theory |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Flexible Unsupervised PolSAR Land Classification System Based on Quaternion Neural Networks |
Sub Title (in English) | |
Keyword(1) | Polarimetric synthetic aperture radar (PolSAR) |
Keyword(2) | unsupervised land classification |
Keyword(3) | quaternion neural network |
Keyword(4) | Poincare parameter |
Keyword(5) | auto-encoder |
Keyword(6) | self-organizing map (SOM) |
1st Author's Name | Hyunsoo Kim |
1st Author's Affiliation | The University of Tokyo(Tokyo Univ.) |
2nd Author's Name | Akira Hirose |
2nd Author's Affiliation | The University of Tokyo(Tokyo Univ.) |
Date | 2017-11-09 |
Paper # | EMT2017-48 |
Volume (vol) | vol.117 |
Number (no) | EMT-289 |
Page | pp.pp.37-42(EMT), |
#Pages | 6 |
Date of Issue | 2017-11-02 (EMT) |