Nonliner Blind Separation of Image Signals Using a Modified Neural Network

カテゴリ: 部門大会

論文No: GS11-3

グループ名: 【C】平成18年電気学会電子・情報・システム部門大会講演論文集

発行日: 2006/09/05

タイトル(英語): Nonliner Blind Separation of Image Signals Using a Modified Neural Network

著者名: 張暁維 (千葉大学),呂建明 (千葉大学),谷萩隆嗣 (千葉大学)

著者名(英語): Xiaowei Zhang(Chiba University),Jianming Lu(Chiba University),Takashi Yahagi(Chiba University)

キーワード: Nonlinear Blind Source Separation|Self-organizing Maps|Expectation-MaximizationExpectation-Maximization|Nonlinear Blind Source Separation|Self-organizing Maps|Expectation-Maximization

要約(日本語): Nonlinear Blind Source Separation (NLBSS) has received much research attention
recently due to the emergence of simple, powerful algorithms that show promise in　practical applications. In this paper, we propose a learning algorithm which is based on Self-Organizing Maps (SOM) for NLBSS problem. We introduce EM(Expectation-Maximization) into the previous SOM. It has the benefits of both EM and SOM algorithms.　We show that the SOM-based approach can provide a solution to the nonlinear BSS problem.

PDFファイルサイズ: 831 Kバイト