日本語
Hiroshima City University 
Graduate School of Information Sciences 
Dept. of Intelligent Systems 

Associate Professor 
Suematsu Nobuo 

Academic background
Kyushu University  Faculty of Science  物理  1988 
Kyushu University  Graduate School, Division of Natural Science  物理学  Doctor prophase  1990 

Research Areas
Machine Learning 

Subject of research
Reinfarcement Learning in Partially Observable Environments 
Ensemble Learning Methods 

Papers
Research paper (international conference proceedings)  Joint  Extending the Full Procrustes Distance to Anisotropic Scale in Shape Analysis  sukasa Okamoto, Kazunori Iwata, Nobuo Suematsu  Proceedings of the 4th IAPR Asian Conference on Pattern Recognition  634-639  2017/11/29 
Research paper (international conference proceedings)  Joint  A Sampling Method for Processing Contours Drawn with an Uncertain Stroke Order and Number  Kazuya Ose, Kazunori Iwata, Nobuo Suematsu  Proceedings of the 15th IAPR International Conference on Machine Vision Applications  438-441  2017/05/08 
Research paper (scientific journal)  Joint  大学におけるクラウド前提の学術情報基盤への移行と分析  前田 香織,末松 伸朗, 北村 俊明  情報処理学会論文誌  57/ 3, 948-957  2016/03 
Research paper (scientific journal)  Joint  A Spatially Correlated Mixture Model for Image Segmentation  Kosei KURISU, Nobuo SUEMATSU, Kazunori IWATA, Akira HAYASHI  IEICE TRANSACTIONS on Information and Systems  E98-D/ 4, 930-937  2015/04/01  1745-1361  In image segmentation, finite mixture modeling has been widely used. In its simplest form, the spatial correlation among neighboring pixels is not taken into account, and its segmentation results can be largely deteriorated by noise in images. We propose a spatially correlated mixture model in which the mixing proportions of finite mixture models are governed by a set of underlying functions defined on the image space. The spatial correlation among pixels is introduced by putting a Gaussian process prior on the underlying functions. We can set the spatial correlation rather directly and flexibly by choosing the covariance function of the Gaussian process prior. The effectiveness of our model is demonstrated by experiments with synthetic and real images. 
Research paper (scientific journal)  Joint  Marginalized Viterbi Algorithm for Hierarchical Hidden Markov Models  Akira Hayashi, Kazunori Iwata, Nobuo Suematsu  Pattern Recognition  46/ 12  2013/12 
View details...

Research presentations
Poster presentation  近傍標本点の対応を考慮した形状間の対応づけとその計算コストの削減  第20回情報論的学習理論ワークショップ  2017/11/09 
Poster presentation  フルプロクルステス距離の拡張とその距離計量学習  第20回情報論的学習理論ワークショップ  2017/11/09 
Oral presentation(general)  ノンパラメトリックベイズ時系列整列法の高速化  平成29年度(第68回)電気・情報関連学会中国支部連合大会  2017/10/21 
Oral presentation(general)  近傍標本点の対応を考慮した形状整合の高速化  平成29年度(第68回)電気・情報関連学会中国支部連合大会  2017/10/21 
Poster presentation  マルコフ連鎖モンテカルロ法を用いた画像レジストレーション  第20回画像の認識・理解シンポジウム  2017/08/09 
View details...

Allotted class
パターン認識 
情報科学基礎実験B 
View details...

Memberships of academic societies
IEICE (The Institute of Electronics, Information and Communication Engineers)  2006/05/01-Present 
情報処理学会 
人工知能学会 
View details...