Shunji Satoh

J-GLOBAL         Last updated: Aug 19, 2011 at 11:28
 
Avatar
Name
Shunji Satoh
Affiliation
The University of Electro-Communications
Section
Graduate School of Information Systems

Research Interests

 
 

Research Areas

 
 

Published Papers

 
Yamazaki T, Ikeno H, Okumura Y, Satoh S, Kamiyama Y, Hirata Y, Inagaki K, Ishihara A, Kannon T, Usui S
Neural networks : the official journal of the International Neural Network Society   24(7) 693-698   Sep 2011   [Refereed]
Shunji Satoh
Neural Computing & Applications      2011   [Refereed]
Sasaki H, Satoh S
Neurocomputing   73(4-6) 867-873   2010   [Refereed]
Sasaki H, Satoh S
Neural networks : the official journal of the International Neural Network Society   22(4) 362-372   May 2009   [Refereed]
Satoh S, Usui S
Cognitive neurodynamics   3 1-8   Mar 2009   [Refereed]
Satoh S, Usui S
Neural networks : the official journal of the International Neural Network Society   21(9) 1261-1271   Nov 2008   [Refereed]
Satoh S
Biological cybernetics   95 259-270   Sep 2006   [Refereed]
SATOH Shunji, MIYAKE Shogo
The transactions of the Institute of Electronics, Information and Communication Engineers. D-II   88(7) 1257-1268   Jul 2005   [Refereed]
背景から物体(図)領域を検出する方法を提案し, 提案方法を実現する神経回路網モデルを提案する.提案方法は, 画像工学の立場から提唱された動的輪郭法の計算原理に基づいており, 物体の境界がもつ幾何的特徴, 及び認知心理学的知見を統合することで得られる.本研究ではまず, 動的輪郭法の定常状態に関する解析をした後, 物体領域検出のためのエネルギーを定義し, エネルギー最小化の原理から神経回路網モデルの結合や動作を導出する.本研究では神経回路網モデルの提案のみならず, 動的輪郭法の重要な問題点を解...
SHIMOMURA Masao, SATOH Shunji, MIYAKE Syogo, ASO Hirotomo
The transactions of the Institute of Electronics, Information and Communication Engineers. D-II   88(4) 769-777   Apr 2005   [Refereed]
ネオコグニトロンは, 高い認識率と拡張性をもつパターン認識用の階層型ニューラルネットワークであるが, その性能を引き出すためには多数存在するパラメータを認識対象に応じて適切に調整する必要があった.そこで本論文では, ネオコグニトロンの各階層で行われている処理が次元圧縮である点に着目し, 統計的な次元圧縮法である主成分分析(PCA), 独立成分分析(ICA)及び部分空間法をネットワークの学習法として導入することで, パラメータ数の削減とパラメータ変動への頑健性向上を図る.また, これらの手法...
SATOH Shunji, MIYAKE Shogo
The transactions of the Institute of Electronics, Information and Communication Engineers. D-II   86(10) 1490-1501   Oct 2003   [Refereed]
神経生理学で得られた知見や,視覚心理学で得られたヒトの視覚特性を考慮した,スケールスペース理論に基づく物体検出のための視覚モデルを提案する.提案モデルは主に(a)初期視覚モデル,(b)注意計算モデルで構成される.本論文ではまず,多重チャネル性,空間不均一性,及び方位選択性を考慮した視覚特性が,離散化されたスケールスペースで記述できることを示し,初期視覚モデルの定式化を行う.次に,スケールスペース理論で得られた知見に基づき,注意計算モデルの定式化,及び動作アルゴリズムの定式化を行う.数値実験...
SATOH Shunji, KUROIWA Jousuke, ASO Hirotomo, MIYAKE Shogo
The transactions of the Institute of Electronics, Information and Communication Engineers   81(6) 1365-1374   Jun 1998   [Refereed]
教師なし学習でパターン認識が可能なネオコグニトロンは位置ずれや変形に頑強な認識をする階層型神経回路モデルであるが, 回転したパターンを認識することはモデルの構成上不可能であった.本論文では, 変形に対する頑強性を実現しているネオコグニトロンの構成をそのまま回転に対して拡張し, 回転パターンも認識可能な回転対応型ネオコグニトロンを提案する.更に, 回転対応型ネオコグニトロンが所期の機能を獲得するための学習法として, 学習段階でしきい値を変化させる.しきい値制御学習法を提案する.また, 実際に...
Shunji Satoh, Hirotomo Aso, Shogo Miyake, Jousuke Kuroiwa
   [Refereed]
. A new model which can recognize rotated, distorted, scaled, shifted and noised patterns is proposed. The model is constructed based on psychological experiments in a mental rotation. The model has two types of processes: (i) one is a bottom-up p...
Shunji Satoh, Jousuke Kuroiwa, Hirotomo Aso, Shogo Miyake
   [Refereed]
A rotation-invariant neocognitron is constructed by extending the neocognitron which can recognize translated, scaled and/or distorted patterns from training ones. In constructing the model two technical methods during the learning, a "threshold-c...

Misc

 
Ohmura Junichi, Shunji Satoh, Akira Egashira, Takefumi Miyoshi, Hidetsugu Irie, Tsutomu Yoshinaga
IPSJ SIG Notes   2011(8) 1-8   Jul 2011
A popular approach to understand our human visual "functions" is performing the computational simlutaions of its linear models created with a focus on input-output relations of cells. However, due to a lot of simulation time for a huge amount of c...
Yusuke Saito, Shunji Satoh, Ohmura Junichi, Takefumi Miyoshi, Hidetsugu Irie, Tsutomu Yoshinaga
IPSJ SIG Notes   2011(4) 1-8   Mar 2011
Numerical simulation for the linear model of visual neurons is the most important approach to understand our visual system from computational viewpoints. We attempt to parallelize the time-consuming simulation on a cluster computer system. We achi...
SASAKI Hiroaki, SATOH Shunji, USUI Shiro
IEICE technical report. Neurocomputing   109(461) 7-12   Mar 2010
It is believed that in primary visual cortex(V1), neurons efficiently encode natural images by the sparse outputs of these neurons. However, such outputs of V1 cells are not sufficiently sparse because natural images have a spatial structure like ...
SATOH Shunji, SAKAGUCHI Yutaka, USUI Shiro
IEICE technical report. Neurocomputing   109(461) 391-396   Mar 2010
We propose a simple but novel mathematical framework dealing with binocular vision; using complex functions, {left image} + i{right image}. This framework expands the solutions of mathematical problems on visual functions into complex functions. I...
SATOH Shunji, USUI Shiro
IEICE technical report. Neurocomputing   108(480) 141-146   Mar 2009
A novel model for spatio-temporal receptive fields of V1 simple neurons is proposed by introducing fractional order derivatives. We consider the computational role of V1 simple neurons, and derive physiologically acceptable and theoretically optim...
KANNON Takayuki, MAKIMURA Koji, KAMIJI Nilton Liuji, SATOH Shunji, USUII Shiro
IEICE technical report. Neurocomputing   108(480) 37-42   Mar 2009
Mathematical model of the brain developed so far have targeted specific brain areas such as, cells, networks and/or phenomena, including microscopic and macroscopic levels. However, in order to understand the whole brain system, integration of suc...
SATOH Shunji
IEICE technical report. Neurocomputing   106(590) 1-6   Mar 2007
A visual model for filling-in at the blind spot is proposed. Standard regularization theory is employed to derive a visual model deductively. First, some problems of the diffusion eqation, which is frequently used for depth perception, are pointed...
SATOH Shunji
IEICE technical report. Neurocomputing   105(658) 55-60   Mar 2006
Neurophysiological experiments have revealed that post-synaptic effect conveyed by long-range horizontal connections (LHC) in V1 can be excitatory or inhibitory, and the effect highly depends on the activity of pre-synaptic neurons (non-linearity ...
SATOH Shunji
IEICE technical report. Neurocomputing   104(760) 1-6   Mar 2005
A visual model for object detection is proposed. The computational principles of active contours and level-set method are utilized to construct the proposed model. A visual model for object detection have been proposed by Satoh (IEICE Tech. Rep. V...
SATOH Shunji
Technical report of IEICE. HIP   104(525) 1-6   Dec 2004
A visual model for object detection is proposed. The dynamics of the proposed model is based on the computational principle of active contours, which is a fundamental method for various image processing. A redefined level set function is utilized ...
Satoh Shunji, Miyake Shogo
情報科学技術フォーラム一般講演論文集   3(2) 409-411   Aug 2004
SASAKI Ryo, MIYAKE Shogo, SATOH Shunji
IEICE technical report. Neurocomputing   103(734) 85-90   Mar 2004
A computational model for working memory which is controled by a reward system is proposed. The control system of working memory is formulated by use of a temporal difference model which estimates future reward. Through the computational experimen...
SHISHIDO Hideaki, MIYAKE Shogo, SATOH Shunji
IEICE technical report. Neurocomputing   103(734) 91-96   Mar 2004
From the recent physiological data, it is shown that acetylcholine controls the synaptic transmission in the hippocampus and induces the theta rhythm and the rhythm of sleep and waking. It is also suggested that acetylcholine plays a very importan...
NAKAMURA Ikuo, SATOH Shunji, MIYAKE Shogo, ASO Hirotomo
IEICE technical report. Neurocomputing   103(733) 61-66   Mar 2004
Multilayer perceptron is widely used as high-performance discrimination circuit. Learned knowledge is expressed as connection weights and output values in hidden and output layer. However it is difficult to gain the comprehensible knowledge from t...
OHIZUMI Yoji, SATOH Shunji, MIYAKE Shogo, ASO Hirotomo
IEICE technical report. Neurocomputing   103(732) 117-122   Mar 2004
In figure-ground reversal perception for ambiguous figures, figure-ground relationship reverses depending on the position and the range of our attention. Our attention also changes depending on figural features. We propose a neural network model t...
Kamiyama Yutaka, Satoh Shunji, Miyake Shogo
Proceedings of the IEICE General Conference   2004    Mar 2004
SHIMOMURA Masao, SATOH Shunji, MIYAKE Syogo, ASO Hirotomo
The transactions of the Institute of Electronics, Information and Communication Engineers. D-II   86(8) 1244-1253   Aug 2003
認識開始時において網膜上にパターン全体が射影されていない場合でも,注視点を動かすことによりパターンを正しく切出し・認識可能なモデルを提案する.本モデルは,まず提示位置に影響されない特徴を用いてパターンの仮の認識結果(仮定)を生成し,続いて提示位置に依存する特徴を用いて,得られた仮定を検証しながら注視点の移動を行うという動作を繰り返すことにより,適切なパターンの切出し・認識を実現する.また,本モデルの計算機シミュレーションを行うことで,実際にパターン一部分の提示から適切な切出し・認識が可能で...
Yamazaki Susumu, Ito Makoto, Miyake Shogo, Satoh Shunji
IEICE technical report. Neurocomputing   102(731) 125-130   Mar 2003
We construct a model of a radial-arms maze task for rats in which a higher function of memory is required. We formulate a radial-arms maze task as a reinforcement learning problem. In this model, agent can successfully learn depnding on past memor...
SATOH Shunji, MIYAKE Shogo
IEICE technical report. Neurocomputing   102(730) 79-84   Mar 2003
Principles of visual grouping and figure-ground relationship about contours of objects have been identified as closure, symmetry, convexity, parallelism, proximity and so on. These principles should be considered in order to construct an abject re...
Kimura Hiroshi, Satoh Shunji, Sugaya Yoshihiro, Miyake Shogo, Aso Hirotomo
Proceedings of the IEICE General Conference   2003    Mar 2003
Fukui Masaaki, Miyake Shogo, Satoh Shunji
Proceedings of the IEICE General Conference   2003(1)    Mar 2003
Satoh Yoshihiro, Satoh Shunji, Miyake Shogo, Aso Hirotomo
Proceedings of the IEICE General Conference   2003(1)    Mar 2003
Sasaki Yoshinori, Satoh Shunji, Miyake Shogo
Proceedings of the IEICE General Conference   2003(1)    Mar 2003
Imazawa Yoshiro, Satoh Shunji, Miyake Shogo
Proceedings of the IEICE General Conference   2003(1)    Mar 2003
Sasaki Yuji, Kato Tsuyoshi, Satoh Shunji, Miyake Shogo, Aso Hirotomo
Proceedings of the IEICE General Conference   2003(2)    Mar 2003
SATOH Shunji, MIYAKE Shogo
IEICE technical report. Neurocomputing   102(627) 37-42   Jan 2003
A model for visual attention based on scale-space theory is proposed by employing results given by neurophysiological experiments and cognitive phenomena of humans on the vision. The major parts of the model are (a) a model of early vision, and (b...
MORISHIMA K., SATOH S., SHIMOMURA M., MIYAKE S., ASO H.
IEICE technical report. Neurocomputing   101(736) 95-102   Mar 2002
A modular-type neural network is proposed in order to solve pattern recognition problems. Neurons in the hidden layer convert an input space to a curved space by use of polynominal functions, and a neuron in the output makes discriminant surfaces ...
Satoh S., Miyake S., Aso H., Takanaka K.
IEICE technical report. Neurocomputing   100(686) 55-62   Mar 2001
A new recognition model for occluded patterns and texture patterns is proposed. The model includes two types of cells; cells that detect the luminance of input patterns and ones which have Gabor-like receptive fields. Generally, features of unknow...
Shimomura M., Satoh S., Miyake S., Aso H.
IEICE technical report. Neurocomputing   100(686) 63-70   Mar 2001
We present a new neural network model for pattern extraction which is able to efficiently recognize patterns allocated in a plane. The model consists of two units, a saccade unit with low-resolution and a gazing unit with high-resolution. Both uni...
SUDOH Takashi, SATOH Shunji, MIYAKE Shogo, ASO Hirotomo
IEICE technical report. Neurocomputing   100(686) 109-116   Mar 2001
The method of feature extraction in usual pattern recognition depends heavily on objects and is heuristic. In order to get high performance in pattern recognition, feature extraction is an important process. There are only few studies which discus...
SHIMOMURA Masao, SATOH Shunji, MIYAKE Shogo, ASO Hirotomo
Proceedings of the Society Conference of IEICE   2000    Sep 2000
Miyano Y., Satoh S., Aso H., Miyake S.
IEICE technical report. Neurocomputing   99(685) 137-144   Mar 2000
Neocognitron is a neural network model which can recognized noised, distorted and shifted patterns, and several extended models based on the neocognitron are proposed. However, the structures of these models become more and more complex in the pro...
Satoh S., Aso H., Miyake S., Kuroiwa J.
IEICE technical report. Neurocomputing   98(674) 231-238   Mar 1999
A new neural network model which can recognize rotated, distorted, scaled, shifted and noised patterns is proposed. The model is constructed based on psychological experiments in a mental rotation. The model has two types of processes. One is a fe...
SATOH Shunji, KUROIWA Jousuke, ASO Hirotomo, MIYAKE Shogo
Proceedings of the IEICE General Conference   1997(1)    Mar 1997
回転対応型ネオコグニトロンは, Fukushimaによって提案されたネオコグニトロンを, 回転したパターンの認識を可能とするように拡張した階層型神経回路モデルである. 我々は, 任意の回転パターンを認識可能とするために, 閾値制御法を用いた学習法を提案し, 最大Θ=120゜の認識が可能であることを数値シミュレーションで示した (Θは学習パターンからの角度ずれを示している). 本研究では任意の角度に回転させたパターン (例えばΘ=180゜) の認識を可能とするために, S層内の同一細胞面群内...
SATOH Shunji, KUROIWA Jousuke, ASO Hirotomo, MIYAKE Shogo
Proceedings of the Society Conference of IEICE   1996    Sep 1996
回転対応型ネオコグニトロンは,Fukushimaによって提案されたネオコグニトロンを拡張し,回転したパターンの認識をも可能にした階層型神経回路モデルである.回転した特徴を効率よく学習するために動的閾値制御法が提案されているが,本論文では学習パターン数に依存しない動的閾値制御法を提案する.また回転パターンの認識率と計算時間を更に向上させるため,細胞面群の生成と消滅を考慮した新しい学習方法を提案する.
Satoh S., Kuroiwa J., Aso H., Miyake S., Inawashiro S.
IEICE technical report. Neurocomputing   95(598) 263-270   Mar 1996
The neocognitron is a hierarchical neural network to recognize patterns after self organizing, and it is robust for deformation and/or shifting of patterns for recognition. However, it is difficult to recognize greatly rotated patterns. We propose...

Books etc

 
Knowledge-based Intelligent Techniques in Character Recognition
Shunji Satoh (Part:Joint Work, Chapter 3)
CRC Press   1999