Organization
Center for Innovative Teaching and Learning Professor
Title
Professor
External link
YAMASHITA YUKIHIKO
Degree
-
Doctor of Engineering ( 1993.4 Tokyo Institute of Technology )
-
Master of Engineering ( 1985.3 Tokyo Institute of Technology )
Research Interests
-
Architecture of processor to add or change its functions
-
Development system of Teaching materials by collaborative intelligence
-
Information system with semi-machine translation by collaborative intelligence
-
Subband moving image coding
-
機能の追加変更が容易なプロセッサ
-
サブバンド動画像符号化
-
asymmetric kernel method
-
Mahalanobis metric
-
非対称カーネル法
-
集合知に基づく教材作成システム
-
集合知に基づく半機械翻訳情報提供システム
-
マハラノビス計量
Research Areas
-
Informatics / Intelligent informatics
-
Informatics / Theory of informatics
-
Manufacturing Technology (Mechanical Engineering, Electrical and Electronic Engineering, Chemical Engineering) / Communication and network engineering
Education
-
Tokyo Institute of Technology Science of Engineering
- 1985
Country： Japan
-
Tokyo Institute of Technology School of Engineering
- 1983
Country： Japan
Research History
-
東京科学大学（東京工業大学） 教育革新センター
2022.3
Country：Japan
-
Tokyo Institute of Technology
2016.4 - 2022.3
Country：Japan
-
Tokyo Institute of Technology
1995
-
東京工業大学 助手
1989
-
株式会社アイザック 研究員
1988
-
日本原子力研究所 研究員
1985
Professional Memberships
-
IEEE
-
IEEE
-
情報処理学会
-
電子情報通信学会
Papers
-
Gaussian Mixture Bandpass Filter Design for Narrow Passband Width by Using a FIR Recursive Filter Reviewed
Yukihiko YAMASHITA
IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences E106.A ( 10 ) 1277 - 1285 2023.10
-
Geometrically local isotropic independence and numerical analysis of the Mahalanobis metric in vector space Reviewed
Joken Son, Naoya Inoue, Yukihiko Yamashtia
Pattern Recognition Letters 31 ( 8 ) 709 - 716 2010.6
-
Relative Karhunen-Loeve transform Reviewed
Y. Yamashita, H. Ogawa
IEEE Transactions on Signal Processing 44 ( 2 ) 371 - 378 1996
-
理工系分野の大学院生を対象とする大学教員準備講座 : 他大学の教職員へのアンケート調査分析を中心に Reviewed
加藤由香里, 門松怜史, 浅山拓哉, 畠山久, 山下幸彦, 室田真男
高等教育ジャーナル―高等教育と生涯学習― 32 47 - 57 2025.3
-
Worameth CHINCHUTHAKUN, David WINDERL, Alvin C.G. VARQUEZ, Yukihiko YAMASHITA, Manabu KANDA
Journal of JSCE 12 ( 2 ) n/a - n/a 2024
-
Designing Faculty Development Programs by a Team from Different Majors Reviewed
Yukari KATO, Yukihiko YAMASHITA, Hisashi HATAKEYAMA, Toshiya OISHI
International Conference on Computers in Education 2023.12
-
Yukihiko Yamashita, Toru Wakahara
2020 25th International Conference on Pattern Recognition (ICPR) 1128 - 1135 2021.1
-
領域画像マッチングのための目標領域特徴量による初期位置探索法 Reviewed
山下幸彦, 若原徹
電子情報通信学会論文誌D J103-D ( 4 ) 330 - 340 2020.4
-
Anisotropic Guided Filtering Reviewed
Carlo Noel Ochotorena, Yukihiko Yamashita
IEEE Transactions on Image Processing 29 1397 - 1412 2020
-
Image registration using 2D projection transformation invariant GPT correlation Reviewed
Toru Wakahara, Shizhi Zhang, Yukihiko Yamashita
International Workshop on Advanced Image Technology (IWAIT) 2019 5 - 5 2019.3
-
Image matching using GPT correlation associated with simplified HOG patterns Reviewed
Shizhi Zhang, Toru Wakahara, Yukihiko Yamashita
2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA) 1 - 6 2017.11
-
Multi-scale structure-preserving image filtering Reviewed
Carlo Noel Ochotorena, Yukihiko Yamashita
2017 IEEE 19th International Workshop on Multimedia Signal Processing (MMSP) 1 - 6 2017.10
-
ESTIMATING NEAR-SURFACE WIND FIELD BY AERIAL THERMAL IMAGE VELOCIMETRY Reviewed
Motoyuki HIJIKATA, Atsushi INAGAKI, Manabu KANDA, Yukihiko YAMASHITA
Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering) 73 ( 4 ) I_457 - I_462 2017
-
Theoretical criterion for image matching using GPT correlation Reviewed
Shizhi Zhang, Toru Wakahara, Yukihiko Yamashita
2016 23rd International Conference on Pattern Recognition (ICPR) 555 - 560 2016.12
-
Affine-transformation and 2D-projection invariant k-NN classification of handwritten characters via a new matching measure Reviewed
Yukihiko Yamashita, Toru Wakahara
Pattern Recognition 52 459 - 470 2016.4
-
Enhanced GPT Correlation for 2D Projection Transformation Invariant Template Matching Reviewed
Toru Wakahara, Yukihiko Yamashita
Lecture Notes in Computer Science 435 - 445 2015.11
-
Smooth nonnegative matrix and tensor factorizations for robust multi-way data analysis Reviewed
Yokota Tatsuya, Zdunek Rafal, Cichocki Andrzej, Yamashita Yukihiko
Signal Processing 113 234 - 249 2015.8
-
GPT Correlation for 2D Projection Transformation Invariant Template Matching Reviewed
Toru Wakahara, Yukihiko Yamashita
2014 22nd International Conference on Pattern Recognition 3810 - 3815 2014.8
-
k-NN Classification of Handwritten Characters Using a New Distortion-Tolerant Matching Measure Reviewed
Yukihiko Yamashita, Toru Wakahara
2014 22nd International Conference on Pattern Recognition 262 - 267 2014.8
-
Multiple Kernel Learning for Quadratically Constrained MAP Classification Reviewed
Washizawa Yoshikazu, Yokota Tatsuya, Yamashita Yukihiko
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS E97D ( 5 ) 1340 - 1344 2014.5
-
k-NN classification of handwritten characters via accelerated GAT correlation Reviewed
Toru Wakahara, Yukihiko Yamashita
Pattern Recognition 47 ( 3 ) 994 - 1001 2014.3
-
A Quadratically Constrained MAP Classifier Using the Mixture of Gaussians Models as a Weight Function Reviewed
Yokota Tatsuya, Yamashita Yukihiko
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 24 ( 7 ) 1127 - 1140 2013.7
-
“Image compression by using vector quantization and vector-embedded Karhunen-Loeve Transfor
Kiung Park, Yukihiko Yamashita
Proceedings of the 28th International Technical Conference on Circuit/Systems, Computers and Communications 635 - 638 2013.6
-
Heteroscedastic Gaussian based Correction term for Fisher Discriminant Analysis and Its Kernel Extension Reviewed
Yokota Tatsuya, Wakahara Toru, Yamashita Yukihiko
2013 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) 2013
-
Acceleration of GAT correlation for distortion tolerant image matching Reviewed
Toru Wakahara, Yukihiko Yamashita
Proceedings of the 21st International Conference on Pattern Recognition (ICPR 2012), 746 - 749 2012.11
-
k-NN Classification of Handwritten Characters via Accelerated GAT Correlation Reviewed
Toru Wakahara, Yukihiko Yamashita
2012 International Conference on Frontiers in Handwriting Recognition 143 - 148 2012.9
-
A new approach to a maximum a posteriori-based kernel classification method Reviewed
Nopriadi, Yukihiko Yamashita
Neural Networks 33 247 - 256 2012.9
-
Nopriadi, Yukihiko Yamashita
The 2012 International Joint Conference on Neural Networks (IJCNN) 1 - 8 2012.6
-
Linked PARAFAC/CP Tensor Decomposition and Its Fast Implementation for Multi-block Tensor Analysis Reviewed
Yokota Tatsuya, Cichocki Andrzej, Yamashita Yukihiko
NEURAL INFORMATION PROCESSING, ICONIP 2012, PT III 7665 84 - 91 2012
-
Simultaneous Learning of Localized Multiple Kernels and Classifier with Weighted Regularization Reviewed
Naoya Inoue, Yukihiko Yamashita
STRUCTURAL, SYNTACTIC, AND STATISTICAL PATTERN RECOGNITION 7626 354 - 362 2012
-
Affine-Invariant Recognition of Handwritten Characters via Accelerated KL Divergence Minimization Reviewed
Toru Wakahara, Yukihiko Yamashita
2011 International Conference on Document Analysis and Recognition 1095 - 1099 2011.9
-
Quadratically Constrained Maximum a Posteriori Estimation for Binary Classifier Reviewed
Tatsuya Yokota, Yukihiko Yamashita
Lecture Notes in Computer Science 6871 1 - 15 2011
-
Support Vector Machines with Weighted Regularization Reviewed
Tatsuya Yokota, Yukihiko Yamashita
Lecture Notes in Computer Science 7063 471 - 480 2011
-
Kernel Wiener Filter and its Application to Pattern Recognition Reviewed
Hirokazu Yoshino, Chen Dong, Yoshikazu Washizawa, Yukihiko Yamashita
IEEE TRANSACTIONS ON NEURAL NETWORKS 21 ( 11 ) 1719 - 1730 2010.11
-
Multi-template GAT/PAT Correlation for Character Recognition with a Limited Quantity of Data Reviewed
Toru Wakahara, Yukihiko Yamashita
2010 20th International Conference on Pattern Recognition 2873 - 2876 2010.8
-
Subspace Methods with Globally/Locally Weighted Correlation Matrix Reviewed
Yukihiko Yamashita, Toru Wakahara
2010 20th International Conference on Pattern Recognition 4259 - 4262 2010.8
-
Nasharuddin ZAINAL, Toshihisa TANAKA, Yukihiko YAMASHITA
IEICE Transactions on Information and Systems E93-D ( 6 ) 1608 - 1617 2010
-
Blind extraction of global signal from multi-channel noisy observations. Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita
IEEE Trans. Neural Networks 21 ( 9 ) 1472 - 1481 2010
-
Maximum a Posteriori Based Kernel Classifier Trained by Linear Programming Reviewed
Nopriadi, Yukihiko Yamashita
Lecture Notes in Computer Science 493 - 502 2010
-
Moving picture coding by subband transformation and edge adaptive deblocking filter Reviewed
Nasharuddin Zainal, Yukihiko Yamashita
Proc. of 2009 RISP International Workshop on Nonlinear Circuits and Signal Processing 451 - 454 2009.3
-
Blind source extraction using spatio-temporal inverse filter Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita, Andrzej Cichocki
ISCAS: 2009 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, VOLS 1-5 2786 - + 2009
-
Numerical analysis of Mahalanobis metric in vector space Reviewed
Son Joken, Naoya Inoue, Yukihiko Yamashita
Proc. of 19th International Conference on Pattern Recognition (ICPR2008) 2008.12
-
Blind global source extraction from noisy observations Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita, Toshihisa Tanaka, Andrzej Cichocki
2008 RISP international workshop on nonlinear circuits and signal processing 2008.3
-
Pattern recognition by kernel Wiener filter Reviewed
Hirokazu Yoshino, Yukihiko Yamashita
Proc. of Signal Processing, Pattern Recognition, and Applications (SPPRA 2008) 7 - 12 2008.2
-
Yoshikazu Washizawa, Yukihiko Yamashita, Toshihisa Tanaka, Andrzej Cichocki
2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society 5449 - 5452 2007.8
-
Moving picture coding by wavelet transform and edge adaptive deblocking filter, Reviewed
Nasharuddine Zainal, Takashi Ohta, Yukihiko Yamashita
Proc. of the 2007 International Conference on Image Processing, Computer Vision, & Pattern Recognition 25 - 29 2007.6
-
Variable-length lapped transforms with a combination of multiple synthesis filter banks for image coding Reviewed
Toshihisa Tanaka, Yasutaka Hirasawa, Yukihiko Yamashita
IEEE Transactions on Image Processing 15 ( 1 ) 81 - 88 2006.1
-
Asymmetric kernel method and its application to Fisher’s discriminant Reviewed
N. Koide, Y. Yamashita
18th International Conference on Pattern Recognition (ICPR'06) 820 - 824 2006
-
Non-linear Wiener filter in reproducing kernel Hilbert space Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita
18TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 1, PROCEEDINGS 967 - + 2006
-
Maxwell Normal Distribution in a Manifold and Mahalanobis Metric Reviewed
Yukihiko Yamashita, Mariko Numakami, Naoya Inoue
Lecture Notes in Computer Science 604 - 612 2006
-
Kernel projection classifiers with suppressing features of other classes Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita
Neural Computation 18 ( 8 ) 1932 - 1950 2006
-
The generalized lapped pseudo-biorthogonal transform: Oversampled linear-phase perfect reconstruction filterbanks with lattice structures Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
IEEE Transactions on Signal Processing 52 ( 2 ) 434 - 446 2004.2
-
On perfect reconstruction with lost channel data in lapped pseudo-orthogonal transform. Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
2004 12th European Signal Processing Conference, Vienna, Austria, September 6-10, 2004 877 - 880 2004
-
Kernel relative principal component analysis for pattern recognition Reviewed
Y Washizawa, K Hikida, T Tanaka, Y Yamashita
STRUCTURAL, SYNTACTIC, AND STATISTICAL PATTERN RECOGNITION, PROCEEDINGS 3138 1105 - 1113 2004
-
Kernel sample space projection classifier for pattern recognition Reviewed
Yoshikazu Washizawa, Yukihiko Yamashita
Proceedings - International Conference on Pattern Recognition 2 435 - 438 2004
-
A time-varying subband transform with projection-based reconstruction Reviewed
T Tanaka, T Saito, Y Yamashita
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES E86A ( 8 ) 1935 - 1941 2003.8
-
A biorthogonal transform with overlapping and non-overlapping basis functions for image coding Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
IEEE Transactions on Signal Processing 51 ( 3 ) 732 - 743 2003.3
-
Projection-based time-varying subband image coding Reviewed
T Tanaka, T Saito, Y Yamashita
2003 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL 1, PROCEEDINGS 649 - 652 2003
-
A novel class of variable-length lapped transform for image coding Reviewed
T Tanaka, Y Hirasawa, Y Yamashita
2003 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL 3, PROCEEDINGS I 201 - 204 2003
-
An adaptive lapped biorthogonal transform and its application in orientation adaptive image coding Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Signal Processing 82 ( 11 ) 1633 - 1647 2002.11
-
Noise robust oversampled linear phase perfect reconstruction filter bank with a lattice structure Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Proc. of 2002 IEEE International Conference on Image Processing CD-ROM 2002
-
Adaptive transforms with overlapping basis functions for image coding Reviewed
T. Tanaka, Y. Yamashita
Journal of Electronic Imaging 10 ( 3 ) 706 - 719 2001.7
-
The orientation adaptive lapped biorthogonal transform for efficient image coding Reviewed
T Tanaka, Y Yamashita
2001 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL III, PROCEEDINGS CD-ROM 214 - 217 2001
-
BISC (Bus Instruction Set Computer) architecture and BISC-1
Kenichi Maruko, Yukihiko Yamashita
Proc. of 2000 International Technical Conference on Circuits/Systems, Computers, and Communications (ITC-CSCC 2000) 2 935 - 938 2001
-
A biorthogonal transform with overlapping and non-overlapping basis functions for image coding (タイトルが同じだとresearchmapがSP論文と同じと判断するための追記) Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Proc. of 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2001) III 1705 - 1708 2001
-
An iterative deblocking method using 2-D directional FIR filters Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Proc. of 2000 International Technical Conference on Circuits/Systems, Computers, and Communications 1 46 - 49 2000.7
-
Vector-embedded Karhunen-Loeve transform and its application in orientation adaptive coding of images Reviewed
T Tanaka, Y Yamashita
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES E83A ( 6 ) 1257 - 1266 2000.6
-
The orientation adaptive lapped orthogonal transform for image coding Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Proc. of 2000 IEEE International Conference on Image Processing 3 829 - 832 2000
-
Admissibility of memorization learning with respect to projection learning in the presence of noise Reviewed
A Hirabayashi, H Ogawa, Y Yamashita
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS E82D ( 2 ) 488 - 496 1999.2
-
Simulator in a virtual space for autonomous robot and vehicle Reviewed
Ryoichi Abe Hirofumi Isobe, Yukihiko Yamashita
Proc. of 1999 IEEE International Conference on System, Man, and Cybernetics 6 625 - 630 1999
-
Image Coding using vector-embedded Karhunen-Leve transform Reviewed
Toshihisa Tanaka, Yukihiko Yamashita
Proc. of 1999 IEEE International Conference on Image Processing (CD-ROM) ( 25PP2.7 ) 1999
-
Relative Karhunen-Loeve transform method for pattern recognition Reviewed
Y. Yamashita, Y. Ikeno, H. Ogawa
Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170) 2 1031 - 1033 1998.8
-
Quadratic optimization of fixed points of nonexpansive mappings in Hilbert space Reviewed
Isao Yamada, Nobuhiko Ogura, Yukihiko Yamashita, Kohichi Sakaniwa
Numerical Functional Analysis and Optimization 19 ( 1-2 ) 165 - 190 1998
-
New networks for linear programming Reviewed
Yukihiko Yamashita
IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E81-A ( 5 ) 931 - 939 1998
-
ウィーナーフィルタによって雑音を除去する場合の最適標本化 Reviewed
山下幸彦
電子情報通信学会論文誌(A) J81-A ( 2 ) 229 - 236 1998
-
Relative Karhunen-Loeve transform and an optimum sampling Reviewed
Yukihiko Yamashita, Hidemitsu Ogawa
Proc. of the 4th International Conference on Optimization : Techniques and Applications 2 1009 - 1016 1998
-
Paley-Wiener multiresolution analysis and Paley-Wiener wavelet frame Reviewed
Mang Li, Hidemitsu Ogawa, Yukihiko Yamashita
IEICE Trans. on Fundamentals of Electronics, Communications and Computer Science E80-A ( 12 ) 2555 - 2561 1997
-
相対KL変換法によるパターン認識 Reviewed
池野靖行, 山下幸彦, 小川英光
電子情報通信学会論文誌(D-II) J80-D-II ( 2 ) S41-S47 1997
-
General frame multiresolution analysis and its wavelet frame representation Reviewed
Mang Li, Hidemitsu Ogawa, Yukihiko Yamashita
IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences E79-A ( 10 ) 1713 - 1719 1996
-
An Optimal fixed point theorem for nonexpansive operator and its application to set theoritic signal imation-Optimazation with in consistent convex constraints
Isao Yamada, Nobuhiko. Ogura Yukihiko Yamashita, Kohichi Sakaniwa
Proceeding of 1996 International Symposium on Information Theory and its application 736 - 742 1996
-
Image compression by weighted Karhumen-Loeve operator Reviewed
Yukihiko Yamashita, Hidemitsu Ogawa
Proceeding the 13th International Conference on Pattern Recognition E 636 - 640 1996
-
Pattern recognition in the presence of noise Reviewed
D. Liu, Y. Yamashita, H. Ogawa
Pattern Recognition 28 ( 7 ) 989 - 995 1995.7
-
A unified theory of the family of projection filters for signal and image estimation Invited
Yuji Koide, Yukihiko Yamashita, Hidemitsu Ogawa
Systems and Computers in Japan 26 ( 4 ) 95 - 105 1995
-
Methods of convex projections with optimum condition Reviewed
Yukihiko Yamashita
The 2nd International Symposium on Inverse Problems in Engineering Sciences C43 - C44 1994.7
-
信号・画像推定のための射影フィルタ族の統一理論 Reviewed
小出祐司, 山下幸彦, 小川英光
電子情報通信学会論文誌 J77-D-II ( 7 ) 1293 - 1301 1994
-
RELATIVE KARHUNEN-LOEVE OPERATOR Reviewed
Y YAMASHITA, H OGAWA
PROCEEDINGS OF THE 12TH IAPR INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL III - CONFERENCE C: SIGNAL PROCESSING, CONFERENCE D: PARALLEL COMPUTING 3 168 - 170 1994
-
Optimum image restoration and topological invariance Invited
Yukihiko Yamashita, Hidemitsu Ogawa
Systems and Computers in Japan 24 ( 5 ) 53 - 63 1993
-
最適画像復元フィルタと一般逆 Reviewed
山下幸彦, 小川英光
電子情報通信学会論文誌 J75-D-2 5 1992
-
Optimum Image Restoration and Topological Invariance Reviewed
YAMASHITA Yukihiko, OGAWA Hidemitsu
The Transactions of the Institute of Electronics,Information and Communication Engineers. J75-D-2 ( 2 ) 306 - 313 1992
-
各種最適画像復元フィルタの相互関係 Reviewed
山下幸彦, 小川英光
電子情報通信学会論文誌 J75-D-2 5 1992
-
平均射影フィルタの諸性質 Reviewed
山下幸彦, 小川英光
電子情報通信学会論文誌 J74-D-2 ( 2 ) 2 - 149 1991
-
平均射影フィルタによる画像復元 Reviewed
山下幸彦, 小川英光
電子情報通信学会論文誌 J74-D-2 ( 2 ) 2 - 157 1991
-
JT-60の周辺プラズマ監視システム Reviewed
山下幸彦, 荻原徳男
核融合研究 59別冊 298 - 302 1988.3
-
ィーナフィルタ法によるアーベル変換 Reviewed
山下幸彦
JAERI-M 87-206 1987
MISC
-
初歩から学ぶ統計的学習 画像処理で体験してみよう, 映像情報メディア学会編/熊沢逸夫著, オーム社刊, (2012年5月25日発行), A5判, 232頁, 定価:2,940円(税込)
山下 幸彦
映像情報メディア学会誌 : 映像情報メディア 66 ( 10 ) 812 - 812 2012.10
Presentations
-
A Novel Design of Variable-Length Lapped Transforms and Their Applications to Image Coding
TANAKA Toshihisa, HIRASAWA Yasutaka, YAMASHITA Yukihiko
Proceedings of the IEICE General Conference 2003.3 社団法人電子情報通信学会
-
J-82 Convex Projection-Based Adaptive Subband Transforms for Image Coding
SAITO Takateru, TANAKA Toshihisa, YAMASHITA Yukihiko
情報科学技術フォーラム一般講演論文集 2002.9 FIT(電子情報通信学会・情報処理学会)推進委員会
-
J-83 Image Coding by Variable Length Lapped Transform Based on Iterative Method
HIRASAWA Yasutaka, TANAKA Toshihisa, YAMASHITA Yukihiko
情報科学技術フォーラム一般講演論文集 2002.9 FIT(電子情報通信学会・情報処理学会)推進委員会
-
A Noise Robust Generalized Lapped Pseudo-Biorthogonal Transform
TANAKA Toshihisa, YAMASHITA Yukihiko
Proceedings of the Society Conference of IEICE 2002.8 社団法人電子情報通信学会
-
D-11-35 交互射影法に基づく適応サブバンド変換による画像符号化
齋藤 高輝, 田中 聡久, 山下 幸彦
電子情報通信学会総合大会講演論文集 2002.3 電子情報通信学会
-
TANAKA Toshihisa, YAMASHITA Yukihiko
Technical report of IEICE. DSP 2001.4 The Institute of Electronics, Information and Communication Engineers
-
A Lapped Biorthogonal Transform with the Optimal Non-Overlapping Basis Functions
TANAKA Toshihisa, YAMASHITA Yukihiko
ITE Technical Report 2000.12 社団法人映像情報メディア学会
-
A Lapped Biorthogonal Transform with the Optimal Non-Overlapping Basis Functions
TANAKA Toshihisa, YAMASHITA Yukihiko
IPSJ SIG Notes 2000.12 一般社団法人情報処理学会
-
A Lapped Biorthogonal Transform with the Optimal Non-Overlapping Basis Functions
TANAKA Toshihisa, YAMASHITA Yukihiko
IEICE technical report. Communication systems 2000.12 社団法人電子情報通信学会
-
Image Coding Using Orientation Adaptive Transforms with Lapped Basis Functions
TANAKA Toshihisa, YAMASHITA Yukihiko
Proceedings of the Society Conference of IEICE 2000.9 社団法人電子情報通信学会
-
An Iterative Deblocking Method for Orientation Adaptive Coding of Images
HASHIMOTO Kentaro, TANAKA Toshihisa, YAMASHITA Yukihiko
Technical report of IEICE. DSP 2000.4 社団法人電子情報通信学会
-
TANAKA Toshihisa, YAMASHITA Yukihiko
Technical report of IEICE. DSP 2000.4 社団法人電子情報通信学会
-
An Iterative Deblocking Method Using Directional Adaptive Filters
TANAKA Toshihisa, YAMASHITA Yukihiko
Proceedings of the IEICE General Conference 2000.3 社団法人電子情報通信学会
Awards
-
第一回LSI IPデザイン・アワード 開発奨励賞
1999
-
電子情報通信学会米沢ファンダーズ・メダル記念特別賞
1992
-
電子情報通信学会論文賞
1992
Research Projects
-
Theory for unified expression of recognition mechanisms and its application to machine learning
Grant number：20H04206 2020.4 - 2023.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
Grant amount：\16640000 （ Direct Cost: \12800000 、 Indirect Cost：\3840000 ）
-
Theory of operator manifold and its application to pattern recognition
Grant number：17H01760 2017.4 - 2020.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
Yukihiko Yamashita
Grant amount：\17940000 （ Direct Cost: \13800000 、 Indirect Cost：\4140000 ）
In this research, as the basic part of the theory for operator manifolds, we defined the differentials for operators and an operation that corresponds to the differential of differentials contained in an operator. By using them, we provided an operator equation expressing the local isotropic independence. We also provided a solution for a simple case and an equation derived from the fact that the operator is space-invariant.
Furthermore, we have developed methods to obtain a curved coordinate system using local correlation given by the Mahalanobis metric defined by using the equation for the local isotropic independence, to obtain the deformation of an image by combining the canonical correlation and local features of images, to adapt to different domains in pattern recognition by using attributes of patterns, to detect outliers of EEG by using spectral distance as a Riemannian distance, and to smooth image using local statistical features of images. -
Comparative analysis on application of ICT to promote 21st century skills in education in Asia
Grant number：15K01017 2015.4 - 2018.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
Yamaguchi Shinobu
Grant amount：\4680000 （ Direct Cost: \3600000 、 Indirect Cost：\1080000 ）
In collaborating with UNESCO Asia and the Pacific, this research aims at investigating the use of ICT to promote 21st Century Skills (transversal skills) in Asia. Based on the country study in the Asia and the Pacific, it identified advantages and issues in introducing ICT in education settings. Specifically, the analysis was conducted on education policy documents related to transversal skills, teacher training programs and local policies, classroom observations and teacher surveys. Regional conferences were organized to discuss the issues among the experts concerned and to disseminate the study results in Seoul (2016) and Ulanbaatar (2018). Participatory education database was also created at UNESCO Bangkok to promote on-line discussion and share the studies related. As few regional studies exist in the Asian region, and thus, this research is expected to contribute to the monitoring of SDG4 promoting quality education in the region.
-
Manifold signal processing theory concerning metric structure and its application to biological signal processing
Grant number：26280054 2014.4 - 2017.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
Yukihiko Yamashita
Grant amount：\16380000 （ Direct Cost: \12600000 、 Indirect Cost：\3780000 ）
For the manifold signal processing theory concerning the metric structure, we constructed a method to solving locally isotropically independence equations by using Gauss normalized kernel functions, and defined an operator equation to express the locally isotropically independent change. We developed a semi-supervised learning method for metrics when the number of training data is not enough. We proposed a method to assign a pattern to a subspace, and a metric suitable to discrimination for the Grassmann manifold. We also developed an anomaly detection method with images by combining the direct estimation for the ration between probability density functions and the convolutional network, and a distance for hand-written character recognition that provides the best recognition rate among those based on k-nearest neighbor method. Furthermore, we improved the performance of recognition of motional intention that enables paretics caused by apoplexy, etc. to reconstruct motor function.
-
signal manifold by extended kernel method and its application
Grant number：23300069 2011.4 - 2014.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
YUKIHIKO Yamashita, SUGIYAMA Masashi, TANAKA Toshihisa, WASHIZAWA Yoshikazu
Grant amount：\20670000 （ Direct Cost: \15900000 、 Indirect Cost：\4770000 ）
In order to improve the performance of signal processing and pattern recognition, it is necessary to obtain a proper metric for signals and patterns and develop processing and recognition methods based on the metric.
According to the concept, we improve the theory of Mahalanobis metric, develop a calculation method of Mahalanobis metric using kernel functions, and provide criteria and calculation methods that can obtain a equivalent result with the maximum posteriori method without estimating a posteriori probability. We also propose a method to directly estimate the L2 distance between two probability density functions and show its properties with respect to convergence and bias. We develop an algorithm to calculate the sample average on Stiefel manifold, and the adaptive kernel principle component analysis that can extract features with a small number of basis functions. The advantages of developed methods are shown by experimental results. -
Machine learning theory based on structure of signal space and its application
Grant number：18300057 2006 - 2008
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
YAMASHITA Yukihiko, SUGIYAMA Masashi, TANAKA Toshihisa, WASHIZAWA Yoshikazu
Grant amount：\16810000 （ Direct Cost: \14500000 、 Indirect Cost：\2310000 ）
-
Time-Varying Filter Banks and Their Applications
Grant number：15560344 2003 - 2005
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
TANAKA Toshihisa, YAMASHITA Yukihiko
Grant amount：\3500000 （ Direct Cost: \3500000 ）
A novel class of time-varying subband transforms and its application in image coding are introduced. We construct the framework in which an analysis subband transform is chosen from a given set of analysis transforms and perfect reconstruction is achieved. To this end, we suggest the reconstruction method from the transformed coefficients by applying the theory of convex projections (POCS). We introduce convex sets for perfect reconstruction from the transformed vector. We further propose other convex sets which assure smoothness of plane regions for image coding applications. We show several image coding examples. The proposed coding method is an extension of conventional one with multiple block-based transforms. At each time instance (block), the transform to be applied is chosen from a given set of orthogonal subband transforms according to a certain criterion. Experimental results show that the use of multiple subband transforms leads to the improvement of coding performance compared to the use of single transform even though there exists side information.
-
Kernel relative principal component analysis for pattern recognition
Grant number：15500101 2003 - 2005
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
YAMASHITA Yukihiko, TANAKA Toshihisa
Grant amount：\3600000 （ Direct Cost: \3600000 ）
Since the accuracy of pattern recognition is not enough, this research is done for making pattern recognition more accurate by applying the kernel method, which can realize complicated discrimination boundary with a non-linear mapping, to the relative principal component analysis, which is proposed by our research group and can extract principal components under the effect of another signal which has to be suppressed.
The results of this research are as follows.
1) The theorem of the kernel principal component analysis (KRPCA) was established and its closed form that can provide the solution of KRPCA with a kernel function and samples were obtained.
2) A simple closed form of KRPCA for a non-singular kernel Gram matrix was provided. Then, KRPCA can be realized by computer more simply.
3) By computer simulation with standard recognition problems, the advantages of KRPCA were shown.
4) The kernel sample space method and the one with suppression feature that are the KRPCAs in a special case were proposed. Its closed forms were provided. Although they are restricted version of KRPCA, they achieved similar performance to KRPCA. Since their solution are very simple, the theory of additive learning for them was provided.
5) The existing kernel method uses a kind of nonlinear function. By extending it, we proposed the theory of asymmetric kernel method that uses two kinds of nonlinear functions. It will be a basis for future progress of kernel method. A classifier by using it was constructed and its advantages were shown.
6) For other researches, we provided a new theory of subband filter bank, showed its advantage in image coding, and researches a computer architecture for recognition and signal processing. -
相対Karhunen-Loeve変換によるパターン認識
Grant number：08780332 1996
日本学術振興会 科学研究費助成事業 奨励研究(A)
山下 幸彦
Grant amount：\1000000 （ Direct Cost: \1000000 ）
工学的なパターン認識の処理過程は、観測、前処理、特徴抽出、識別の4過程に分けることができる。また、部分空間法と呼ばれる一連の識別手法においては、識別過程においても、それぞれのカテゴリーの特徴(類内特徴)の抽出を行なう。そして、入力パターンが最も多く含む類内特徴に対応するカテゴリーを識別結果とする。このような認識手法では、類内特徴抽出は他のカテゴリーと区別するための特徴を抽出する過程であり、重要な役割を担っている。代表的な類内特徴抽出法であるCLAFIC(CLAss-Featuring Information Compression)法では、自分のカテゴリーに平均的に多く含まれる特徴を抽出する。しかしながら、そのような特徴は他のカテゴリーと区別するために適したものであるとは限らない。識別に適した特徴とは、自分のカテゴリーに含まれ、他のカテゴリーには含まれないような特徴である.このような観点から、まず、雑音を抑制しながら信号の特徴を抽出するために開発された相対Karhunen-Loeve変換の評価基準を拡張し、着目するカテゴリーの特徴の中で他のカテゴリーのパターンに対してはその大きさが小さくなるような特徴を抽出することができる評価基準を求めた。そして、その評価基準を満たす類内特徴抽出法(相対KL変換法)を具体的に計算する方法を開発した。また、計算機による手書き文字認識実験を行ない、相対KL変換法によってCLAFIC法よりも高い認識精度を得ることができた。このときの実験から得られた認識精度は十分実用的なものであった。さらに、相対KL変換法のための行列を設計するための計算時間は、CLAFIC法の2倍程度であり、1つのパターンに対する処理時間はCLAFIC法とほぼ同程度であった。この実験によって、相対KL変換法は文字認識のための類内特徴抽出法として有効であることを示すことができた。
-
最適条件付き凸射影法による画像修復
Grant number：07780307 1995
日本学術振興会 科学研究費助成事業 奨励研究(A)
山下 幸彦
Grant amount：\1000000 （ Direct Cost: \1000000 ）
最適条件付き凸射影フィルタを用いた、雑音を含んだ画像から雑音を除去する画像修復手法を確立し、計算機実験によってその有効性を示した。本研究の成果を次に記す。
1.画像を表現するために適した凸集合を考案した。そのために、計算速度を高めるためばかりではなく、画像の局所的な性質を捉えるために画像のブロック化を行なう方法を提案した。さらに、画像ブロックの重複部分を作ることによって、上記方法で生じるブロック歪みを軽減する方法を提案した。
2.雑音を除去するための最適化基準を考案した。この最適化基準と上記の凸集合を組み合わせることによって、画像の劣化が少ない画像修復法提案した。さらに、最適条件付き凸射影法によってその画像修復を実現する理論を構築した。
3.画像に人工的に作成した白色雑音、インパルス性雑音等の各種雑音を加え、設備備品として購入しパーソナルコンピュータとワークステーションに設備備品として購入したモニタを用いて修復実験を行ない、本研究で提案した手法の有効性を確認した。
4.また、この理論および最適条件付き凸射影法をコンピュータトモグラフィの分野へも応用し、制限角投影からの再構成における理論を構築するともに、購入した設備備品よってその有効性を確認した。 -
Study about a construction of optimally generalizing neural networks
Grant number：06452399 1994 - 1995
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for General Scientific Research (B)
OGAWA Hidemitsu, YAMASHITA Yukihiko
Grant amount：\6000000 （ Direct Cost: \6000000 ）
In this sutdy, we first formalized the problem of training a neural network as one of an inverse problem in function approximation. Next, we provided necessary and sufficient conditions for optimal generalization capability in terms of the number of hidden units, the basis functions, and the weights connected to each hidden unit. Furthermore, we gave a methodology to construct neural networks with optimal generalization capability.
From this methodology, we can see that are an infinite numbers of neural networks with the same generalization capability. From among this infinite number, we specified the ones which are most robust with respect to some kinds of faults which may occur in actual usage. Concretely, we gave methods to decide weights in neural networks which optimally suppress the influences of each of the following faults : a proportional errors in the weights, a connection fault, and a stuck-at gamma fault. Moreover, we gave the methods to decide not only weights but also basis functions for the hidden units which optimally suppress the above three faults.
Next we constructed a method for incremental learning in which only the current network and one new datum are used to obtain a new network at each step, while maintaining the property that the neural network is optimally generalizing with respect to all of the data learned so far. We think our results have the potential for being applied to the problem of active learning.
Moreover, we gave a solution to the problem of over-learning which occurs in training of neural networks using the error-backpropagation algorithm, i.e., we introduced the concept of admissibility defined by relation-ship between two learning criteria. According to the admissibility, we devised methods for choosing training data to prevent over-leaning. -
凸射影フィルタによる線形計画法に関する研究
Grant number：05780234 1993
日本学術振興会 科学研究費助成事業 奨励研究(A)
山下 幸彦
Grant amount：\900000 （ Direct Cost: \900000 ）
線形計画法の解法について、次のような成果を得た。
1.凸射影法を最大勾配法の考え方を使って拡張した方法を提案した。そこで使われる点列が、必ず収束することを証明し、線形計画法の解を与えることを示した。また、この解法の1スッテプの収束が単なる凸射影法を使った線形計画法の解法よりも速いことを証明した。
2.凸射影法を共役勾配法の考え方を使って拡張した方法を提案した。そこで使われる点列が、必ず収束することを証明し、線形計画法の解を与えることを示した。この解法の1スッテプの収束が単なる凸射影法を使った解法や最大勾配法を使って拡張した凸射影法による解法よりも速いことを証明した。
3.線形計画法の拡張である凸2次形式の最小化問題の解を、凸射影法を拡張した方法によって与える方法を提案した。また、そこで使われる点列が必ず収束することを証明した。
4.ここで提案した方法のアルゴリズムは簡単なので、並列計算に適していることを示した.
5.1および2で提案した解法が、実際に収束すること、1の解法の収束が単なる凸射影法を使った線形計画法の解法よりも速いこと、2の解法の収束が1の解法や単なる凸射影による解法よりも速いことを、計算機実験で確かめた。 -
A study on optimal generalizing learning schema for neural networks based on theories of image processing filters
Grant number：02452155 1990 - 1991
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for General Scientific Research (B)
OGAWA Hidemitsu, YAMASHITA Yukihiko, KUMAZAWA Itsuo
Grant amount：\5700000 （ Direct Cost: \5700000 ）
A theory for neural network learning which is very effective for analyzing the generalization ability and the overlearning problem is developed. This theory is based on the image restoration theories proposed by the head investigator of this project. Although the problems of generalization and image restoration seem to have nothing in common, we have shown that both problems can be dealt with under the same methodology if we formalize them as a kind of inverse problem. This novel approach provides an analytical and quantitative method for the problems of generalization and over-learning which have been so far treated qualitatively. Followings are the major results obtained in this project.
・A new framework of generalization which can extract general structures among training samples is developed based on the image restoration theories.
・We analyze the generalization ability of the back-propagation using the framework.
・We provide a way of choosing, training samples which does not cause the over-learning problem and gives an optimal generalizing ability.
・Above theoretical results are examined by some computer simulations. -
A Research for Novel Computerized Topography Technologies for Moving Objects.
Grant number：63460133 1988 - 1989
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for General Scientific Research (B)
OGAWA Hidemitsu, IMIYA Atsushi, YAMASHITA Yukihiko, KUMAZAWA Itsuo
Grant amount：\7400000 （ Direct Cost: \7400000 ）
The Computerized Topography is said to be one of the most revolutional medical technologies for the last decade. However, the ability of the algorithms proposed so far is restricted only to the reconstruction of two dimensional cross section images and the reconstruction of 3-dimensional images has been carried out by applying those algorithms independently to each section image of the 3-D objects. It has been thought to be impossible to reconstruct moving objects from their projections by using those algorithms, because the independent application of those algorithms to each section of a object is computationally too much burden and can not make use of the 3-D correlations among the section images of a moving object for saving computation power or for restoring blurred and noisy projections In our research, we proposed novel reconstruction algorithms to eliminate above problems. (1) We proposed a direct reconstruction algorithm for cone beam projection schema by which we can collect 3-D data faster than traditional methods. This algorithm was also designed to have the capability of restoring blurred data. (2) We developed a noise canceling technique which uses the redundant structure among projections. This technique is effective when the data is noisy due to the fast rotation of the X-ray detector. (3) We proposed a new filter which is called a partial projection filter. This filter utilizes the 3-D correlation of images to save the number of samplings of the data or to improve the quality of reconstructed images.
-
Image Processing
Grant type：Competitive
-
画像処理
Grant type：Competitive