Organization
School of Engineering Researcher
Title
Researcher
External link
UTO KUNIAKI
Degree
-
Doctor of Engineering ( Tokyo Institute of Technology )
Research Interests
-
パターン認識
-
Pattern recognition
Research Areas
-
Informatics / Perceptual information processing
Education
-
Tokyo Institute of Technology Interdisciplinary Science and Engineering
- 1999
Country： Japan
-
Tokyo Institute of Technology Graduate School, Division of Integrated Science and Engineering
- 1999
-
Tokyo Institute of Technology School of Engineering
- 1995
Country： Japan
Research History
-
-:Tokyo Institute of Technology Interdisciplinary Graduate Schlol of Science and Engineering Research Associate
2001
-
-:東京工業大学 大学院総合理工学研究科 助手
2001
Professional Memberships
-
電子情報通信学会
-
Inc.
-
Inc.
-
the Institute of Electronics
-
Information and Communication Engineers
-
the Institute of Electrical and Electronics Engineers
-
the Institute of Electrical and Electronics Engineers
Papers
-
Coastal observation using new hyperspectral imager for UAVs Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
International Geoscience and Remote Sensing Symposium (IGARSS) 2017- 3614 - 3617 2017.12
-
Leaf parameter estimation based on shading distribution in leaf scale hyperspectral images Reviewed
Kuniaki Uto, Yukio Kosugi
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2013- 2017.10
-
Development of hyperspectral imaging system using optical fiber bundle and swing mirror Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2015- 2017.10
-
Development of lightweight hyperspectral imaging system for UAV observation Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2014- 2017.10
-
Measurement of a coastal area by a hyperspectral imager using an optical fiber bundle, a swing mirror and compact spectrometers Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2017.10
-
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 9 ( 9 ) 3909 - 3925 2016.9
-
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 9 ( 2 ) 861 - 875 2016.2
-
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi, Teruhisa Komatsu
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 9 ( 2 ) 861 - 875 2016.2
-
Kuniaki Uto, Yukio Kosugi, Genya Saito
2016 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 6914 - 6917 2016
-
Development of hyperspectral imaging sensor which mounted on unmanned aircraft for coastal zone studies
Genya Saito, Haruyuki Seki, Kuniaki Uto, Yukio Kosugi, Teruhisa Komatsu
ACRS 2015 - 36th Asian Conference on Remote Sensing: Fostering Resilient Growth in Asia, Proceedings 2015
-
SEMI-SUPERVISED HYPERSPECTRAL MANIFOLD LEARNING FOR REGRESSION Reviewed
Kuniaki Uto, Yukio Kosugi, Genya Saito
2015 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 9 - 12 2015
-
Kuniaki Uto, Yukio Kosugi, Genya Saito
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 7 ( 6 ) 2583 - 2599 2014.6
-
HYPERSPECTRAL SUBSPACE LEARNING OF FOREST PHENOLOGY UNDER ORDER CONSTRAINTS Reviewed
Kuniaki Uto, Yukio Kosugi, Genya Saito
2014 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 4652 - 4655 2014
-
Development of hyperspectral imaging sensor, which mounted on UAV for environmental study at coastal zone
Genya Saito, Haruyuki Seki, Kuniaki Uto, Yukio Kosugi, Teruhisa Komatsu
35th Asian Conference on Remote Sensing 2014, ACRS 2014: Sensing for Reintegration of Societies 2014
-
Leaf Parameter Estimation Based on Leaf Scale Hyperspectral Imagery Reviewed
Kuniaki Uto, Yukio Kosugi
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 6 ( 2 ) 699 - 707 2013.4
-
Characterization of Rice Paddies by a UAV-Mounted Miniature Hyperspectral Sensor System Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 6 ( 2 ) 851 - 860 2013.4
-
Development of mounted hyperspectral sensor on UAV for environmental and agricultural studies
Genya Saito, Haruyuki Seki, Kuniaki Uto, Yukio Kosugi
34th Asian Conference on Remote Sensing 2013, ACRS 2013 4 2884 - 2887 2013
-
DEVELOPMENT OF UAV-MOUNTED MINIATURURE HYPERSPECTRAL SENSOR SYSTEM FOR AGRICULTURAL MONITORING Reviewed
Kuniaki Uto, Haruyuki Seki, Genya Saito, Yukio Kosugi
2013 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 4415 - 4418 2013
-
Estimation of Lambert parameter based on leaf-scale hyperspectral images using dichromatic model-based PCA Reviewed
Kuniaki Uto, Yukio Kosugi
INTERNATIONAL JOURNAL OF REMOTE SENSING 34 ( 4 ) 1386 - 1412 2013
-
Shinya ODAGAWA, Yukio KOSUGI, Genya SAITO, Kuniaki UTO, Yuka SASAKI, Kunio ODA, Masatane KATO
Journal of the Japan society of photogrammetry and remote sensing 51 ( 5 ) 270 - 284 2012.9
-
EXTRACTION OF LAMBERT PARAMETER FROM LEAF SCALE HYPERSPECTRAL IMAGES Reviewed
Kuniaki Uto, Yukio Kosugi
2012 4TH WORKSHOP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING (WHISPERS) 2012
-
RICE GROWTH STATE ESTIMATION BY HYPERSPECTRAL MANIFOLD LEARNING Reviewed
Kuniaki Uto, Takahiro Harano, Yukio Kosugi
2012 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 4178 - 4181 2012
-
Hyperspectral Band Selection for Human Detection Reviewed
Kuniaki Uto, Yukio Kosugi, Toru Murase, Sigenori Takagishi
2012 IEEE 7TH SENSOR ARRAY AND MULTICHANNEL SIGNAL PROCESSING WORKSHOP (SAM) 501 - 504 2012
-
Hyperspectral band selection for human detection Reviewed
Kuniaki Uto, Yukio Kosugi, Toru Murase, Sigenori Takagishi
Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop 501 - 504 2012
-
Human detection based on active infrared illumination Reviewed
Kuniaki Uto, Haruyuki Seki, Yukio Kosugi, Toru Murase, Sigenori Takagishi
Proceedings - 2012 IEEE Global Humanitarian Technology Conference, GHTC 2012 47 - 52 2012
-
Extraction of Lambert parameter from leaf scale hyperspectral images Reviewed
Kuniaki Uto, Yukio Kosugi
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2012
-
Analyses for soil salinity at strawberry farm by the tsunami of 2011 off the pacific coast earthquake of tohoku using multitemporal TerraSAR-x data
Genya Saito, Kuniaki Uto, Shigeru Kakumoto, Yukio Kosugi, Haruyuki Seki, Shinya Odagawa, Hideki Shimamura
33rd Asian Conference on Remote Sensing 2012, ACRS 2012 3 2452 - 2455 2012
-
Tree disease detection based on hyperspectral manifold learning of phenological transition on forest
Kuniaki Uto, Yukio Kosugi
32nd Asian Conference on Remote Sensing 2011, ACRS 2011 1 386 - 391 2011
-
LOW-ALTITUDE HYPERSPECTRAL OBSERVATION OF PADDY USING RADIO-CONTROLLED HELICOPTER Reviewed
Yukio Kosugi, Shinji Mukoyama, Yuji Takabayashi, Kuniaki Uto, Kunio Oda, Genya Saito
2011 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 1748 - 1751 2011
-
Spectral Evaluations of Color Rendering Properties of LED Light Sources by Using Hyper Spectral Sensor Reviewed
Haruyuki Seki, Kuniaki Uto, Yukio Kosugi
IDW'11: PROCEEDINGS OF THE 18TH INTERNATIONAL DISPLAY WORKSHOPS, VOLS 1-3 1439 - 1442 2011
-
EVALUATION OF PADDY YIELD AND PROTEIN ESTIMATION METHODS BASED ON VARIOUS VEGETATION INDICES, NDSI AND PLS USING AN AIRBORNE HYPERSPECTRAL SENSOR AISA IN SHONAI PLAIN, YAMAGATA, JAPAN Reviewed
Shinya Odagawa, Kuniaki Uto, Yukio Kosugi, Genya Saito, Yuka Sasaki, Kunio Oda, Masatane Kato
2011 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) 1926 - 1929 2011
-
Band selection for Japanese oak wilt extraction in autumnal tints of forest based on NWI Reviewed
Kuniaki Uto, Toshihiro Massaki, Yukio Kosugi, Genya Saito, Toshinari Ogata
Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing 2011
-
Normalized Wilt Index based on Visible/Near-Infrared Hyperspectral Analysis of Japanese Oak Wilt Reviewed
Kuniaki UTO, Yukio KOSUGI, Toshinari OGATA, Shinya ODAGAWA
Journal of the Japan society of photogrammetry and remote sensing 49 ( 5 ) 294 - 309 2010.9
-
Statistical analysis of growth levels of rice paddy based on hyperspectral imagery with high spatial resolution Reviewed
Kuniaki Uto, Yukio Kosugi, Jiro Sasaki
2nd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS 2010 - Workshop Program 2010
-
Evaluation of paddy yield and quality estimation methods based on various vegetation indices, NDSI and PLS using BRDF-corrected airborne hyperspectral data Reviewed
Shinya Odagawa, Masatane Kato, Tomoyuki Suhama, Jiro Sasaki, Uto Kuniaki, Yukio Kosugi, Genya Saito
International Geoscience and Remote Sensing Symposium (IGARSS) 3 III565 - III568 2009
-
EVALUATION OF OAK WILT INDEX BASED ON GENETIC PROGRAMMING Reviewed
Kuniaki Uto, Yukio Kosugi, Toshinari Ogata
2009 FIRST WORKSHOP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING: EVOLUTION IN REMOTE SENSING 267 - + 2009
-
LEAF AREA INDEX ESTIMATION FROM HYPERSPECTRAL DATA USING GROUP DIVISION METHOD Reviewed
Taro Asano, Yukio Kosugi, Kuniaki Uto, Naoko Kosaka, Shinya Odagawa, Kunio Oda
2009 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-5 3197 - + 2009
-
Hyperspectral image classification of grass species in northeast Japan Reviewed
Sildomar T. Monteiro, Kuniaki Uto, Yukio Kosugi, Kunio Oda, Yoshiyuki Iino, Genya Saito
International Geoscience and Remote Sensing Symposium (IGARSS) 4 ( 1 ) IV399 - IV402 2008
-
Development of rice yield estimation method based on spaceborne hyperspectral data: preliminary study using airborne hyperspectral data Reviewed
Shinya Odagawa, Mastane Kato, Tomoyuki Suhama, Jiro Sasaki, Kuniaki Uto, Yukio Kosugi, Genya Saito
International Geoscience and Remote Sensing Symposium (IGARSS) 3 ( 1 ) III716 - III719 2008
-
Hyperspectral analysis of Japanese Oak wilt to determine normalized wilt index Reviewed
Kuniaki Uto, Yuji Takabayashi, Yukio Kosugi, Toshinari Ogata
International Geoscience and Remote Sensing Symposium (IGARSS) 2 ( 1 ) II295 - II298 2008
-
Self-organizing property of nonlinear mapping for change detection Reviewed
Kuniaki Uto, Yukio Kosugi
IGARSS: 2007 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-12 1962 - 1965 2007
-
Segmentation of High Resolution Satellite Images based on Spatial Pattern Dynamics Model Reviewed
Kuniaki Uto, Yasumasa Imai, Yukio Kosugi
2006 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-8 2814 - 2817 2006
-
Development of a low-altitude hyperspectral imaging system for measuring ground truth in agricultural fields Reviewed
Yohei Minekawa, Takayuki Edanaga, Kuniaki Uto, Yukio Kosugi, Kunio Oda
2006 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-8 2052 - + 2006
-
Monitoring of Bacterial Pustule on Soybean by Neural Network Using Hyperspectral Data Reviewed
Naoko Kosaka, Yohei Minekawa, Kuniaki Uto, Yukio Kosugi, Kunio Oda, Genya Saito
2006 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, VOLS 1-8 3879 - + 2006
-
ICA-aided mixed-pixel analysis of hyperspectral data in agricultural land Reviewed
Naoko Kosaka, Kuniaki Uto, Yukio Kosugi
IEEE Geoscience and Remote Sensing Letters 2 ( 2 ) 220 - 224 2005.4
-
Analysis of salt-damaged paddy field using SPOTS satellite images in Yamagata Prefecture Reviewed
Shoichi Hoshino, Naoko Kosaka, Kuniaki Uto, Youhei Minekawa, Yukio Kosugi, Genya Saito, Kunio Oda
International Geoscience and Remote Sensing Symposium (IGARSS) 3 1830 - 1833 2005
-
Salt-damaged paddy fields analyses using high-spatial-resolution hyperspectral imaging system Reviewed
Yohei Minekawa, Kuniaki Uto, Naoko Kosaka, Yukio Kosugi, Ho Ando, Yuka Sasaki, Kunio Oda, Shizuka Mori, Genya Saito
International Geoscience and Remote Sensing Symposium (IGARSS) 3 2153 - 2156 2005
-
Synthesis and analysis of periodically distributed satellite image components by independent component analysis
Kuniaki Uto, Naoko Kosaka, Yukio Kosugi
International Geoscience and Remote Sensing Symposium (IGARSS) 4 2431 - 2434 2004
-
Ill-posed problems arising in image-guided navigation systems and a network realization based on spline interpolation Reviewed
Kuniaki Uto, Yukio Kosugi
Systems and Computers in Japan 29 ( 10 ) 36 - 45 1998
MISC
-
B-2-39 Rice Blast Detection using a Hyperspectral Group Division Method
Asano Taro, Kosugi Yukio, Utsu Kuniaki, Odagawa Shinya, Sasaki Jiro
Proceedings of the IEICE General Conference 2009 ( 1 ) 296 - 296 2009.3
-
Extraction of movement-related potentials from EEG based on DT-aided independent component analysis
K Uto, K Hibi, Y Kosugi
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS E86D ( 8 ) 1464 - 1469 2003.8
-
Real-time extracton of readiness-potentials from EEG
K Uto, K Hibi, Y Kosugi
PROCEEDINGS OF THE 23RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-4 23 1197 - 1200 2001
-
Segmented representation of approximation surface
K Uto, Y Kosugi
ICONIP'98: THE FIFTH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING JOINTLY WITH JNNS'98: THE 1998 ANNUAL CONFERENCE OF THE JAPANESE NEURAL NETWORK SOCIETY - PROCEEDINGS, VOLS 1-3 Proc. ICONIP'98 75 - 78 1998
-
Acquisition of the unique solution based on spline network
K Uto, Y Kosugi
IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE Proc. IJCNN'98 349 - 352 1998
-
画像誘導ナビゲーションにおける不適切性とスプライン補間に基づくネットワークの検討
宇都有昭
電子情報通信学会論文誌 J81-D-II ( 2 ) 361 - 369 1998
-
Ill-Posed Problems Arising in Image-Guided Navigation Systems and a Network Realization Based on Spline Interpolation
Kuniaki Uto
J81-D-II ( 2 ) 361 - 369 1998
Presentations
-
DEVELOPMENT OF RICE YIELD ESTIMATION METHOD BASED ON SPACEBORNE HYPERSPECTRAL DATA
PRELIMINARY STUDY USING AIRBORNE HYPERSPECTRAL DATA 2008
-
Low-Altitude Hyperspectral Imaging of Naruko Integrated Field for the Interpretation of High-Altitude Observations
Abst. of 6th Internationa Symposium on Integrated Field Science A2 2008
-
指標NWIによるASRER画像からのナラ枯れ発生域抽出
日本写真測量学会平成20年度年次学術講演会講演論文集 2008
-
ナラ枯れのハイパースペクトル・リモートセンシングの可能性を探る
第119回日本森林学会大会抄録 2008
-
Low-Altitude Hyperspectral Imaging of Naruko Integrated Field for the Interpretation of High-Altitude Observations
Abst. of 6th Internationa Symposium on Integrated Field Science A2 2008
-
HYPERSPECTRAL IMAGE CLASSIFICATION OF GRASS SPECIES IN NORTHEAST JAPAN
Proc. IGARSS2008 2008
-
ハイパースペクトル画像を用いた潮風被害の推定
第26回日本シミュレーション学会大会抄録 2007
-
変化域抽出精度の向上と大規模処理の高速化の研究
大都市大震災軽減化特別プロジェクト・ｽV 平成18年度報告書（防災科研編） 2007
-
クレーン搭載ハイパースペクトルセンサデータによる大豆葉焼病の推定
写真測量とリモートセンシング 2007
-
短波長赤外領域スペクトル情報を用いた人肌領域抽出に関する研究
写真測量とリモートセンシング 2007
-
HYPERSPECTRAL IMAGE CLASSIFICATION OF GRASS SPECIES IN NORTHEAST JAPAN
Proc. IGARSS2008 2008
-
航空機搭載ハイパースペクトルセンサを用いたナラ枯れの観測と解析
第120回日本森林学会抄録 2009
-
Leaf Area Index Estimation from Hyperspectral Data Using A Group Division Method
Proc. IGARSS 2009 2009
-
赤外ハイパースペクトル情報を用いた路上障害物認識に関する基礎的検討
日本写真測量学会年次学術講演会論文集 2009
-
ハイパースペクトルデータにおけるグループ分割手法を用いたいもち病検出に関する研究
電子情報通信学会2009年総合大会抄録 2-pages(CD-ROM)(2009) 2009
-
グループ分割手法を用いたハイパースペクトルデータの葉面積指数推定
第１６回リモートセンシングフォーラム抄録 2009
-
DEVELOPMENT OF RICE YIELD ESTIMATION METHOD BASED ON SPACEBORNE HYPERSPECTRAL DATA
PRELIMINARY STUDY USING AIRBORNE HYPERSPECTRAL DATA 2008
-
Discrimination of Obstacles on the Road Using Hyper-spectral Dada
Proc. ACRS (Beijing 2009) 2009
-
Evaluation of oak wilt index based on genetic programming
Proc. First Workshop on Hyperspectral Image and Signal Processing 2009
-
Evaluation of Paddy Yield and Quality Estimation Methods Based on Various Vegetation Indices, NDSI and PLS Using BRDF-Corrected Airbone Hyperspectral Data
Proc. IGARSS 2009 2009
-
Discrimination of Obstacles on the Road Using Hyper-spectral Dada
Proc. ACRS (Beijing 2009) 2009
-
Evaluation of oak wilt index based on genetic programming
Proc. First Workshop on Hyperspectral Image and Signal Processing 2009
-
Evaluation of Paddy Yield and Quality Estimation Methods Based on Various Vegetation Indices, NDSI and PLS Using BRDF-Corrected Airbone Hyperspectral Data
Proc. IGARSS 2009 2009
-
Leaf Area Index Estimation from Hyperspectral Data Using A Group Division Method
Proc. IGARSS 2009 2009
-
Discrimination of Obstacles on the Road Using Hyper-spectral
Asian Journal of Geoinformatics 2010
-
Discrimination of Obstacles on the Road Using Hyper-spectral
Asian Journal of Geoinformatics 2010
-
グループ分割手法によるハイパースペクトルデータからの葉面積指数推定
写真測量とリモートセンシング（印刷中）
-
無人ヘリ搭載ハイパースペクトルセンサによる水稲の植生計測
システム農学会2007年度秋季大会講演論文集 2007
Research Projects
-
Hierarchical Cordinated Control of Multiple Robots for Smart/Autonomous Agriculture and Proof of Concept
Grant number：24K00906 2024.4 - 2027.3
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)
Grant amount：\17550000 （ Direct Cost: \13500000 、 Indirect Cost：\4050000 ）
-
マルチエージェント深層学習による音声因子分解
Grant number：19H04133 2019.4 - 2022.3
日本学術振興会 科学研究費助成事業 基盤研究(B)
篠田 浩一, 井上 中順, 岩野 公司, 宇都 有昭
Grant amount：\17420000 （ Direct Cost: \13400000 、 Indirect Cost：\4020000 ）
前年度に「(A)音韻性と雑音の分離」と「(B)音韻性と話者性の分離」について、ある程度の性能向上が達成されたため、今年度はこれらを活用して、「(D)音源分離」の応用に着手した。より具体的には、これまで開発してきた波形ベースの音声信号処理システムをベースに、複数の話者の音声を分離する音源分離のフレームワークをまず構築した。そして、「(A)音韻性と雑音の分離」の成果をもとに、音源から雑音を除去する仕組みをこのシステムに追加し、雑音の影響に対して頑健な音声分離システムを構築した。既存のデータベースに雑音を重畳したデータを構築し、それを用いて評価を行った。従来法に比べ有意に高い性能を得た。この成果は、信号処理関連の国際会議IEEE APSIPA 2021に採択され、発表した。さらに「(E)言語認識、感情認識」において、まず感情認識に着手した。ここでは、まず、「(B)音韻性と話者性の分離」の成果に基づき、音声から音韻性と話者性を分離する、disentanglement（もつれをほどく）のフレームワークを構築し、残された成分を入力として感情の認識を行うシステムを開発する。今年度は、既存のデータベースを用いたベースラインを構築した。disentanglementには、オートエンコーダー(自己符号化器）を利用した声質変換を用いる方法を採用し、実装を行った。最終年度に評価を行う予定である。評価には、感情認識の分野で広く用いられているIEMOCAPデータベースを用いる予定である。
-
Estimation of crop vitality based on Tensor decomposition and data fusion of multimodal, multitemporal leaf-scale aerial images
2017.10 - 2021.3
Japan Science and Technology Agency PRESTO
Kuniaki UTO
Authorship：Principal investigator Grant type：Competitive
-
Research on the Upgrading of Hyperspectral Image Processing
Grant number：20246069 2008 - 2011
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)
KOSUGI Yukio, UTO Kuniaki, SAITO Genya, KOJIMA Kazuyuki, OOYA Tosiki
Grant amount：\46280000 （ Direct Cost: \35600000 、 Indirect Cost：\10680000 ）
In this research, on the basis of the data observed using hyperspectral sensors, the feature quantity was extracted by optimizing the operation among wavelengths on each observed value, while image representation is carried out to satisfy the requirement for each application field such as agriculture and forestry, medical, or disaster field. Especially, efforts was also made for the research which contributed to the earthquake disaster reconstruction in the last fiscal year.
-
Study on hyperspectral data acquisition system for agriculture using industrial RC helicopter
Grant number：20500153 2008 - 2010
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
UTO Kuniaki
Grant amount：\4680000 （ Direct Cost: \3600000 、 Indirect Cost：\1080000 ）
Under the humid subtropical climate in Japan, it is difficult to obtain optical remote sensing data during the rainy season in growing period of vegetation. In this study, a low altitude hyperspectral observation system mounted on industrial RC helicopter is developed. Estimation methods of quality and quantity of vegetation, e.g. LAI, SPAD, based on hyperspectral data are developed.
-
Research on the Effective Use of apriori Information in Image Interpretation
Grant number：15100003 2003 - 2007
Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (S)
KOSUGI Yukio, KAMEYAMA Keisuke, UTO Kuniaki, KOSAKA Naoko
Grant amount：\62660000 （ Direct Cost: \48200000 、 Indirect Cost：\14460000 ）
For the accurate land-use and land-cover analysis of satellite and aerial images, as well as for automatic diagnosis and surgical planning based on medical images, we are aiming at establishing image-interpretation frameworks using a priori knowledge inherent to the subject under consideration. In the study of high-resolution images as well as hyperspectral images, Models with excessive freedom sometimes mislead us to inconsistent results. Sufficient information can be obtained by introducing models with adequately limited degree of freedom, as shown in the following results :
Damaged area estimation : The satellite image photographed before disaster was handled as a priori information, and the image photographed after the disaster was compared in nonlinear map algorithm based on the principle of "coincidence enhancement", and change detection of damaged houses was carried out. This system was applied to satellite images of the Bam earthquake of December 26, 2003, and the damaged area was extracted. The processing result was provided to a Japanese field investigation group, and the validity of the detection result was verified via the field study.
Sugar content estimation of crops : The flavor of crops is dependent on the fraction of contents such as sugar, amino acid and protein etc. In this study, we tried to estimate these quantities from the hyperspectral image of the crops before the harvest. In the estimation system, we optimized the mathematical operation among large number of wavelength of the hyperspectral images by combining the neural network with the PSO(particle swarm optimization) algorithm. In this estimation, the set of measured values obtained after the harvest was used as a tutorial data.
The derivation of human skin extraction index : We analyzed hyperspectral image of short wavelength infrared region, to make use of a priori information effectively in detecting the human victims in the water or lying on the gravel or ground after the disaster. As a result, NHI(normalized human index) was proposed to make the skin tangible by utilizing the spectral information inherent to the human skin.
Independent Component Analysis of Periodically Distributed Vegetation Images : As a basic study for the development of high-altitude observation of hyperspectral images of Japanese agricultural fields, we experimentally and theoretically consider the use of a priori knowledge in the effective use of independent component analysis for resolving mixture pixels found in images observed beyond the limitation of physical resolution. -
高解像度地理画像の変化域抽出
Grant number：13309006 2001 - 2003
日本学術振興会 科学研究費助成事業 基盤研究(A)
小杉 幸夫, 宇都 有昭, 土居原 健
Grant amount：\28990000 （ Direct Cost: \22300000 、 Indirect Cost：\6690000 ）
本研究では、高解像度衛星画像や空撮画像をもとに、地物の変化を有効に捉える方式の開発を目的とするが、本年度は主として、変化域抽出方法の安定化および変化域情報の防災応用の2点について研究を進めた。
先ず、変化域抽出方法の安定化については、初年度より実施している2画像間の非線形写像から地理画像上の変化域を抽出する方法が探査用微小領域のサイズDSの調整に敏感で、不安定要因となっていた点を改善すべく、解像度を変えた種々の空撮画像についてROC評価を基準とした実験的検討を行った。その結果、最適なDSのピクセル数は地上換算で約15m付近ではあるものの、市街の構成内容、対象とする変化の内容によってある程度のばらつきを有することが明らかになった。そこで、このアルゴリズムの検出性能を安定化するために複数のDS値について、それぞれ非線形写像を実施し、複数の処理結果を二乗誤差の形で統合する方式を提案した。この方式では、アプリオリには得られない最適なDSを持った計算ユニットが地物変化について大きな二乗誤差を出すのに対して、不適切なDSを持ったユニットでは変化が顕著に検出されないため大きな二乗差出力は生じず、結果として最適のDSを有するユニットの意見が重んじられ、安定した抽出結果が得られた。
また、変化域抽出の防災応用についても防災の専門家を交えて検討を行い、非線形写像の結果得られる移動ベクトル分布に対して発散の演算処理を施して地盤の液状化に伴う側方流動等の変化を顕在化させるなど、具体的な応用可能性についてシミュレーション実験を実施した。