novel probability Pattern Recognition derived from a density function consisting with normal samples and their mirror ones

• Project/Area Number: 24500338
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: Tokyo University of Agriculture and Technology
• Principal Investigator: SEIJI Hotta 東京農工大学, 工学(系)研究科(研究院), 准教授 (90346932)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Project Status: Completed (Fiscal Year 2014)
• Budget Amount: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000); Fiscal Year 2014: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000); Fiscal Year 2013: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000); Fiscal Year 2012: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
• Keywords: パターン認識 / 部分空間法 / Pattern Recognition

Outline of Final Research Achievements

This research presents a novel probability density function (PDF) consisting with normal samples and their mirror ones. We can derive heuristic classifiers such as simple and multiple subspace classifiers from this PDF. The performance of our approach is verified by experiments on big data recognition such as image, audio, and video.

Research Products

• [Journal Article] Mahalanobis distance on extended Grassmann manifolds for variational pattern analysis (2014)
  • Author(s): Y. Washizawa and S. Hotta
  • Journal Title: IEEE Trans. Neural Networks and Learning Systems Volume: - Issue: 11 Pages: 1980-1990
  • DOI: 10.1109/tnnls.2014.2301178
  • Peer Reviewed / Open Access
• [Presentation] 部分空間法と最近の幾何学的不変性へのアプローチ (2014)
  • Author(s): 堀田 政二
  • Organizer: JEITA認識形入力方式標準化専門委員会集中審議
  • Place of Presentation: 御殿場高原・時之栖（静岡県御殿場市）
  • Year and Date: 2014-09-25 – 2014-09-26
  • Invited
• [Presentation] 具体例に基づく画像の幾何学的不変性の実現 (2014)
  • Author(s): 堀田 政二
  • Organizer: 第５９回 STARC アドバンストセミナー
  • Place of Presentation: 川崎市産業振興会館（神奈川県川崎市）
  • Year and Date: 2014-09-19
  • Invited
• [Presentation] 機械学習による Facebook に基づく株価予測 (2014)
  • Author(s): 及川 健一郎 堀田 政二
  • Organizer: 精密工学会サマーセミナー
  • Place of Presentation: 岩松旅館 （宮城県仙台市）
  • Year and Date: 2014-08-19 – 2014-08-20
• [Presentation] 漫画からの顔検出とその応用に関する基礎検討 (2014)
  • Author(s): 並木 駿也, 佐山 裕一, 中尾 駿太、堀田 政二
  • Organizer: MIRU2014
  • Place of Presentation: 岡山コンベンションセンター（岡山県岡山市）
  • Year and Date: 2014-07-28 – 2014-07-31
• [Presentation] Appearance-Based Character Recognition in Natural Scenes Using Subspace Classifier (2014)
  • Author(s): Kenta Tsuishu and Seiji Hotta
  • Organizer: ICT-ISPC 2014
  • Place of Presentation: バンコク タイ
• [Presentation] Local Subspace Classifier with Gabor Filter Decomposition for Image Classification (2013)
  • Author(s): Kouta Tanaka and Seiji Hotta
  • Organizer: Recent Advances in Computer Vision and Pattern Recognition
  • Place of Presentation: 沖縄 Loisir Hotel & SPA Tower Naha
• [Presentation] Local Subspace Classifier with Transformation Invariance for Appearance-Based Character Recognition in Natural Images (2013)
  • Author(s): Keisuke Higa and Seiji Hotta
  • Organizer: ICDAR
  • Place of Presentation: ワシントンＤＣ アメリカ
• [Presentation] Character Recognition and Human Localization with Subspace Classifier (2013)
  • Author(s): Keisuke Higa, Takashi Higa, and Seiji Hotta
  • Organizer: IGS
  • Place of Presentation: 奈良 東大寺
• [Presentation] Mahalanobis distance on Grassmann manifold and its application to brain signal processing (2012)
  • Author(s): Washizawa and Seiji Hotta
  • Organizer: Machine Learning for Signal Processing
  • Place of Presentation: スペイン The Royal Palace of The Magdalena
• [Presentation] 複合部分空間法を用いた動画像の意味的索引付け (2012)
  • Author(s): 田中 福嗣，堀田 政二
  • Organizer: IAIP サマーセミナー2012
  • Place of Presentation: 愛知 愛知犬山温泉
• [Presentation] 画像・映像認識におけるパターン認識技術の動向
  • Author(s): 堀田政二
  • Organizer: 撮像技術と画像特徴抽出・認識技術の最前線
  • Place of Presentation: 東京 グランパーク プラザ
  • Invited
• [Presentation] 例題ではじめる部分空間法 ～パターン認識へのいざない～
  • Author(s): 堀田政二
  • Organizer: 第69回 ロボット工学セミナー
  • Place of Presentation: 東京 東京大学 本郷キャンパス
  • Invited
• [Remarks] 部分空間法のページ
  • URL: http://www.tuat.ac.jp/~s-hotta/RSJ2012/
• [Remarks] 部分空間法のページ
  • URL: http://www.tuat.ac.jp/~s-hotta/RSJ2012/