Research on unified models and techniques for recognizing multilingual handwriting from touch based input devices

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00225
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Tokyo University of Agriculture and Technology
• Principal Investigator: Zhu Bilan 東京農工大学, 工学(系)研究科(研究院), 助教 (50466918)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 手書き認識 / オンライン認識 / 確率モデル

Outline of Final Research Achievements

Based on the highest recognition technologies by explicit segmentation methods established in Japanese online handwriting, we proposed a comprehensive model and applied it to multilingual handwriting recognition such as English, Chinese and Arabic, and we realized performances and advantages more than each technique while containing the merits of the existing techniques for each language.
By using large-scale databases made in Institute of Automation, Chinese Academy of Sciences and University at Buffalo, Buffalo, NY, USA, and turning a PDCA cycle on comparison with the existing techniques, we have achieved the goal. We presented the research results at the places such as international conferences and article magazines and exchanged opinions necessary for studies.

Free Research Field

パタン認識