Recognition of Cursive/Blind Kanji Handwriting Utilizing the Contuinous Speech Recognition Approach

利用连续语音识别方法识别草书/盲汉字手写体

• 批准号: 11480074
• 负责人: SAGAYAMA Shigeki
• 金额: $ 6.78万
• 依托单位: The University of Tokyo (2001)Japan Advanced Institute of Science and Technology (1999-2000)
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 1999
• 资助国家: 日本
• 起止时间: 1999 至 2001
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-11480074/
• 关键词: Online Handwriting Recognition; Cursive, Blind Writing; Continuous Speech Recognition Approach; Hidden Markov Model; Stroke-based Modeling; Hierarchical Kanji Structure Lexicon; Viterbi Algorithm; Handwritten by the Blind; 漢字データベース

In this project, continuous speech recognition (CSR) approach was successfully applied to online handwriting recognition and produced enormous results. Replacing phonemes and grammar in CSR with 25 basic strokes and Kanji structure lexicon and utilizing highly advanced CSR techniques to make an efficient search in the solution space, the newly developed technology is able to recognize cursive and blind hand-written Kanji characters. By observing the velocity vector sequence of the pen movements, this technology estimates the intension of what Kanji character is being written instead of recognizing the drawn figure so that it may recognize heavily deformed characters even human can not read.After establishing the above fundamental structure of the technology, we investigated a number of research items such as velocity-based features (comparing several feature parameters), pen pressure as a feature, stroke context-dependent modeling (considering stroke deformation caused by previous and following strokes), stroke macro model (adjacent strokes are combined to form a macro-stroke), adaptation to writers (adapting HMMs to the writer's characteristics), recognition of blind writing (written without seeing), recognizing overlapped writing (multiple characters on the same area), acceleration of recognition speed (improved search algorithm), curved strokes for numerics and Hirakana, and collection of cursive, blind writing and ones from blind people.Utilizing the research results, our R&D proposal of "Handwritten Communication for the Blind" was approved by Ishikawa Prefecture with a support of 100 million yen. The two projects cooperated well with each other so that more than 1 million handwritten character data were collected including cursive, blind, and sight-impaired characters.

在这个项目中，连续的语音识别（CSR）方法成功地应用于在线手写识别，并产生了巨大的结果。用25个基本笔触和汉字结构词典代替CSR中的音素和语法，并利用高级CSR技术在解决方案空间中进行有效的搜索，这项新开发的技术能够识别出粗略的和盲目的手工编写的汉字角色。 By observing the velocity vector sequence of the pen movements, this technology estimates the intension of what Kanji character is being written instead of recognizing the drawn figure so that it may recognize heavily deformed characters even human can not read.After establishing the above fundamental structure of the technology, we investigated a number of research items such as velocity-based features (comparing several feature parameters), pen pressure as a feature, stroke context-dependent modeling (considering stroke deformation caused by previous and following strokes), stroke macro model (adjacent strokes are combined to form a macro-stroke), adaptation to writers (adapting HMMs to the writer's characteristics), recognition of blind writing (written without seeing), recognizing overlapped writing (multiple characters on the same area), acceleration of recognition speed (improved search algorithm), curved strokes for数字和广场纳，以及草书，盲目写作和盲人的收集。利用研究结果，我们的“盲人手写交流”的研发提案得到了Ishikawa县的批准，并支持了1亿日元。这两个项目相互合作，因此收集了超过100万个手写角色数据，包括草书，盲目和视力障碍角色。

项目成果

嵯峨山 茂樹, 中井 満, 下平 博: "ストロークHMMに基づくオンライン手書き文字認識方式"電子情報通信学会技術研究報告. PRMU2000-35. 1-8 (2000)

Shigeki Sagayama，Mitsuru Nakai，Hiroshi Shimodaira：“基于笔画HMM的在线手写字符识别方法”IEICE技术研究报告1-8（2000）。

Junko Tokuno, Nobuhito Inami, Shigeki Matsuda, Mitsuru Nakai, Hiroshi Shimodaira, Shigeki Sagayama: "Context-Dependent Substroke Model for HMM-based On-line Handwriting Recognition"Proceedings of 8th International Workshop on Frontiers in Handwriting Reco

Junko Tokuno、Nobuhito Inami、Shigeki Matsuda、Mitsuru Nakai、Hiroshi Shimodaira、Shigeki Sagayama：“基于 HMM 的在线手写识别的上下文相关子笔画模型”第八届手写识别前沿国际研讨会论文集

井波暢人,松田繁樹,中井満,下平博,嵯峨山茂樹: "環境依存型ストロークHMMによるオンライン手書き文字認識"平成12年電気関係学会北陸支部大会講演論文集,F-93. 394 (2000)

Nobuto Inami、Shigeki Matsuda、Mitsuru Nakai、Hiroshi Shimodaira、Shigeki Sagayama：“使用环境相关笔画 HMM 进行在线手写字符识别”日本电气工程师学会 2000 年北陆分会会议记录，F-93（2000 年） ）

Naoto Akira, Mitsuru Nakai, Hiroshi Shimodaira, and Shigeki Sagayama: "On-line Handwritten Character Recognition based on Stroke-HMM in non Visual-feedback Writing Condition"IEICE Technical Report. PRMU2000-206. 39-46 (2001)

Naoto Akira、Mitsuru Nakai、Hiroshi Shimodaira 和 Shigeki Sagayama：“非视觉反馈书写条件下基于笔画-HMM 的在线手写字符识别”IEICE 技术报告。

中井満,嵯峨山茂樹,秋良直人,小場久雄,下平博: "ストロークHMMによるオンライン手書き文字認識の性能評価"信学技報,PRMU2000-36. 9-16 (2000)

Mitsuru Nakai、Shigeki Sagayama、Naoto Akira、Hisao Oba、Hiroshi Shimodaira：“使用笔画 HMM 进行在线手写字符识别的性能评估”IEICE 技术报告，PRMU2000-36 (2000)。