[論文查詢：請進入全國博碩士論文網：檢索字串輸入林彰鏗，欄位選擇研究生，進行檢索]
論文名稱：
研究生: 林彰鏗
(以研究生姓名查詢國家圖書館索書號 ，未查獲者表國圖尚未典藏)
(以研究生姓名查詢國科會科資中心微片資料庫)
(連結至全國圖書聯合目錄) (連結至政大圖書館館藏目錄)
論文名稱: 中文文字知識應用於線上手寫中文識別之研究
指導教授: 范子儀
范國清
學位類別: 博士
校院名稱: 國立中央大學
系所名稱: 資訊及電子工程研究所
學年度: 81
語文別: 中文
論文頁數: 129
關鍵字: 中文文字
手寫中文識別
全文影像: (依著作權法相關規定，以下全文影像僅限國家圖書館館內下載)
目錄
摘要
英文摘要
第一章 緒論
1.1. 研究動機
1.2. 影響研究之因素
1.3. 現在和未來之識別策略
1.4. 正楷字識別之研究範圍
第二章 實驗識別系統之架構
2.1. 學習子系統之架構
2.2. 識別子系統之架構
第三章 建立特徵模板資料庫
3.1. 預處理
3.2. 抽取基本之筆劃
3.3. 正楷字之變異分析
3.4. 中文字知識之抽取
第四章 動態規劃比對別法
4.1. 比對圖
4.2. 類似度量測函數
4.3. 動態規劃比對
4.4. 實驗結果
第五章 A*淀算法為基礎之比對識別
5.1. 比對樹
5.2. 類似度量測函數
5.3. A*演算法為基礎之比對比
5.4. 實驗結果
第六章 應用人工智慧技術於混淆字集之識別
6.1. 識別原理
6.2. 實驗結果
第七章 結論和展望
7.1. 研究結論
7.2. 系統改進與整合
7.3. 草寫字識別之未來策略
1 Introduction
1.1 Motivations for On-Line Chinese Character Recognition (OLCCR)
1.2 Introduction to OLCCR
1.3 Factors Affecting OLCCR
1.4 Our Recognition strategies for the Present and Future
1.5 Research Scope of Square Writing Character Recognition
1.6 Main Results of This Dissertation
1.7 Outline of This Dissertation
2 Introduction to Experimental Recognition System
2.1 Motivations of Presenting the System
2.2 Architecture of Learning Subsystem
2.3 Architecture of Recognition Subsystem
2.4 Objectives
2.5 System's Approach
2.6 System Flow of Learning Subsystem
2.7 System Flow of Recognition Subsystem
3 Establishment of Feature Template Data Base
3.1 Preprocessing
3.2 Primitive Stroke Extraction
3.3 Variation Analysis for Square Writing Character
3.4 Chinese Character Knowledge Extraction
3.5 Preclassification for Templates
4 Matching Recognition by Dynamic Programming Matching Method
4.1 Matching Graph
4.2 Similarity Measure Function
4.3 Stroke Similarity Measure
4.4 Dynamic Programming Matching
4.5 Decision Making
4.6 Experimental Results
5 Matching Recognition by A*Algorithm Based Method
5.1 Matching Tree
5.2 Matching Criterion
5.3 Stroke Similarity Measure
5.4 A*Algorithm Based Matching
5.5 Decision Making
5.6 Experimental Results
6 Confusion Set Recognition by Using Artificial Intelligence Based Techniques
6.1 Problem Description
6.2 The System Architecture
6.3 Character Knowledge Extraction
6.4 Pattern Pair Matching
6.5 Experiments and Discussion
7 The Conclusion and Beyond
7.1 Research Conclusions
7.2 System Improvement and Integration
7.3 Future Recognition Strategies for Cursive Script Recognition
References
Publications List and Patents List
[摘要]
中文在目前電腦資訊普及化時代所遭遇最大的挑戰是資訊入自動化的問題，線上手寫
中文文字識別技術是解決此類問題本的辦法。提出能有效識別草寫字的線上文字識別技術
和實用系統是研究發展之理想目標，廣泛應用作為中文輸入工具之趨是可以預期。為符合
未來趨勢之需要，本文即以線上中文正字識別為研究之目標。
由於中文字本身特性字數多、結構複雜，以及使用者使用性因人而異，成功研發出草
寫字識別科技非一蹴可及，為國因應一問題特性，本文擬訂一個研究發展策略，並依此策
略，提出用中文文字啟示知識做為代表中文特徵之構想，實作於一套縐手寫中文正楷字識
別實驗系統，期證明本構想為未來草寫字識別之有效途徑，並顯現此一研發領域累積研究
發經驗之重要性。
傳統的特徵抽取方法中，係針對大字彙中文字抽取其共通特徵為主，用此特徵代表多
變異的草寫字並不適用，應改採啟示識觀念針對各別小字彙或各別文字抽取各別特徵為方
式。在識驗系統中，我們提出兩種知識與三種比對法，作正楷文字之識，其識別率在取
前10位最相似候選字時可達98%，取第一位準確率可達90%以上。作為識別相似字，正確率
則可達99.9%。實驗結果證實本論文構想是合理可行， 未來可爰此識別策略進行研發。
[摘要]
The greatest challenge facing the popularization of the Chinese computer
in the age of information is the automation of Chinese information input. On-
line Chinese character recognition technique is the fundamental solution to
this problem. The proposal of an on-line Chinese character recognition
technique able to effectively recognize cursive script, and a practical system
is the ideal goal of research and development. The trend of applying the two
proposals in question as the Chinese input tools is well in sight. For the
demand in the future trend, the on-line square writing Chinese character
recognition is selected as our research objectives.
For Chinese, due to its huge number of characters, complicated structure
and the variations of usage characteristics depending on individual user, it
will not be easy to successfully research and develop the cursive script
recognition technique. To cope with the characteristics of this problem, we
specified a research and development strategy. Based on this strategy, we
proposed the idea of adopting the Chinese character heuristic knowledge to
represent Chinese feature, and actually put into practice in an experimental
system of on-line Chinese square writing character recognition. We hope to
prove that this idea is an effective step stone to the development of cursive
script recogition, and demonstrate that the importance of the accumulation of
research and development experiences in this field.
The conventional feature extraction methods mainly rely on extracting the
common features shares by characters in a large group of vocabulary. But this
is not appropriate for cursive scripts, whose features features represent
numerous variations. A better way is to use the concept of heuristic knowledge
for extracting individual features from small groups of vocabulary or
individual characters. In our experimental system, we proposed two kinds of
knowledge and three types of matching methods for the recognition of square
writing. The accuracy can reach 98% when the first ten most similar candidates
are selected, and over 90% when only the first is selected. Used in the
recognition of similar characters, the accuracy can reach 99.9%. The
experimental results prove that the idea in this thesis is reasonable,
feasible, and sufficient to act the recognition strategy of further research
and development.