Atjaunināt sīkdatņu piekrišanu

E-grāmata: Multimodal Interactive Handwritten Text Transcription [World Scientific e-book]

(Univ Politecnica De Valencia, Spain), (Univ Politecnica De Valencia, Spain), (Univ Politecnica De Valencia, Spain)
Citas grāmatas par šo tēmu:
  • World Scientific e-book
  • Cena: 91,68 €*
  • * this price gives unlimited concurrent access for unlimited time
Multimodal Interactive Handwritten Text TranscriptionPalielināt
Citas grāmatas par šo tēmu:
World Scientific e-booksMore info about World Scientific e-books
Rokasgrāmatas mājas lapa: http://www.worldscientific.com/worldscibooks/10.1142/8394#t=toc
This book presents an interactive multimodal approach for efficient transcription of handwritten text images. This approach, rather than full automation, assists the expert in the recognition and transcription process.Until now, handwritten text recognition (HTR) systems are far from being perfect and heavy human intervention is often required to check and correct the results of such systems. The interactive scenario studied in this book combines the efficiency of automatic handwriting recognition systems with the accuracy of the experts, leading to a cost-effective perfect transcription of the handwritten text images.The interactive system here allows the user to repeatedly interact with the system. Hence, the quality and ergonomy of the interactive process is crucial for the success of the system. Moreover, more ergonomic multimodal interfaces are used to obtain an easier and more comfortable human-machine interaction.
Preface v
1 Preliminaries
1(28)
1.1 Introduction
1(4)
1.2 State of the Art
5(4)
1.2.1 Optical Character Recognition
6(1)
1.2.2 Handwritten Text Recognition
7(2)
1.3 Formal Background
9(18)
1.3.1 Hidden Markov Models
9(6)
1.3.2 Language models: N-grams
15(3)
1.3.3 Interactive Pattern Recognition
18(4)
1.3.4 Word-graphs
22(5)
1.4 Assessing Computer Assisted Transcription of Handwritten Text Images
27(2)
2 Corpora
29(12)
2.1 Introduction
29(1)
2.2 CS
29(3)
2.3 ODEC
32(4)
2.4 IAMDB
36(3)
2.5 UNIPEN
39(2)
3 Handwritten Text Recognition
41(32)
3.1 Introduction
41(2)
3.2 Off-line Handwritten Text Recognition
43(25)
3.2.1 Preprocessing
43(4)
3.2.2 Feature Extraction
47(2)
3.2.3 Recognition
49(5)
3.2.4 Experimental Framework
54(2)
3.2.5 Meta-parameter Adjustment Experiments
56(4)
3.2.6 Discussion of Results
60(8)
3.3 On-line Handwritten Text Recognition
68(3)
3.3.1 Preprocessing
68(1)
3.3.2 Feature Extraction
68(2)
3.3.3 Recognition
70(1)
3.3.4 Experimental Framework
70(1)
3.3.5 Results
71(1)
3.4 Summary and Conclusions
71(2)
4 Computer Assisted Transcription of Handwritten Text Images
73(36)
4.1 Introduction
73(2)
4.2 Formal Framework
75(2)
4.3 Adapting the Language Model
77(1)
4.4 Searching
78(7)
4.4.1 Direct Viterbi-based Approach
78(2)
4.4.2 Word-graph based Approach
80(5)
4.5 Increasing Interaction Ergonomy
85(4)
4.5.1 Language Modelling and Search
88(1)
4.6 Interacting at the Character Level
89(3)
4.6.1 Language Modelling and Search
91(1)
4.7 Experimental Framework
92(4)
4.7.1 Assessment Measures
93(2)
4.7.2 Parameters and Meta-Parameters
95(1)
4.8 Results
96(9)
4.9 Conclusions and Future Work
105(4)
5 Multimodal Computer Assisted Transcription of Hand-written Text Images
109(14)
5.1 Introduction
109(2)
5.2 Formal Framework
111(3)
5.3 Adapting the Language Model
114(1)
5.4 Searching
115(1)
5.5 Experimental Framework
116(2)
5.5.1 Corpora
116(1)
5.5.2 Assessment Measures
117(1)
5.6 Results
118(3)
5.7 Conclusions
121(2)
6 A Web-based Demonstrator of Interactive Multimodal Transcription
123(22)
6.1 Introduction
123(1)
6.2 User Interaction Protocol
124(1)
6.3 System Description
125(9)
6.3.1 Application Programming Interface
126(1)
6.3.2 MM-CATTI Server
126(1)
6.3.3 Web Interface
127(2)
6.3.4 Electronic Pen or Touchscreen Interaction
129(4)
6.3.5 Keyboard and Mouse Interaction
133(1)
6.4 Evaluation
134(3)
6.4.1 Assessment Measures
135(1)
6.4.2 Corpus
135(1)
6.4.3 Participants
135(1)
6.4.4 Apparatus
135(1)
6.4.5 Procedure
136(1)
6.4.6 Design
137(1)
6.5 Results and Discussion
137(7)
6.5.1 Quantitative Analysis
138(2)
6.5.2 Qualitative Analysis
140(1)
6.5.3 Correlation Analysis
141(1)
6.5.4 Limitations of the Study
142(2)
6.6 Conclusions
144(1)
7 Conclusions and Outlook
145(4)
7.1 Conclusions
145(2)
7.2 Outlook
147(2)
Appendix A Symbols and Acronyms
149(6)
A.1 Symbols
149(3)
A.2 Acronyms
152(3)
Bibliography 155(10)
Index 165
Permanent link: https://www.kriso.lv/db/9789814390347_pe.html