Muutke küpsiste eelistusi

E-raamat: Handwritten Historical Document Analysis, Recognition, And Retrieval - State Of The Art And Future Trends

Edited by (Univ Of Fribourg, Switzerland), Edited by (Univ Of Fribourg, Switzerland), Edited by (Univ Of Fribourg, Switzerland)
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 81,90 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Handwritten Historical Document Analysis, Recognition, And Retrieval - State Of The Art And Future TrendsSuurem pilt
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

In recent years, libraries and archives all around the world have increased their efforts to digitize historical manuscripts. To integrate the manuscripts into digital libraries, pattern recognition and machine learning methods are needed to extract and index the contents of the scanned images. The unique compendium describes the outcome of the HisDoc research project, a pioneering attempt to study the whole processing chain of layout analysis, handwriting recognition, and retrieval of historical manuscripts. This description is complemented with an overview of other related research projects, in order to convey the current state of the art in the field and outline future trends. This must-have volume is a relevant reference work for librarians, archivists and computer scientists.

1 Introduction
1(10)
Andreas Fischer
Marcus Liwicki
Rolf Ingold
The HisDoc Project
9(2)
2 IAM-HistDB: A Dataset of Handwritten Historical Documents
11(14)
Andreas Fischer
2.1 Introduction
11(1)
2.2 Related Work
12(1)
2.3 The IAM-HistDB
13(6)
2.3.1 Saint Gall Database
15(1)
2.3.2 Parzival Database
15(2)
2.3.3 George Washington Database
17(2)
2.4 Semi-Automatic Ground Truth Creation
19(2)
2.5 Conclusions
21(4)
References
22(3)
3 DIVA-HisDB: A Precisely Annotated Dataset of Challenging Medieval Manuscripts
25(20)
Foteini Simistira Liwicki
3.1 Introduction
25(1)
3.2 Description
26(5)
3.2.1 CSG18
29(1)
3.2.2 CSG863
29(1)
3.2.3 CB55
30(1)
3.3 Creation
30(1)
3.4 Competition
31(14)
3.4.1 Evaluation and Results
34(4)
3.4.2 Discussion
38(3)
References
41(4)
4 Layout Analysis in Handwritten Historical Documents
45(22)
Mathias Seuret
4.1 Introduction
45(1)
4.2 Segmentation in Regions of Interest
46(2)
4.3 Region Description
48(2)
4.4 Typical Processing Steps
50(4)
4.4.1 Binarization
50(2)
4.4.2 Grouping Entities
52(1)
4.4.3 Cutting
53(1)
4.4.4 Labeling Data
53(1)
4.5 Layout Analysis Methods
54(8)
4.5.1 Content Identification
54(3)
4.5.2 Text Line Segmentation
57(5)
4.6 Open Problems
62(5)
4.6.1 Semantical Analysis of the Layout
62(1)
4.6.2 Reading Order
63(1)
4.6.3 Rare Occurrences
63(1)
References
64(3)
5 Automatic Handwriting Recognition in Historical Documents
67(14)
Andreas Fischer
5.1 Introduction
67(2)
5.2 Image Preprocessing and Feature Extraction
69(1)
5.3 Character Modeling
70(4)
5.3.1 HMM Character Models
71(1)
5.3.2 LSTM Character Models
72(2)
5.4 Automatic Transcription
74(3)
5.5 Extensions
77(1)
5.6 Conclusions
78(3)
References
79(2)
6 Handwritten Keyword Spotting in Historical Documents
81(20)
Volkmar Frinken
Shriphani Palakodety
6.1 Introduction
81(1)
6.2 Related Work
82(3)
6.2.1 Example-Based Search Queries
83(1)
6.2.2 String-Based Search Queries
84(1)
6.2.3 Embedding-Based Search Queries
84(1)
6.3 LSTM NN-Based Keyword Spotting
85(8)
6.3.1 Document Representation
86(1)
6.3.2 LSTM Neural Networks
87(1)
6.3.3 Connectionist Temporal Classification
88(2)
6.3.4 Extending CTC for Efficient Keyword Spotting
90(1)
6.3.5 Experimental Evaluation
91(2)
6.4 Remarks and Further Research
93(2)
6.5 Common Databases
95(1)
6.6 Conclusion
95(6)
References
97(4)
7 DIVA Services - Transforming Document Analysis Methods into Web Services
101(20)
Marcel Gygli
7.1 Abstract
101(1)
7.2 Introduction
102(1)
7.3 Related Work
103(1)
7.3.1 Web Services in Document Image Analysis
103(1)
7.3.2 Web Services in Other Fields
104(1)
7.4 DIVAServices - The RESTful Web Service Framework
104(1)
7.5 Core Interactions with DivaServices
105(4)
7.5.1 Accessing Method Information
106(1)
7.5.2 Providing Data
107(1)
7.5.3 Execution of a Method
108(1)
7.6 Example Use of DivaServices
109(5)
7.6.1 Upload the Original Image
110(1)
7.6.2 Binarize the Image
111(1)
7.6.3 Extracting Text Lines
112(1)
7.6.4 Performing Optical Character Recognition (OCR)
113(1)
7.7 The Ecosystem of DIVAServices
114(3)
7.7.1 DlVAServices-Spotlight
115(1)
7.7.2 DiVAServices-Weblnterface
116(1)
7.7.3 DiVAServices-Management
117(1)
7.8 Conclusion and Future Work
117(4)
References
118(3)
8 GraphManuscribble: Interactive Annotation of Historical Manuscripts
121(36)
Angelika Garz
8.1 Introduction
121(4)
8.2 Related Work
125(7)
8.2.1 Document Segmentation and Annotation Systems
125(4)
8.2.2 Human-Computer Interaction in Image Segmentation
129(3)
8.3 Document Graphs
132(12)
8.3.1 Basic Definitions for Graphs
132(1)
8.3.2 Graph Nodes
133(3)
8.3.3 Graph Edges
136(1)
8.3.4 Edge Weights
137(1)
8.3.5 Graph Clustering
138(1)
8.3.6 Split Layout Elements
139(1)
8.3.7 Polygonal Graph Representation
139(2)
8.3.8 Graph Evaluation
141(3)
8.4 Graph-User-Interaction: Scribbling
144(4)
8.4.1 Scribbling as User Interaction Pattern
145(2)
8.4.2 User Interaction Evaluation
147(1)
8.5 Conclusions and Outlook
148(9)
References
149(6)
Related Research Projects
155(2)
9 OldDocPro: Old Greek Document Recognition
157(18)
Basilis Gatos
Georgios Louloudis
Nikolaos Stamatopoulos
George Retsinas
Giorgos Sfikas
Angelos P. Giotis
Foteini Simistira Liwicki
Vassilis Papavassiliou
Vassilis Katsouros
9.1 Introduction
157(2)
9.2 The GRPOLY-DB Database
159(3)
9.3 Page Segmentation
162(3)
9.3.1 Performance Evaluation of Page Segmentation
162(1)
9.3.2 Word Segmentation
162(2)
9.3.3 Document Image Segmentation Representation
164(1)
9.4 Text Recognition
165(3)
9.4.1 Isolated Character Recognition
165(2)
9.4.2 Text Line Recognition
167(1)
9.5 Keyword Spotting
168(3)
9.6 Conclusions
171(4)
References
172(3)
10 Advances in Handwritten Keyword Indexing and Search Technologies
175(20)
Joan Puigcerver
Alejandro H. Toselli
Enrique Vidal
10.1 Introduction
175(3)
10.2 Proposed Indexing and Search Technology
178(4)
10.2.1 Pixel-Level Word Relevance Probabilities: the "Posteriorgram"
179(1)
10.2.2 Image Region Word Relevance Probabilities
180(1)
10.2.3 Minimal Searchable Image Regions: Line-Level KWS
181(1)
10.2.4 Efficient Computation of Posteriorgrams and Relevance Probabilities
181(1)
10.3 Datasets
182(3)
10.4 Experimental Framework
185(2)
10.4.1 System Setup
185(1)
10.4.2 Dataset Usage Details and Query Set Selection
185(1)
10.4.3 Evaluation Measures
186(1)
10.5 Laboratory Results
187(1)
10.6 Demonstration Systems
188(1)
10.7 Conclusion and Outlook
189(6)
References
191(4)
11 Browsing of the Social Network of the Past: Information Extraction from Population Manuscript Images
195(26)
Alicia Fornes
Josep Llados
Joana Maria Pujadas-Mora
11.1 Introduction
195(3)
11.2 Population Records and Datasets
198(2)
11.3 System Architecture
200(2)
11.4 Image Capture and Document Enhancement
202(2)
11.4.1 Layout Analysis and Text Line Extraction
202(2)
11.5 Annotation Space
204(4)
11.5.1 Key Word Spotting
204(3)
11.5.2 Handwritten Text Recognition
207(1)
11.6 Semantic Space
208(3)
11.6.1 Named Entity Recognition
208(1)
11.6.2 Context-Aware Transcription
209(2)
11.6.3 Record Linkage
211(1)
11.7 User Space
211(4)
11.7.1 Crowdsourcing Applications
212(2)
11.7.2 The Browsers
214(1)
11.8 Conclusions
215(2)
11.9 Acknowledgements
217(4)
References
218(3)
12 Lifelong Learning for Text Retrieval and Recognition in Historical Handwritten Document Collections
221(28)
Lambert Schomaker
12.1 Introduction
221(5)
12.2 Expectation Management
226(5)
12.3 Deep Learning
231(1)
12.4 The Ball-Park Principle
232(3)
12.5 Technical Realization
235(5)
12.5.1 Work Flow
236(1)
12.5.2 Quality and Quantity of Material
236(1)
12.5.3 Industrialization and Scalability
237(1)
12.5.4 Human Effort
237(1)
12.5.5 Algorithms
237(1)
12.5.6 Object of Recognition: Whole-Word Approaches
237(1)
12.5.7 Processing Pipeline
238(2)
12.6 Performance
240(3)
12.7 Compositionality
243(1)
12.8 Conclusion
243(6)
References
246(3)
13 Conclusions and Future Trends
249(4)
Andreas Fischer
Marcus Liwicki
Rolf Ingold
Index 253
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97898112032516e.html
Märksõnad: