Handwritten historical document analysis, recognition, and retrieval - state of the art and future trends
Intro -- Contents -- 1. Introduction -- The HisDoc Project -- 2. IAM-HistDB: A Dataset of Handwritten Historical Documents -- 2.1 Introduction -- 2.2 Related Work -- 2.3 The IAM-HistDB -- 2.3.1 Saint Gall Database -- 2.3.2 Parzival Database -- 2.3.3 George Washington Database -- 2.4 Semi-Automatic G...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Weitere Verfasser: | , |
| Format: | UnknownFormat |
| Sprache: | eng |
| Veröffentlicht: |
New Jersey ; London ; Singapore ; Beijing ; Shanghai ; Hong Kong ; Taipei ; Chennai ; Tokyo
World Scientific Publishing Co. Pte. Ltd.
2021
|
| Schriftenreihe: | Series in machine perception and artificial intelligence
89 |
| Schlagworte: | |
| Online Zugang: | Inhaltsverzeichnis |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Tag hinzufügen | Zusammenfassung: | Intro -- Contents -- 1. Introduction -- The HisDoc Project -- 2. IAM-HistDB: A Dataset of Handwritten Historical Documents -- 2.1 Introduction -- 2.2 Related Work -- 2.3 The IAM-HistDB -- 2.3.1 Saint Gall Database -- 2.3.2 Parzival Database -- 2.3.3 George Washington Database -- 2.4 Semi-Automatic Ground Truth Creation -- 2.5 Conclusions -- References -- 3. DIVA-HisDB: A Precisely Annotated Dataset of Challenging Medieval Manuscripts -- 3.1 Introduction -- 3.2 Description -- 3.2.1 CSG18 -- 3.2.2 CSG863 -- 3.2.3 CB55 -- 3.3 Creation -- 3.4 Competition -- 3.4.1 Evaluation and Results 3.4.1.1 Task-1: Layout Analysis -- 3.4.1.2 Task-2: Baseline Extraction -- 3.4.1.3 Task-3: Line Segmentation -- 3.4.2 Discussion -- References -- 4. Layout Analysis in Handwritten Historical Documents -- 4.1 Introduction -- 4.2 Segmentation in Regions of Interest -- 4.3 Region Description -- 4.4 Typical Processing Steps -- 4.4.1 Binarization -- 4.4.2 Grouping Entities -- 4.4.3 Cutting -- 4.4.4 Labeling Data -- 4.5 Layout Analysis Methods -- 4.5.1 Content Identification -- 4.5.2 Text Line Segmentation -- 4.6 Open Problems -- 4.6.1 Semantical Analysis of the Layout -- 4.6.2 Reading Order 4.6.3 Rare Occurrences -- References -- 5. Automatic Handwriting Recognition in Historical Documents -- 5.1 Introduction -- 5.2 Image Preprocessing and Feature Extraction -- 5.3 Character Modeling -- 5.3.1 HMM Character Models -- 5.3.2 LSTM Character Models -- 5.4 Automatic Transcription -- 5.5 Extensions -- 5.6 Conclusions -- References -- 6. Handwritten Keyword Spotting in Historical Documents -- 6.1 Introduction -- 6.2 Related Work -- 6.2.1 Example-Based Search Queries -- 6.2.2 String-Based Search Queries -- 6.2.3 Embedding-Based Search Queries -- 6.3 LSTM NN-Based Keyword Spotting 6.3.1 Document Representation -- 6.3.2 LSTM Neural Networks -- 6.3.3 Connectionist Temporal Classification -- 6.3.4 Extending CTC for Efficient Keyword Spotting -- 6.3.5 Experimental Evaluation -- 6.4 Remarks and Further Research -- 6.5 Common Databases -- 6.6 Conclusion -- References -- 7. DIVAServices -- Transforming Document Analysis Methods into Web Services -- 7.1 Abstract -- 7.2 Introduction -- 7.3 Related Work -- 7.3.1 Web Services in Document Image Analysis -- 7.3.2 Web Services in Other Fields -- 7.4 DIVAServices -- The RESTful Web Service Framework 7.5 Core Interactions with DivaServices -- 7.5.1 Accessing Method Information -- 7.5.2 Providing Data -- 7.5.3 Execution of a Method -- 7.6 Example Use of DivaServices -- 7.6.1 Upload the Original Image -- 7.6.2 Binarize the Image -- 7.6.3 Extracting Text Lines -- 7.6.4 Performing OCR -- 7.7 The Ecosystem of DivaServices -- 7.7.1 DivaServices-Spotlight -- 7.7.2 DivaServices-WebInterface -- 7.7.3 DivaServices-Management -- 7.8 Conclusion and Future Work -- References -- 8. GraphManuscribble: Interactive Annotation of Historical Manuscripts -- 8.1 Introduction -- 8.2 Related Work |
|---|---|
| Beschreibung: | 8.2.1 Document Segmentation and Annotation Systems |
| Beschreibung: | xiii, 254 Seiten Illustrationen, Diagramme, Faksimiles |
| ISBN: | 9789811203237 978-981-120-323-7 |