Muutke küpsiste eelistusi

E-raamat: Graph-based Keyword Spotting

Edited by (Univ Of Applied Sciences & Arts Northwestern, Switzerland), Edited by (Univ Of Fribourg, Switzerland), Edited by (Univ Of Bern, Switzerland & Univ Of Applied Sciences & Arts Northwestern, Switzerland)
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 87,75 €*
  • * 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.
Graph-based Keyword SpottingSuurem 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. 

"Keyword Spotting (KWS) has been proposed as a flexible and more error-tolerant alternative to full transcriptions. In most cases, it allows to retrieve arbitrary query words in handwritten historical document. This comprehensive compendium gives a self-contained preamble and visually attractive description to the field of graph-based KWS. The volume highlights a profound insight into each step of the whole KWS pipeline, viz. image preprocessing, graph representation and graph matching. Written by two world-renowned co-authors, this unique title combines two very current research fields of graph-based pattern recognition and document analysis. The book serves as an attractive teaching material for graduate students, as well as a useful reference text for professionals, academics and researchers."

Preface vii
List of Figures xiii
List of Tables xxiii
1 Introduction 1(12)
1.1 Keyword Spotting (KWS) as a Scientific Discipline
1(5)
1.2 Limitations
6(1)
1.3 Research Questions
6(2)
1.4 Contributions
8(2)
1.5 Outline and Organisation of the Book
10(3)
2 Related Work 13(28)
2.1 A Taxonomy of KWS Systems
13(4)
2.2 Image Preprocessing
17(3)
2.3 Formal Representation
20(7)
2.3.1 Statistical Representation
20(6)
2.3.2 Structural Representation
26(1)
2.4 Querying
27(11)
2.4.1 Template-based Approaches for KWS
29(5)
2.4.2 Learning-based Approaches for KWS
34(4)
2.5 Summary
38(3)
3 Datasets 41(20)
3.1 Historical Manuscripts
41(4)
3.1.1 George Washington (GW)
44(1)
3.1.2 Parzival (PAR)
44(1)
3.1.3 Alvermann Konzilsprotokolle (AK)
44(1)
3.1.4 Botany (BOT)
45(1)
3.2 Preprocessing the Datasets
45(13)
3.2.1 Gaussian Filtering
50(2)
3.2.2 Binarisation
52(1)
3.2.3 Segmentation
53(1)
3.2.4 Skew Correction
54(1)
3.2.5 Morphological Filtering
55(2)
3.2.6 Skeletonisation
57(1)
3.3 Summary
58(3)
4 Graph Representations 61(22)
4.1 Introduction and Basic Definitions
61(1)
4.2 Graph-Based Pattern Recognition
62(2)
4.3 From Handwriting to Graphs
64(11)
4.3.1 Keypoint-Based Graphs
65(2)
4.3.2 Grid-Based Graphs
67(2)
4.3.3 Projection-Based Graphs
69(4)
4.3.4 Split-Based Graphs
73(2)
4.4 Preliminary Experimental Evaluation
75(5)
4.5 Summary
80(3)
5 Graph Matching 83(28)
5.1 Overview and Broader Perspective
83(1)
5.2 Graph Matching for KWS
84(6)
5.2.1 Exact Graph Matching
84(3)
5.2.2 Inexact Graph Matching
87(2)
5.2.3 Formal Requirements
89(1)
5.3 Graph Edit Distance
90(13)
5.3.1 Suboptimal Algorithms
93(1)
5.3.2 Bipartite Graph Edit Distance (BP)
94(3)
5.3.3 Hausdorff Edit Distance (HED)
97(4)
5.3.4 Bipartite Graph Edit Distance 2 (BP2)
101(2)
5.4 Polar Graph Dissimilarity (PGD)
103(5)
5.4.1 Polar Segmentation of Graphs
104(1)
5.4.2 Histogram-Based Representation
104(2)
5.4.3 Histogram-Based Dissimilarity Measure
106(2)
5.5 Summary
108(3)
6 Graph-Based Keyword Spotting 111(18)
6.1 Overview of our Framework for KWS
111(2)
6.2 From Graph Edit Distance (GED) to Retrieval Indices
113(6)
6.2.1 Cost Model for Keyword Spotting
113(5)
6.2.2 Retrieval Indices
118(1)
6.3 Extensions of the Basic Framework
119(6)
6.3.1 Quadtree Segmentations
120(2)
6.3.2 Fast Rejection Methods
122(1)
6.3.3 Ensemble Methods
123(2)
6.4 Evaluation Metric
125(2)
6.5 Summary
127(2)
7 Experiments 129(70)
7.1 Overview of Experimental Evaluation
129(2)
7.2 Experimental Setup
131(7)
7.2.1 Validation of Parameters
132(6)
7.3 The Baseline KWS System Using BP
138(3)
7.4 Speeding Up the Baseline KWS System
141(15)
7.4.1 KWS System Based on BP with Quadtree Segmentations
142(4)
7.4.2 KWS System Based on BP with Fast Rejection Methods
146(5)
7.4.3 KWS System Based on PGD
151(5)
7.5 KWS System Based on HED
156(9)
7.5.1 Hausdorff Edit Distance
156(5)
7.5.2 Context-aware Hausdorff Edit Distance
161(4)
7.6 KWS System Based on BP2
165(4)
7.7 Graph-Based Ensembles for KWS
169(7)
7.7.1 Ensemble Methods Based on BP
170(5)
7.7.2 Ensemble Methods Based on HED, CED, and BP2
175(1)
7.8 Cross-Evaluation of Graph-Based Systems
176(4)
7.9 Quantitative and Qualitative Summary of Graph-Based Systems
180(9)
7.9.1 Quantitative Summary
180(3)
7.9.2 Qualitative Summary
183(6)
7.10 Comparison of Our Framework with Reference Systems
189(5)
7.10.1 Reference Systems
189(2)
7.10.2 Comparison with Template-Based Reference Systems
191(2)
7.10.3 Comparison with Learning-Based Reference Systems
193(1)
7.11 Summary
194(5)
8 Conclusion & Future Work 199(8)
8.1 Conclusion
199(5)
8.2 Future Work
204(3)
9 Visualisation of Graph Representations 207(12)
9.1 Visualisation of GW
209(2)
9.2 Visualisation of PAR
211(2)
9.3 Visualisation of AK
213(2)
9.4 Visualisation of BOT
215(4)
10 Optimisation of the Parameters 219(6)
11 Ensemble Methods 225(16)
11.1 Ensemble Methods Based on HED
226(4)
11.2 Ensemble Methods Based on CED
230(6)
11.3 Ensemble Methods Based on BP2
236(5)
Bibliography 241(22)
Index 263
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97898112066416e.html
Märksõnad: