Descombes Stochastic Geometry for Image Analysis

1. Auflage 2013
ISBN: 978-1-118-60113-6
Verlag: John Wiley & Sons
Format: EPUB
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 384 Seiten, E-Book

This book develops the stochastic geometry framework for imageanalysis purpose. Two main frameworks are described: markedpoint process and random closed sets models. We derive the mainissues for defining an appropriate model. The algorithms forsampling and optimizing the models as well as for estimatingparameters are reviewed. Numerous applications, coveringremote sensing images, biological and medical imaging, aredetailed. This book provides all the necessary tools fordeveloping an image analysis application based on modern stochasticmodeling.

Autoren/Hrsg.

Descombes, Xavier

Inhaltsverzeichnis

Chapter 1. Introduction 1
X. DESCOMBES
Chapter 2. Marked Point Processes for Object Detection11
X. DESCOMBES
2.1. Principal definitions 11
2.2. Density of a point process 15
2.3. Marked point processes 21
2.4. Point processes and image analysis 22
Chapter 3. Random Sets for Texture Analysis 29
C. LANTUÉJOUL and M. SCHMITT
3.1. Introduction 29
3.2. Random sets 33
3.3. Some geostatistical aspects 42
3.4. Some morphological aspects 51
3.5. Appendix: demonstration of Miles' formulae for theBoolean model 61
Chapter 4. Simulation and Optimization 65
F. LAFARGE, X. DESCOMBES, E. ZHIZHINA and R. MINLOS
4.1. Discrete simulations: Markov chain Monte Carlo algorithms66
4.2. Continuous simulations 91
4.3. Mixed simulations 105
4.4. Simulated annealing 106
Chapter 5. Parametric Inference for Marked Point Processes inImage Analysis 113
R. STOICA, F. CHATELAIN and M. SIGELLE
5.1. Introduction 113
5.2. First question: what and where are the objects in theimage? 117
5.3. Second question: what are the parameters of the pointprocess that models the objects observed in the image? 129
5.4. Conclusion and perspectives 158
5.5. Acknowledgments 159
Chapter 6. How to Set Up a Point Process? 161
X. DESCOMBES
6.1. From disks to polygons, via a discussion of segments162
6.2. From no overlap to alignment 167
6.3. From the likelihood to a hypothesis test 172
6.4. From Metropolis-Hastings to multiple births anddeaths 176
Chapter 7. Population Counting 179
X. DESCOMBES
7.1. Detection of Virchow-Robin spaces 180
7.2. Evaluation of forestry resources 192
7.3. Counting a population of flamingos 207
7.4. Counting the boats at a port 229
Chapter 8. Structure Extraction 249
F. LAFARGE and X. DESCOMBES
8.1. Detection of the road network 250
8.2. Extraction of building footprints 262
8.3. Representation of natural textures 269
Chapter 9. Shape Recognition 287
F. LAFARGE and C. MALLET
9.1. Modeling of a LIDAR signal 287
9.2. 3D reconstruction of buildings 308
Bibliography 325
List of Authors 341
Index 343