Segmentation of Stochastic Images using ... - Jacobs University
Segmentation of Stochastic Images using ... - Jacobs University
Segmentation of Stochastic Images using ... - Jacobs University
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Contents<br />
Acknowledgement<br />
Abstract<br />
Notation<br />
iii<br />
v<br />
ix<br />
1 Introduction 1<br />
2 Image <strong>Segmentation</strong> and Limitations 7<br />
2.1 Mathematical <strong>Images</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8<br />
2.2 Random Walker <strong>Segmentation</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9<br />
2.3 Mumford-Shah and Ambrosio-Tortorelli <strong>Segmentation</strong> . . . . . . . . . . . . . . . . 12<br />
2.4 Level Sets for Image <strong>Segmentation</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . 16<br />
2.5 Why is Classical Image Processing not Enough? . . . . . . . . . . . . . . . . . . . . 21<br />
2.6 Work Related to the <strong>Stochastic</strong> Framework . . . . . . . . . . . . . . . . . . . . . . . 23<br />
3 SPDEs and Polynomial Chaos Expansions 25<br />
3.1 Basics from Probability Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25<br />
3.2 <strong>Stochastic</strong> Partial Differential Equations . . . . . . . . . . . . . . . . . . . . . . . . 26<br />
3.3 Polynomial Chaos Expansions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29<br />
3.4 Relation to Interval Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35<br />
4 Discretization <strong>of</strong> SPDEs 37<br />
4.1 Sampling Based Discretization <strong>of</strong> SPDEs . . . . . . . . . . . . . . . . . . . . . . . 37<br />
4.2 <strong>Stochastic</strong> Finite Difference Methods . . . . . . . . . . . . . . . . . . . . . . . . . . 38<br />
4.3 <strong>Stochastic</strong> Finite Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39<br />
4.4 Generalized Spectral Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . 40<br />
4.5 Adaptive Grids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45<br />
5 <strong>Stochastic</strong> <strong>Images</strong> 47<br />
5.1 Polynomial Chaos for <strong>Stochastic</strong> <strong>Images</strong> . . . . . . . . . . . . . . . . . . . . . . . . 47<br />
5.2 Generation <strong>of</strong> <strong>Stochastic</strong> <strong>Images</strong> from Samples . . . . . . . . . . . . . . . . . . . . 48<br />
5.3 Comparison <strong>of</strong> the Space from [130] and the Space Used in this Thesis . . . . . . . . 52<br />
5.4 Visualization <strong>of</strong> <strong>Stochastic</strong> <strong>Images</strong> . . . . . . . . . . . . . . . . . . . . . . . . . . . 53<br />
6 <strong>Segmentation</strong> <strong>of</strong> <strong>Stochastic</strong> <strong>Images</strong> Using Elliptic SPDEs 57<br />
6.1 Random Walker <strong>Segmentation</strong> on <strong>Stochastic</strong> <strong>Images</strong> . . . . . . . . . . . . . . . . . 57<br />
6.2 Ambrosio-Tortorelli <strong>Segmentation</strong> on <strong>Stochastic</strong> <strong>Images</strong> . . . . . . . . . . . . . . . 67<br />
7 <strong>Stochastic</strong> Level Sets 79<br />
7.1 Derivation <strong>of</strong> a <strong>Stochastic</strong> Level Set Equation . . . . . . . . . . . . . . . . . . . . . 79<br />
7.2 Discretization <strong>of</strong> the <strong>Stochastic</strong> Level Set Equation . . . . . . . . . . . . . . . . . . 83<br />
7.3 Reinitialization <strong>of</strong> <strong>Stochastic</strong> Level Sets . . . . . . . . . . . . . . . . . . . . . . . . 84<br />
7.4 Numerical Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85<br />
vii