| |
| |
| Preface | |
| |
| |
| |
| Introduction | |
| |
| |
| |
| General introduction | |
| |
| |
| |
| Image processing tasks | |
| |
| |
| |
| Statistical decision and estimation theory | |
| |
| |
| |
| An application-oriented approach | |
| |
| |
| |
| Outline of the book | |
| |
| |
| |
| Linear Filters: Heuristic Theory and Stability | |
| |
| |
| |
| The different approaches to filter design | |
| |
| |
| |
| Heuristic criteria and optimal filters | |
| |
| |
| |
| Noise robustness characterization and matched filter | |
| |
| |
| |
| Sharpness of the correlation peak and inverse filter | |
| |
| |
| |
| Optical efficiency and phase-only filter | |
| |
| |
| |
| Discrimination capability | |
| |
| |
| |
| Optimal SDF filters | |
| |
| |
| |
| Optimal Trade-off filters | |
| |
| |
| |
| Optimal Trade-off SDF filters | |
| |
| |
| |
| Analysis of the stability of linear filters | |
| |
| |
| |
| Regularization of filters | |
| |
| |
| |
| Truncature method for regularization | |
| |
| |
| |
| Stabilizing functional | |
| |
| |
| |
| Some processing examples with stabilized filters | |
| |
| |
| |
| An example of application: Angle estimation of two-dimensional objects | |
| |
| |
| |
| Conclusion | |
| |
| |
| |
| Appendix--Definitions and notation | |
| |
| |
| |
| Appendix--Lagrange multipliers | |
| |
| |
| |
| Statistical Correlation Techniques | |
| |
| |
| |
| Some sources of noise in images | |
| |
| |
| |
| Nonoverlapping noise | |
| |
| |
| |
| Fluctuations of the target's gray levels | |
| |
| |
| |
| Additive noise with unknown PSD | |
| |
| |
| |
| Background on statistical decision and estimation theory | |
| |
| |
| |
| Decision and estimation theory without nuisance parameters | |
| |
| |
| |
| Decision and estimation theory in presence of nuisance parameters | |
| |
| |
| |
| Two-hypothesis testing | |
| |
| |
| |
| Matched filtering and statistical decision theory | |
| |
| |
| |
| Optimal filter for unknown PSD | |
| |
| |
| |
| ML estimation of the spectral density | |
| |
| |
| |
| MAP estimation of the spectral density | |
| |
| |
| |
| Marginal Bayesian approach | |
| |
| |
| |
| Examples of application | |
| |
| |
| |
| Target location in nonoverlapping noise | |
| |
| |
| |
| The SIR image model and optimal location algorithms | |
| |
| |
| |
| Targets with known graylevels | |
| |
| |
| |
| Targets with fluctuating graylevels | |
| |
| |
| |
| Partially fluctuating targets | |
| |
| |
| |
| Conclusion | |
| |
| |
| |
| Appendix--MAP location algorithm in the presence of uniform prior | |
| |
| |
| |
| Applications of Statistical Correlation Techniques to Different Physical Noises | |
| |
| |
| |
| A general framework for designing image processing algorithms | |
| |
| |
| |
| Generalization of the SIR image model | |
| |
| |
| |
| The exponential family | |
| |
| |
| |
| Performing object location with algorithms based on the SIR image model | |
| |
| |
| |
| The whitening process | |
| |
| |
| |
| The generalized likelihood ratio test (GLRT) approach | |
| |
| |
| |
| The implementation issue | |
| |
| |
| |
| Application to binary images: Comparison of optimal and linear techniques | |
| |
| |
| |
| The GLRT algorithm for binary images | |
| |
| |
| |
| A linear approximation to the GLRT algorithm | |
| |
| |
| |
| Application to edge extraction in SAR images | |
| |
| |
| |
| GLRT adapted to speckled images | |
| |
| |
| |
| Bias on edge location | |
| |
| |
| |
| Conclusion | |
| |
| |
| |
| Appendix--Basics of estimation theory | |
| |
| |
| |
| Statistical Snake-based Segmentation Adapted to Different Physical Noises | |
| |
| |
| |
| Active contours | |
| |
| |
| |
| Snake energy | |
| |
| |
| |
| The limits of the classical snake | |
| |
| |
| |
| Geodesic snakes | |
| |
| |
| |
| The level set implementation of snakes | |
| |
| |
| |
| Region-based approaches | |
| |
| |
| |
| The SIR Active Contour and its fast implementation | |
| |
| |
| |
| Solution for exponential family laws | |
| |
| |
| |
| Implementation of a fast statistic calculation | |
| |
| |
| |
| Application to polygonal active contour | |
| |
| |
| |
| Regularization of the contour | |
| |
| |
| |
| Minimization procedure | |
| |
| |
| |
| Discussion | |
| |
| |
| |
| Some examples of application | |
| |
| |
| |
| Applications to tracking in video sequences | |
| |
| |
| |
| Fixed camera | |
| |
| |
| |
| Moving camera | |
| |
| |
| |
| Application to accuracy improvement of edge location | |
| |
| |
| |
| Appendix--Crossing tests | |
| |
| |
| |
| Some Developments of the Polygonal Statistical Snake and Their Applications | |
| |
| |
| |
| Generalization of the statistical snake to multichannel images | |
| |
| |
| |
| MDL-based statistical snake | |
| |
| |
| |
| The MDL principle | |
| |
| |
| |
| Application of the MDL principle to the polygonal statistical snake | |
| |
| |
| |
| Two-step optimization process | |
| |
| |
| |
| Results obtained with different types of noises | |
| |
| |
| |
| Quantitative evaluation of the segmentation performance | |
| |
| |
| |
| Statistical active grid and application to SAR image segmentation | |
| |
| |
| |
| Statistical active grid | |
| |
| |
| |
| Implementation issues | |
| |
| |
| |
| Some segmentation examples | |
| |
| |
| |
| Conclusion | |
| |
| |
| |
| An Example of Application: Processing of Coherent Polarimetric Images | |
| |
| |
| |
| Basics of polarimetric imaging | |
| |
| |
| |
| The representation of polarized light | |
| |
| |
| |
| Active polarimetric imaging systems | |
| |
| |
| |
| Model of coherent polarimetric images | |
| |
| |
| |
| Processing degree of polarization (DOP) images | |
| |
| |
| |
| Principle of DOP imaging | |
| |
| |
| |
| Influence of illumination nonuniformity on segmentation performance | |
| |
| |
| |
| The statistics of the OSCI and its natural representation | |
| |
| |
| |
| Speckle and multiplicative noise | |
| |
| |
| |
| The probability density function of the OSCI | |
| |
| |
| |
| The natural representation of the OSCI | |
| |
| |
| |
| Applications to image processing of the OSCI | |
| |
| |
| |
| Target and edge detection | |
| |
| |
| |
| Statistical snake segmentation of OSCI | |
| |
| |
| |
| Defining a contrast in Stokes images | |
| |
| |
| |
| Position of the problem | |
| |
| |
| |
| Contrast parameters for coherent polarimetric signals | |
| |
| |
| |
| Detection and segmentation in Stokes images | |
| |
| |
| |
| Target detection/localization | |
| |
| |
| |
| Statistical snake-based segmentation | |
| |
| |
| |
| Contrast parameters and detection performance | |
| |
| |
| |
| Conclusion | |
| |
| |
| |
| Appendix--Statistical properties of the OSCI | |
| |
| |
| |
| Appendix--GLRT and statistical snake for Gaussian noise with common variance | |
| |
| |
| |
| Appendix--Interpretation of the contrast parameters | |
| |
| |
| Credits | |
| |
| |
| References | |
| |
| |
| Index | |