Scale-Space Theories in Computer Vision

This volume constitutes the refereed proceedings of the Second International Conference on Scale-Space Theories in Computer Vision, Scale-Space'99, held in Corfu, Greece, in September 1999.
The 36 revised full papers and the 18 revised posters presented in the book were carefully reviewed and selected from 66 high-quality submissions. The book addresses all current aspects of this young and active field, in particular geometric Image flows, nonlinear diffusion, functional minimization, linear scale-space, etc.

Inhalt

Oral Presentations.- Blur and Disorder.- Applications of Locally Orderless Images.- Scale Space Technique for Word Segmentation in Handwritten Documents.- Fast Geodesic Active Contours.- Morphing Active Contours.- Unfolding the Cerebral Cortex Using Level Set Methods.- Reconciling Distance Functions and Level Sets.- Computation of Ridges via Pullback Metrics from Scale Space.- The Maximal Scale Ridge - Incorporating Scale into the Ridge Definition.- Detection of Critical Structures in Scale Space.- Qualitative Multi-scale Feature Hierarchies for Object Tracking.- Riemannian Drums, Anisotropic Curve Evolution and Segmentation.- An Active Contour Model without Edges.- A Compact and Multiscale Image Model Based on Level Sets.- Morphological Scale Space and Mathematical Morphology.- Scale-Space from a Level Lines Tree.- Morphological Scale-Space Representation with Levelings.- Numerical Solution Schemes for Continuous-Scale Morphology.- Scale-Space Properties of Regularization Methods.- An Adaptive Finite Element Method for Large Scale Image Processing.- A Scale-Space Approach to Nonlocal Optical Flow Calculations.- Scales in Natural Images and a Consequence on Their Bounded Variation Norm.- Edges as Outliers: Anisotropic Smoothing Using Local Image Statistics.- The Hausdorff Dimension and Scale-Space Normalisation of Natural Images.- Long Posters.- Lattice Boltzmann Models for Nonlinear Diffusion Filtering.- Geometric-Variational Approach for Color Image Enhancement and Segmentation.- A Level Set Model for Image Classification.- Calculations on Critical Points under Gaussian Blurring.- Region Tracking on Surfaces Deforming via Level-Sets Methods.- Geometric Multiscale Representation of Numerical Images.- Multiscale Morphological Segmentations Based on Watershed, Flooding, and Eikonal PDE.- Nonlinear PDEs and Numerical Algorithms for Modeling Levelings and Reconstruction Filters.- Proper Scales of Shapes - A Curved Scale Space.- Nonlinear Anisotropic Diffusion in Three-Dimensional Electron Microscopy.- Polygon Evolution by Vertex Deletion.- A Scale-Space Based Approach for Deformable Contour Optimization.- Short Posters.- Self-similarity of Noise in Scale-Space.- A New Time Dependent Model Based on Level Set Motion for Nonlinear Deblurring and Noise Removal.- Curvature Scale Space with Affine Length Parametrisation.- A Stochastic Scale Space for Multiscale Image Representation.- Fast Marching to Moving Object Location.- A Windows-Based User Friendly System for Image Analysis with Partial Differential Equations.- Color Invariant Edge Detection.- Scale Space Range Image Aspect Graph.- Hierachical Density Non-rigid Object Analysis.- Discrete Mean Curvature Flow.- An Adaptive Local Smoothing for Contour Figure Approximation.- Negative Results for Multilevel Preconditioners in Image Deblurring.- Decomposition and Hierarchy: Efficient StructuralMatching of Large Multiscale Representations.- Tracing of Curvilinear Structures in 3D Images with Single Scale Diameter Measurement.- A Geometric Functional for Derivatives Approximation.- Segmenting by Compression Using Linear Scale-Space and Watersheds.- A Multiscale Taylor Series Approaches to Optic Flow and Stereo: A Generalization of Optic Flow under the Aperture.- Global Convergence Rates of Nonlinear Diffusion for Time-Varying Images.