Michaelsen / Meidow Hierarchical Perceptual Grouping for Object Recognition

Theoretical Views and Gestalt-Law Applications
1. Auflage 2019
ISBN: 978-3-030-04040-6
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 200 Seiten

Reihe: Advances in Computer Vision and Pattern Recognition

This unique text/reference presents a unified approach to the formulation of Gestalt laws for perceptual grouping, and the construction of nested hierarchies by aggregation utilizing these laws. The book also describes the extraction of such constructions from noisy images showing man-made objects and clutter. Each Gestalt operation is introduced in a separate, self-contained chapter, together with application examples and a brief literature review. These are then brought together in an algebraic closure chapter, followed by chapters that connect the method to the data - i.e., the extraction of primitives from images, cooperation with machine-readable knowledge, and cooperation with machine learning.Topics and features: offers the first unified approach to nested hierarchical perceptual grouping; presents a review of all relevant Gestalt laws in a single source; covers reflection symmetry, frieze symmetry, rotational symmetry, parallelism and rectangular settings, contour prolongation, and lattices; describes the problem from all theoretical viewpoints, including syntactic, probabilistic, and algebraic perspectives; discusses issues important to practical application, such as primitive extraction and any-time search; provides an appendix detailing a  general adjustment model with constraints.This work offers new insights and proposes novel methods to advance the field of machine vision, which will be of great benefit to students, researchers, and engineers active in this area.

Dr.-Ing. Eckart Michaelsen is a researcher at the Object Recognition Department of Fraunhofer IOSB, Ettlingen, Germany.Dr.-Ing. Jochen Meidow is a researcher at the Scene Analysis Department of the same institution.

Autoren/Hrsg.

Michaelsen, Eckart

Meidow, Jochen

Inhaltsverzeichnis

1;Preface;6
2;Contents;8
3;Notations;12
4;1 Introduction;13
4.1;1.1 Examples of Pictures with Hierarchical Gestalt;13
4.2;1.2 The State of the Art of Automatic Symmetry and Gestalt Recognition;17
4.3;1.3 The Gestalt Domain;23
4.4;1.4 Assessments for Gestalten;26
4.5;1.5 Statistically Best Mean Direction or Axis;30
4.6;1.6 The Structure of this Book;31
4.7;References;33
5;2 Reflection Symmetry;35
5.1;2.1 Introduction to Reflection Symmetric Gestalten;35
5.2;2.2 The Reflection Symmetry Constraint as Defined for Extracted Primitive Objects;37
5.3;2.3 Reformulation of the Constraint as a Continuous Score Function;39
5.4;2.4 Optimal Fitting of Reflection Symmetry Aggregate Features;41
5.5;2.5 The Role of Proximity in Evidence for Reflection Symmetry;43
5.6;2.6 The Role of Similarity in Evidence for Reflection Symmetry and How to Combine the Evidences;45
5.7;2.7 Nested Symmetries Reformulated as Successive Scoring on Rising Scale;47
5.8;2.8 Clustering Reflection Symmetric Gestalten with Similar Axes;53
5.9;2.9 The Theory of A Contrario Testing and its Application to Finding Reflection Symmetric Patches in Images;58
5.10;2.10 The Minimum Description Length Approach for Nested Reflection Symmetry;60
5.11;2.11 Projective Symmetry;60
5.12;References;62
6;3 Good Continuation in Rows or Frieze Symmetry;64
6.1;3.1 Related Work on Row Gestalt Grouping;66
6.2;3.2 The Row Gestalt as Defined on Locations;67
6.3;3.3 Proximity for Row Gestalten;69
6.4;3.4 The Role of Similarity in Row Gestalten;70
6.4.1;3.4.1 Vector Features;71
6.4.2;3.4.2 Scale Features;73
6.4.3;3.4.3 Orientation Features;74
6.5;3.5 Sequential Search;75
6.5.1;3.5.1 The Combinatorics of Row Gestalten;75
6.5.2;3.5.2 Greedy Search for Row Prolongation;76
6.6;3.6 The A Contrario Approach to Row Grouping;78
6.7;3.7 Perspective Foreshortening of Rows;78
6.8;References;80
7;4 Rotational Symmetry;82
7.1;4.1 The Rotational Gestalt Law as Defined on Locations;83
7.2;4.2 Fusion with Other Gestalt Laws;86
7.2.1;4.2.1 Proximity Assessments for Rotational Gestalten;86
7.2.2;4.2.2 Similarity Assessments for Rotational Gestalten;88
7.3;4.3 Search for Rotational Gestalten;89
7.3.1;4.3.1 Greedy Search for Rotational Gestalten;89
7.3.2;4.3.2 A Practical Example with Rotational Gestalten of Level 1;90
7.4;4.4 The Rotational Group and the Dihedral on Group;93
7.5;4.5 Perspective Foreshortening of Rotational Gestalts;93
7.6;References;95
8;5 Closure—Hierarchies of Gestalten;96
8.1;5.1 Gestalt Algebra;97
8.2;5.2 Empirical Experiments with Closure;101
8.3;5.3 Transporting Evidence through Gestalt Algebra Terms;103
8.3.1;5.3.1 Considering Additional Features;104
8.3.2;5.3.2 Propagation of Adjustments through the Hierarchy;106
8.4;References;111
9;6 Search;112
9.1;6.1 Stratified Search;112
9.2;6.2 Recursive Search;113
9.3;6.3 Monte Carlo Sampling with Preferences;114
9.4;6.4 Any-time Search Using a Blackboard;115
9.5;References;116
10;7 Illusions;118
10.1;7.1 Literature about Illusions in Seeing;118
10.2;7.2 Deriving Illusion from Top-down Search;119
10.3;7.3 Illusion as Tool to Counter Occlusion;119
10.4;References;120
11;8 Prolongation in Good Continuation;121
11.1;8.1 Related Work on Contour Chaining, Line Prolongation, and Gap Filling;122
11.2;8.2 Tensor Voting;122
11.3;8.3 The Linear Prolongation Law and Corresponding Assessment Functions;126
11.4;8.4 Greedy Search for Maximal Line Prolongation and Gap Closing;131
11.5;8.5 Prolongation in Good Continuation as Control Problem;131
11.6;8.6 Illusory Contours at Line Ends;133
11.7;References;135
12;9 Parallelism and Rectangularity;136
12.1;9.1 Close Parallel Contours;136
12.2;9.2 Drawing on Screens as Graphical User Interface;138
12.3;9.3 Orthogonality and Parallelism for Polygons;139
12.4;References;142
13;10 Lattice Gestalten;143
13.1;10.1 Related Work on Lattice Grouping;144
13.2;10.2 The Lattice Gestalt as Defined on Locations;144
13.3;10.3 The Role of Similarity in Lattice Gestalt Grouping;146
13.4;10.4 Searching for Lattices;147
13.5;10.5 An Example from SAR Scatterers;149
13.6;10.6 Projective Distortion;151
13.7;References;151
14;11 Primitive Extraction;153
14.1;11.1 Threshold Segmentation;154
14.2;11.2 Super-Pixel Segmentation;156
14.3;11.3 Maximally Stable Extremal Regions;158
14.4;11.4 Scale-Invariant Feature Transform;160
14.5;11.5 Multimodal Primitives;162
14.6;11.6 Segmentation by Unsupervised Machine Learning;162
14.6.1;11.6.1 Learning Characteristic Colors from a Standard Three Bytes Per Pixel Image;163
14.6.2;11.6.2 Learning Characteristic Spectra from a Hyper-Spectral Image;164
14.7;11.7 Local Non-maxima Suppression;167
14.8;References;169
15;12 Knowledge and Gestalt Interaction;170
15.1;12.1 Visual Inference;170
15.2;12.2 A Small Review on Knowledge-Based Image Analysis;173
15.3;12.3 An Example from Remotely Sensed Hyper-spectral Imagery;176
15.4;12.4 An Example from Synthetic Aperture RADAR Imagery;178
15.5;References;180
16;13 Learning;181
16.1;13.1 Labeling of Imagery for Evaluation and Performance Improvement;181
16.2;13.2 Learning Assessment Weight Parameters;184
16.3;13.3 Learning Proximity Parameters with Reflection Ground Truth;185
16.4;13.4 Assembling Orientation Statistics with Frieze Ground Truth;187
16.5;13.5 Estimating Parametric Mixture Distributions from Orientation Statistics;189
16.6;References;193
17;A General Adjustment Model with Constraints;195
17.1;References;197
18;Index;198