Principles of Visual Information Retrieval

by Lew, Michael S.

• ISBN13: 9781852333812
• ISBN10: 1852333812
• Format: Hardcover
• Copyright: 2001-02-01
• Publisher: Springer Verlag

Summary

Principles of Visual Information Retrieval introduces the basic concepts and techniques in VIR and develops a foundation that can be used for further research and study.Divided into 2 parts, the first part describes the fundamental principles. A chapter is devoted to each of the main features of VIR, such as colour, texture and shape-based search. There is coverage of search techniques for time-based image sequences or videos, and an overview of how to combine all the basic features described and integrate context into the search process.The second part looks at advanced topics such as multimedia query, specification, visual learning and semantics, and offers state-of-the-art coverage that is not available in any other book on the market.This book will be essential reading for researchers in VIR, and for final year undergraduate and postgraduate students on courses such as Multimedia Information Retrieval, Multimedia Databases, Computer Vision and Pattern Recognition.

Table of Contents

List of Contributors	xvii
PART I: Fundamental Principles	1	(196)
Visual Information Retrieval: Paradigms, Applications, and Research Issues	3	(8)
Michael S. Lew
Thomas S. Huang
Introduction	3	(1)
Retrieval Paradigms	3	(2)
Applications	5	(2)
Architecture, Real Estate, and Interior Design	6	(1)
Biochemical	6	(1)
Digital Catalog Shopping	6	(1)
Education	7	(1)
Film and Video Archives	7	(1)
Medicine	7	(1)
Research Issues	7	(2)
Summary	9	(2)
References	9	(2)
Color-Based Retrieval	11	(40)
Theo Gevers
Introduction	11	(2)
Color Fundamentals	13	(2)
Color Appearance	15	(9)
The Light Source	16	(2)
The Object	18	(1)
The Observer	19	(5)
Colorimetry	24	(8)
XYZ System	24	(3)
RGB System	27	(1)
HSI System	28	(1)
YIQ and YUV Systems	29	(1)
Color Order Systems	30	(2)
Color Invariance	32	(7)
Reflection from Inhomogeneous Dielectric Materials	33	(1)
Reflectance with White Illumination	34	(3)
Color Constancy	37	(2)
Color System Taxonomy	39	(8)
Gray-Value System	39	(1)
RGB Color System	40	(1)
rgb Color System	40	(1)
XYZ Color System	41	(1)
xyz Color System	42	(1)
U*V*W* Color System	42	(1)
L*a*b* Color System	43	(1)
I1I2I3 Color System	44	(1)
YIQ and YUV Color Systems	45	(1)
HSI Color System	45	(1)
Color Ratios	46	(1)
Color and Image Search Engines	47	(1)
Conclusion	47	(4)
References	48	(3)
Texture Features for Content-Based Retrieval	51	(36)
Nicu Sebe
Michael S. Lew
Introduction	51	(4)
Human Perception of Texture	52	(2)
Approaches for Analyzing Textures	54	(1)
Texture Models	55	(21)
Parametric PDF Methods	57	(5)
Non-Parametric PDF Methods	62	(4)
Harmonic Methods	66	(2)
Primitive Methods	68	(4)
Fractal Methods	72	(2)
Line Methods	74	(1)
Structural Methods	75	(1)
Texture in Content-Based Retrieval	76	(5)
Texture Segmentation	77	(2)
Texture Classification and Indexing	79	(1)
Texture for Annotation	80	(1)
Summary	81	(6)
References	82	(5)
State of the Art in Shape Matching	87	(34)
Remco C. Veltkamp
Michiel Hagedoorn
Introduction	87	(2)
Approaches	89	(8)
Global Image Transforms	90	(1)
Global Object Methods	90	(3)
Voting Schemes	93	(1)
Computational Geometry	94	(3)
Finite Point Sets	97	(3)
Bottleneck Matching	98	(1)
Minimum Weight Matching	99	(1)
Uniform Matching	99	(1)
Minimum Deviation Matching	99	(1)
Hausdorff Distance	99	(1)
Transformation Space Subdivision	100	(1)
Curves	100	(7)
Turning Function	102	(1)
Signature Function	103	(1)
Affine Arclength	104	(1)
Reflection Metric	104	(1)
Hausdorff Distance	105	(1)
Frechet Distance	106	(1)
Size Function	107	(1)
Pixel Chains	107	(1)
Regions	107	(5)
Turning Function	107	(1)
Frechet Distance	108	(1)
Hausdorff Distance	109	(2)
Area of Overlap and Symmetric Difference	111	(1)
Robustness	112	(2)
Software	114	(7)
References	115	(6)
Feature Similarity	121	(24)
Jean-Michel Jolion
Introduction	121	(1)
From Images to Similarity	121	(4)
Pre-attentive vs. Attentive Features and Similarities	121	(1)
Distance vs. Similarity	122	(3)
From Features to Similarity	125	(5)
Complete vs. Partial Feature	125	(1)
Global vs. Local Feature	126	(1)
About Histograms	127	(2)
Global vs. Individual Comparison of Features	129	(1)
Similarity Between Two Sets of Points	130	(3)
Multidimensional Voting	130	(1)
Graph-Based Matching	131	(2)
Similarity Between Histograms	133	(4)
Classic Distances	133	(2)
The Unfolded Distance	135	(1)
The Earth Mover's Distance	135	(2)
Merging Similarities	137	(2)
An Information Retrieval Approach	137	(1)
A Probabilistic Model	138	(1)
Similarity Improvement	139	(2)
Conclusions	141	(4)
References	141	(4)
Feature Selection and Visual Learning	145	(18)
Michael S. Lew
Introduction	145	(1)
Simple Semantics	145	(1)
Complex Backgrounds	146	(1)
Feature Selection	147	(8)
Class Separability	148	(1)
Probabilistic Separation	148	(1)
Mahalanobis Distance	149	(1)
Optimal Search	150	(1)
Sequential Forward Selection (SFS)	151	(1)
Generalized Sequential Forward Selection (GSFS)	151	(1)
Sequential Backward Selection (SBS)	152	(1)
Generalized Sequential Backward Selection (GSBS)	152	(1)
Plus L Take Away R Selection (LRS)	152	(1)
Floating Methods	152	(1)
Stochastic Methods	153	(1)
Neural Network Methods	154	(1)
Direct Methods	154	(1)
Summary of the Feature Selection Algorithms	155	(1)
Case Studies on Performance of Feature Selection Methods	155	(1)
Content-Based Retrieval	156	(3)
Discussion	159	(1)
Summary	160	(3)
References	161	(2)
Video Indexing and Understanding	163	(34)
Michael S. Lew
Nicu Sebe
Paul C. Gardner
Introduction	163	(5)
Video Query Formulation	164	(1)
Video Categorization	164	(1)
Searching	165	(2)
Browsing	167	(1)
Viewing	168	(1)
Video Analysis and Processing	168	(24)
Video Shots	168	(21)
Video Stories	189	(1)
Video Parsing	190	(1)
Video Summaries	191	(1)
Automatic Attributes Generation for Video	191	(1)
Discussion	192	(5)
References	193	(4)
PART II: Advanced Topics	197	(154)
Query Languages for Multimedia Search	199	(20)
Shi-Kuo Chang
Erland Jungert
Introduction	199	(2)
Query Languages for Multimedia Databases	201	(1)
Query Languages for Heterogeneous Multimedia Databases	202	(1)
The σ-Query Language	203	(12)
basic Concepts	204	(2)
A General σ-Operator for σ-Queries	206	(2)
The ΣQL Query Language	208	(1)
Querying for Multisensor Data Fusion	209	(6)
Discussion	215	(4)
References	215	(4)
Relevance Feedback Techniques in Image Retrieval	219	(40)
Yong Rui
Thomas S. Huang
Introduction	219	(3)
The Retrieval Process Based on Relevance Feedback	222	(2)
Parameter Updating: a Heuristic Approach	224	(3)
Update of the Query Vector qi	224	(1)
Update of wi	225	(1)
Update of u	226	(1)
Parameter Updating: Optimal Approaches	227	(4)
Parameter Update at the Representation Level	227	(3)
Practical Considerations	230	(1)
Parameter Update at Both Levels: Optimal Approaches	231	(12)
Quadratic in both φ() and ψi()	231	(6)
Linear in φ() and Quadratic in ψi()	237	(6)
Case Studies	243	(11)
Data Sets	243	(1)
Experiments for Algorithms in Sections 9.3.1 and 9.3.2	244	(1)
Experiments for Algorithm in Section 9.3.3	245	(8)
Experiments for Algorithms in Section 9.5.2	253	(1)
Conclusions	254	(2)
Acknowledgment	256	(3)
References	257	(2)
Mix and Match Features in the ImageRover Search Engine	259	(20)
Stan Sclaroff
Marco La Cascia
Saratendu Sethi
Leonid Taycher
Introduction	259	(2)
Relevance Feedback: Background	261	(1)
Unified Feature Representation	262	(6)
Textual Statistics	262	(2)
Visual Statistics	264	(3)
Combined Vector Space	267	(1)
Dimensionality Reduction	267	(1)
Relevance Feedback	268	(3)
Mixing Features and Distance Metrics	269	(1)
Relevance Feedback Algorithm	270	(1)
System Implementation	271	(2)
User Interface	271	(1)
Example Search	271	(2)
Summary	273	(6)
References	275	(4)
Integrating Analysis of Context and Image Content	279	(18)
Vassilis Athitsos
Charles Frankel
Michael J. Swain
Introduction	279	(1)
Extracting Relevant Text	279	(3)
Relevant Text in HTML Documents	280	(2)
Other Document Formats	282	(1)
Image Classification: Photographs vs. Graphics	282	(12)
What are Photographs and Graphics?	283	(1)
Differences Between Photographs and Graphics	283	(1)
Image Metrics	284	(4)
Constructing the Decision Trees	288	(1)
Preparation of Training and Testing Sets	288	(1)
Reasons for Using Multiple Decision Trees	289	(1)
Results	290	(1)
Improved Learning Techniques	291	(1)
Image Context Metrics	291	(1)
Mixed Images	292	(1)
Other Approaches	293	(1)
Other Types of Image Classification	294	(1)
Detecting Portraits	294	(1)
Distinguishing Color from Gray-Scale Images	294	(1)
Filtering Pornographic Images	294	(1)
User Interface Issues in Images Search	295	(2)
References	295	(2)
Semantic-Based Retrieval of Visual Data	297	(22)
Clement Leung
Simon So
Audrey Tam
Dwi Sutanto
Philip Tse
Introduction	297	(1)
Two Kinds of Visual Contents	297	(2)
Primitive Contents	297	(1)
Complex Contents	298	(1)
Levels of Complex Contents	298	(1)
Multiple Search Space Pruning	299	(4)
Using Feature Vectors for Primitive Content Search	303	(3)
Inverted Image Indexing Using Composite Bitplane Signature	306	(3)
Using Semantic Data Models for Complex Content Search	309	(4)
The Basic Ternary Fact Model	310	(1)
Natural Language Indexing	311	(1)
Logical and Physical Index Structure	312	(1)
Discussion and Future Work	313	(6)
References	316	(3)
Trademark Image Retrieval	319	(32)
John P. Eakins
Introduction	319	(1)
Trademark Search and Registration	320	(5)
General Principles	320	(1)
Searching for Word Marks	320	(1)
Searching for Device (Image) Marks	321	(2)
Other Applications of Trademark Image Matching	323	(1)
System Requirements for Trademark Image Retrieval	323	(2)
Techniques for Image Retrieval	325	(8)
The Growth of CBIR	325	(1)
Techniques for Shape Retrieval	325	(4)
Other Types of Image Retrieval Techniques	329	(1)
Retrieval Efficiency	330	(1)
Effectiveness of CBIR Techniques	331	(2)
Trademark Image Retrieval Systems	333	(11)
System Design Issues	333	(1)
Manually Based Techniques for Similarity Retrieval	334	(1)
CBIR Techniques for Similarity Retrieval	334	(9)
Logo Recognition	343	(1)
Conclusions	344	(7)
Effectiveness of Current Techniques	344	(1)
Future Prospects	345	(1)
References	346	(5)
Author Index	351	(2)
Subject Index	353