Edit by Dalton Lin Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 1 Vision, the Challenge.

Slides:



Advertisements
Similar presentations
1 Copyright © 2010, Elsevier Inc. All rights Reserved Fig 4.1 Chapter 4.
Advertisements

Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 9 Binary Shape Analysis.
Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 19 Motion.
1 Chapter 24 - Renal Cortical and Medullary Microcirculations: Structure and Function Copyright © 2013 Elsevier Inc. All rights reserved.
Modeling Constraints with Parametrics
Copyright © 2016 Elsevier Inc. All rights reserved.
Chapter 41 Work-Related Musculo-Skeletal Disorders
Data Mining, Second Edition, Copyright © 2006 Elsevier Inc.
Chapter 65 - The Hormonal Regulation of Calcium Metabolism
Copyright © 2012, Elsevier Inc. All rights Reserved.
Chapter 11.
The University of Adelaide, School of Computer Science
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
The University of Adelaide, School of Computer Science
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
The University of Adelaide, School of Computer Science
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 10.
Chapter 28 - Renal Hyperplasia and Hypertrophy
Copyright © 2014, 2000, 1992 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Analog-to-Digital Conversion
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
The University of Adelaide, School of Computer Science
Copyright © 2013 Elsevier Inc. All rights reserved.
The University of Adelaide, School of Computer Science
The University of Adelaide, School of Computer Science
Modeling Cross-Cutting Relationships with Allocations
Portable Biotechnology
© 2012 Elsevier, Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Modeling Text-Based Requirements and their Relationship to Design
Modeling Functionality with Use Cases
Customizing SysML for Specific Domains
Copyright © 2012, Elsevier Inc. All rights Reserved.
The University of Adelaide, School of Computer Science
Copyright © 2013 Elsevier Inc. All rights reserved.
IntroductionMolecular Structure and Reactivity
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 12.
Chapter 6.
Copyright © 2012, Elsevier Inc. All rights Reserved.
© 2012 Elsevier, Inc. All rights reserved.
Chapter 103 Long-Term Care: The Global Impact
Chapter 01.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Modeling Message-Based Behavior with Interactions
Chapter 08.
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 64 - Renal Calcium Metabolism
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 57 - Renal Ammonium Ion Production and Excretion
© 2015 Elsevier, Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Chapter 15 Contraception
Chapter 29 - Stem Cells and Generation of New Cells in the
Chapter 78 Vaccination and Screening in Cervical Cancer
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 15.
Chapter 77 Cervical Cancer: Burden of Disease
Chapter 20 Assisted Reproductive Technologies
Chapter 3.
© 2015 Elsevier, Inc. All rights reserved.
Presentation transcript:

Edit by Dalton Lin Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 1 Vision, the Challenge

1.1 Introduction – Man and his Senses Vision provides most of the data human receives. Not only do the input pathways from the eyes provide megabits of information at each glance but the data rates for continuous viewing probably exceed 10 megabits per second (mbits/s). Another feature of human visual system is the ease with which interpretation is carried out. In the present day and age, man is trying to get machines to do much of his work for him. Copyright © 2012 Elsevier Inc. All rights reserved.. 2

1.2 The nature of Vision Tackling the Recognition Problem An obvious means of tackling the recognition problem is to standardize the images in some way. Clearly, normalizing the position and orientation of any 2-D picture object would help considerably. Here, we can make additional use of what is known about the structure of characters -- namely, that they consist of limbs of roughly constant width. So the patterns can be thinned to stick figures (called skeletons). Copyright © 2012 Elsevier Inc. All rights reserved.. 3

Copyright © 2012 Elsevier Inc. All rights reserved.. FIGURE 1.1 Some simple 25-bit patterns and their recognition classes used to illustrate some of the basic problems of recognition: (a) training set patterns (for which the known classes are indicated); (b) test patterns.

Copyright © 2012 Elsevier Inc. All rights reserved.. FIGURE 1.2 Use of thinning to regularize character shapes. Here, character shapes of different limb widths - or even varying limb widths - are reduced to stick figures or skeletons. Thus, irrelevant information is removed and at the same time recognition is facilitated. Then, hopefully, there is an even greater chance that the test patterns will be similar to appropriate training set patterns.

Recognition is necessarily a problem of discriminating between patterns of different classes. There is also a problem of generalizing over patterns of the same class. First, objects were positioned and orientated appropriately. Second, they were cleaned of noise spots. Third, they were thinned to skeleton figures with the same width. Copyright © 2012 Elsevier Inc. All rights reserved.. 6

Copyright © 2012 Elsevier Inc. All rights reserved.. FIGURE 1.3 Two-Stage Recognition Paradigm

Copyright © 2012 Elsevier Inc. All rights reserved.. Search is a task that can involve prodigious amounts of computation and which is also subject to a combinatorial explosion Object Location

Formula 1.1 Copyright © 2012 Elsevier Inc. All rights reserved.. 9 One of the most natural ways of checking for a match is to measure the Hamming distance between the template and the local nxn region of the image. The places with a low Hamming distance are places where the match is good.

1.2.4 Scene Analysis Scenes contain a multitude of objects, and it is their interrelationships and relative positions that matter as much as identifying what they are. Scene analysis involves enormous amounts of input data, complex relationships between objects within scenes and, ultimately, descriptions of these complex relationships. Copyright © 2012 Elsevier Inc. All rights reserved.. 10