Feature extraction image processing 2nd Edition

Mark Nixon pdf download

https://ebookfinal.com/download/feature-extraction-image-
processing-2nd-edition-mark-nixon/

Explore and download more ebooks or textbooks

at ebookfinal.com
 We believe these products will be a great fit for you. Click
the link to download now, or visit ebookfinal
to discover even more!

Feature Extraction Image Processing for Computer Vision

3ed. Edition Mark Nixon

https://ebookfinal.com/download/feature-extraction-image-processing-
for-computer-vision-3ed-edition-mark-nixon/

Digital Image Processing preview 2nd Edition Rafael C.

Gonzalez

https://ebookfinal.com/download/digital-image-processing-preview-2nd-
edition-rafael-c-gonzalez/

Software Engineering for Image Processing Systems Image

Processing Series 1st Edition Philip A. Laplante

https://ebookfinal.com/download/software-engineering-for-image-
processing-systems-image-processing-series-1st-edition-philip-a-
laplante/

Digital Image Processing Using Matlab 2nd Edition Rafael

C. Gonzalez

https://ebookfinal.com/download/digital-image-processing-using-
matlab-2nd-edition-rafael-c-gonzalez/
Digital Image Processing 1st Edition S.K. Ghosh

https://ebookfinal.com/download/digital-image-processing-1st-edition-
s-k-ghosh/

Field Guide to Image Processing Khan Iftekharuddin

https://ebookfinal.com/download/field-guide-to-image-processing-khan-
iftekharuddin/

Image Processing The Fundamentals Second Edition Maria

Petrou

https://ebookfinal.com/download/image-processing-the-fundamentals-
second-edition-maria-petrou/

Digital Color Image Processing 1st Edition Andreas Koschan

https://ebookfinal.com/download/digital-color-image-processing-1st-
edition-andreas-koschan/

Digital image processing for medical applications

Dougherty G.

https://ebookfinal.com/download/digital-image-processing-for-medical-
applications-dougherty-g/
Feature extraction image processing 2nd Edition Mark
Nixon Digital Instant Download
Author(s): Mark Nixon, Alberto S Aguado
ISBN(s): 9780123725387, 0123725380
Edition: 2
File Details: PDF, 7.72 MB
Year: 2008
Language: english
Feature Extraction
and
Image Processing
 Dedication
We would like to dedicate this book to our parents:
To Gloria and Joaquin Aguado, and to Brenda and the late Ian Nixon.
Feature Extraction
and
Image Processing
Second edition

Mark S. Nixon
Alberto S. Aguado

Amsterdam • Boston • Heidelberg • London • New York • Oxford

Paris • San Diego • San Francisco • Singapore • Sydney • Tokyo

Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier
Linacre House, Jordan Hill, Oxford OX2 8DP, UK
84 Theobald’s Road, London WC1X 8RR, UK

First edition 2002

Reprinted 2004, 2005
Second edition 2008

Copyright © 2008 Elsevier Ltd. All rights reserved

No part of this publication may be reproduced, stored in a retrieval system

or transmitted in any form or by any means electronic, mechanical, photocopying,
recording or otherwise without the prior written permission of the publisher

Permissions may be sought directly from Elsevier’s Science & Technology Rights
Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333;
email: permissions@elsevier.com. Alternatively you can submit your request online by
visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting
Obtaining permission to use Elsevier material

Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons
or property as a matter of products liability, negligence or otherwise, or from any use
or operation of any methods, products, instructions or ideas contained in the material
herein. Because of rapid advances in the medical sciences, in particular, independent
verification of diagnoses and drug dosages should be made

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

ISBN: 978-0-12372-538-7

For information on all Academic Press publications

visit our web site at books.elsevier.com

Printed and bound in Hungary

08 09 10 10 9 8 7 6 5 4 3 2 1

Working together to grow

libraries in developing countries
www.elsevier.com | www.bookaid.org | www.sabre.org
. .
Contents

Preface xi

1 Introduction 1
1.1 Overview 1
1.2 Human and computer vision 1
1.3 The human vision system 3
1.3.1 The eye 4
1.3.2 The neural system 6
1.3.3 Processing 7
1.4 Computer vision systems 9
1.4.1 Cameras 10
1.4.2 Computer interfaces 12
1.4.3 Processing an image 14
1.5 Mathematical systems 15
1.5.1 Mathematical tools 16
1.5.2 Hello Mathcad, hello images! 16
1.5.3 Hello Matlab! 21
1.6 Associated literature 24
1.6.1 Journals and magazines 24
1.6.2 Textbooks 25
1.6.3 The web 28
1.7 Conclusions 29
1.8 References 29

2 Images, sampling and frequency domain processing 33

2.1 Overview 33
2.2 Image formation 34
2.3 The Fourier transform 37
2.4 The sampling criterion 43
2.5 The discrete Fourier transform 47
2.5.1 One-dimensional transform 47
2.5.2 Two-dimensional transform 49
2.6 Other properties of the Fourier transform 54
2.6.1 Shift invariance 54
2.6.2 Rotation 56
2.6.3 Frequency scaling 56
2.6.4 Superposition (linearity) 57

v
 2.7 Transforms other than Fourier 58
2.7.1 Discrete cosine transform 58
2.7.2 Discrete Hartley transform 59
2.7.3 Introductory wavelets: the Gabor wavelet 61
2.7.4 Other transforms 63
2.8 Applications using frequency domain properties 64
2.9 Further reading 65
2.10 References 66

3 Basic image processing operations 69

3.1 Overview 69
3.2 Histograms 70
3.3 Point operators 71
3.3.1 Basic point operations 71
3.3.2 Histogram normalization 74
3.3.3 Histogram equalization 75
3.3.4 Thresholding 77
3.4 Group operations 81
3.4.1 Template convolution 81
3.4.2 Averaging operator 84
3.4.3 On different template size 87
3.4.4 Gaussian averaging operator 88
3.5 Other statistical operators 90
3.5.1 More on averaging 90
3.5.2 Median filter 91
3.5.3 Mode filter 94
3.5.4 Anisotropic diffusion 96
3.5.5 Force field transform 101
3.5.6 Comparison of statistical operators 102
3.6 Mathematical morphology 103
3.6.1 Morphological operators 104
3.6.2 Grey-level morphology 107
3.6.3 Grey-level erosion and dilation 108
3.6.4 Minkowski operators 109
3.7 Further reading 112
3.8 References 113

4 Low-level feature extraction (including edge detection) 115

4.1 Overview 115
4.2 First order edge detection operators 117
4.2.1 Basic operators 117
4.2.2 Analysis of the basic operators 119
4.2.3 Prewitt edge detection operator 121
4.2.4 Sobel edge detection operator 123
4.2.5 Canny edge detection operator 129

vi Contents
 4.3 Second order edge detection operators 137
4.3.1 Motivation 137
4.3.2 Basic operators: the Laplacian 137
4.3.3 Marr–Hildreth operator 139
4.4 Other edge detection operators 144
4.5 Comparison of edge detection operators 145
4.6 Further reading on edge detection 146
4.7 Phase congruency 147
4.8 Localized feature extraction 152
4.8.1 Detecting image curvature (corner extraction) 153
4.8.1.1 Definition of curvature 153
4.8.1.2 Computing differences in edge direction 154
4.8.1.3 Measuring curvature by changes in intensity
(differentiation) 156
4.8.1.4 Moravec and Harris detectors 159
4.8.1.5 Further reading on curvature 163
4.8.2 Modern approaches: region/patch analysis 163
4.8.2.1 Scale invariant feature transform 163
4.8.2.2 Saliency 166
4.8.2.3 Other techniques and performance issues 167
4.9 Describing image motion 167
4.9.1 Area-based approach 168
4.9.2 Differential approach 171
4.9.3 Further reading on optical flow 177
4.10 Conclusions 178
4.11 References 178

5 Feature extraction by shape matching 183

5.1 Overview 183
5.2 Thresholding and subtraction 184
5.3 Template matching 186
5.3.1 Definition 186
5.3.2 Fourier transform implementation 193
5.3.3 Discussion of template matching 196
5.4 Hough transform 196
5.4.1 Overview 196
5.4.2 Lines 197
5.4.3 Hough transform for circles 203
5.4.4 Hough transform for ellipses 207
5.4.5 Parameter space decomposition 210
5.4.5.1 Parameter space reduction for lines 210
5.4.5.2 Parameter space reduction for circles 212
5.4.5.3 Parameter space reduction for ellipses 217
5.5 Generalized Hough transform 221
5.5.1 Formal definition of the GHT 221
5.5.2 Polar definition 223

Contents vii
 5.5.3 The GHT technique 224
5.5.4 Invariant GHT 228
5.6 Other extensions to the Hough transform 235
5.7 Further reading 236
5.8 References 237

6 Flexible shape extraction (snakes and other techniques) 241

6.1 Overview 241
6.2 Deformable templates 242
6.3 Active contours (snakes) 244
6.3.1 Basics 244
6.3.2 The greedy algorithm for snakes 246
6.3.3 Complete (Kass) snake implementation 252
6.3.4 Other snake approaches 257
6.3.5 Further snake developments 257
6.3.6 Geometric active contours 261
6.4 Shape skeletonization 266
6.4.1 Distance transforms 266
6.4.2 Symmetry 268
6.5 Flexible shape models: active shape and active
appearance 272
6.6 Further reading 275
6.7 References 276

7 Object description 281

7.1 Overview 281
7.2 Boundary descriptions 282
7.2.1 Boundary and region 282
7.2.2 Chain codes 283
7.2.3 Fourier descriptors 285
7.2.3.1 Basis of Fourier descriptors 286
7.2.3.2 Fourier expansion 287
7.2.3.3 Shift invariance 289
7.2.3.4 Discrete computation 290
7.2.3.5 Cumulative angular function 292
7.2.3.6 Elliptic Fourier descriptors 301
7.2.3.7 Invariance 305
7.3 Region descriptors 311
7.3.1 Basic region descriptors 311
7.3.2 Moments 315
7.3.2.1 Basic properties 315
7.3.2.2 Invariant moments 318
7.3.2.3 Zernike moments 320
7.3.2.4 Other moments 324
7.4 Further reading 325
7.5 References 326

viii Contents
8 Introduction to texture description, segmentation and classification 329
8.1 Overview 329
8.2 What is texture? 330
8.3 Texture description 332
8.3.1 Performance requirements 332
8.3.2 Structural approaches 332
8.3.3 Statistical approaches 335
8.3.4 Combination approaches 337
8.4 Classification 339
8.4.1 The k-nearest neighbour rule 339
8.4.2 Other classification approaches 343
8.5 Segmentation 343
8.6 Further reading 345
8.7 References 346

9 Appendix 1: Example worksheets 349

9.1 Example Mathcad worksheet for Chapter 3 349
9.2 Example Matlab worksheet for Chapter 4 352

10 Appendix 2: Camera geometry fundamentals 355

10.1 Image geometry 355
10.2 Perspective camera 355
10.3 Perspective camera model 357
10.3.1 Homogeneous coordinates and projective geometry 357
10.3.1.1 Representation of a line and duality 358
10.3.1.2 Ideal points 358
10.3.1.3 Transformations in the projective space 359
10.3.2 Perspective camera model analysis 360
10.3.3 Parameters of the perspective camera model 363
10.4 Affine camera 364
10.4.1 Affine camera model 365
10.4.2 Affine camera model and the perspective projection 366
10.4.3 Parameters of the affine camera model 368
10.5 Weak perspective model 369
10.6 Example of camera models 371
10.7 Discussion 379
10.8 References 380

11 Appendix 3: Least squares analysis 381

11.1 The least squares criterion 381
11.2 Curve fitting by least squares 382

Contents ix
12 Appendix 4: Principal components analysis 385
12.1 Introduction 385
12.2 Data 385
12.3 Covariance 386
12.4 Covariance matrix 388
12.5 Data transformation 389
12.6 Inverse transformation 390
12.7 Eigenproblem 391
12.8 Solving the eigenproblem 392
12.9 PCA method summary 392
12.10 Example 393
12.11 References 398

Index 399

x Contents
. .
Preface

Why did we write this book?

We will no doubt be asked many times: why on earth write a new book on computer vision?
Fair question: there are already many good books on computer vision in the bookshops, as you
will find referenced later, so why add to them? Part of the answer is that any textbook is a
snapshot of material that exists before it. Computer vision, the art of processing images stored
within a computer, has seen a considerable amount of research by highly qualified people and
the volume of research would appear even to have increased in recent years. This means that a
lot of new techniques have been developed, and many of the more recent approaches have yet
to migrate to textbooks.
But it is not just the new research: part of the speedy advance in computer vision technique
has left some areas covered only in scanty detail. By the nature of research, one cannot publish
material on technique that is seen more to fill historical gaps, rather than to advance knowledge.
This is again where a new text can contribute.
Finally, the technology itself continues to advance. This means that there is new hardware,
and there are new programming languages and new programming environments. In particular for
computer vision, the advance of technology means that computing power and memory are now
relatively cheap. It is certainly considerably cheaper than when computer vision was starting as
a research field. One of the authors here notes that the laptop that his portion of the book was
written on has more memory, is faster, and has bigger disk space and better graphics than the
computer that served the entire university of his student days. And he is not that old! One of
the more advantageous recent changes brought about by progress has been the development of
mathematical programming systems. These allow us to concentrate on mathematical technique
itself, rather than on implementation detail. There are several sophisticated flavours, of which
Mathcad and Matlab, the chosen vehicles here, are among the most popular. We have been using
these techniques in research and teaching, and we would argue that they have been of consider-
able benefit there. In research, they help us to develop technique more quickly and to evaluate
its final implementation. For teaching, the power of a modern laptop and a mathematical system
combines to show students, in lectures and in study, not only how techniques are implemented,
but also how and why they work with an explicit relation to conventional teaching material.
We wrote this book for these reasons. There is a host of material that we could have included
but chose to omit. Our apologies to other academics if it was your own, or your favourite,
technique. By virtue of the enormous breadth of the subject of computer vision, we restricted the
focus to feature extraction and image processing in computer vision, for this not only has been
the focus of our research, but is also where the attention of established textbooks, with some
exceptions, can be rather scanty. It is, however, one of the prime targets of applied computer
vision, so would benefit from better attention. We have aimed to clarify some of its origins
and development, while also exposing implementation using mathematical systems. As such,
we have written this text with our original aims in mind.

xi
Why did we produce another edition?

There are many reasons why we have updated the book to provide this new edition. First,
despite its electronic submission, some of the first edition was retyped before production. This
introduced errors that we have now corrected. Next, the field continues to move forward: we
now include some techniques which were gaining appreciation when we first wrote the book,
or have been developed since. Some areas move more rapidly than others, and this is reflected
in the changes made. Also, there has been interim scholarship, especially in the form of new
texts, and we include these new ones as much as we can. Matlab and Mathcad are still the
computational media here, and there is a new demonstration site which uses Java. Finally, we
have maintained the original format. It is always tempting to change the format, in this case even
to reformat the text, but we have chosen not to do so. Apart from corrections and clarifications,
the main changes from the previous edition are:

• Chapter 1: updating of eye operation, camera technology and software, updating and exten-
sion of web material and literature
• Chapter 2: very little (this is standard material), except for an excellent example of aliasing
• Chapter 3: inclusion of anisotropic diffusion for image smoothing, the force field operator
and mathematical morphology
• Chapter 4: extension of frequency domain concepts and differentiation operators; inclusion
of phase congruency, modern curvature operators and the scale invariant feature transform
(SIFT)
• Chapter 5: emphasis of the practical attributes of feature extraction in occlusion and noise,
and some moving-feature techniques
• Chapter 6: inclusion of geometric active contours and level set methods, inclusion of skele-
tonization, extension of active shape models
• Chapter 7: extension of the material on moments, particularly Zernike moments, including
reconstruction from moments
• Chapter 8: clarification of some of the detail in feature-based recognition
• Appendices: these have been extended to cover camera models in greater detail, and principal
components analysis.

As already mentioned, there is a new JAVA-based demonstration site, at http://www.ecs.soton.

ac.uk/∼msn/book/new_demo/, which has some of the techniques described herein and some
examples of how computer vision-based biometrics work. This webpage will continue to be
updated.

The book and its support

Each chapter of the book presents a particular package of information concerning feature
extraction in image processing and computer vision. Each package is developed from its origins
and later referenced to more recent material. Naturally, there is often theoretical development
before implementation (in Mathcad or Matlab). We have provided working implementations of
most of the major techniques we describe, and applied them to process a selection of imagery.
Although the focus of our work has been more in analysing medical imagery or in biometrics

xii Preface
(the science of recognizing people by behavioural or physiological characteristics, like face
recognition), the techniques are general and can migrate to other application domains.
You will find a host of further supporting information at the book’s website
(http://www.ecs.soton.ac.uk/∼msn/book/). First, you will find the worksheets (the Matlab and
Mathcad implementations that support the text) so that you can study the techniques described
herein. There are also lecturing versions that have been arranged for display via an overhead
projector, with enlarged text and more interactive demonstration. The example questions (and,
eventually, their answers) are also there. The demonstration site is there too. The website will
be kept as up to date as possible, for it also contains links to other material such as websites
devoted to techniques and to applications, as well as to available software and online literature.
Finally, any errata will be reported there. It is our regret and our responsibility that these will
exist, but our inducement for their reporting concerns a pint of beer. If you find an error that
we do not know about (not typos such as spelling, grammar and layout) then use the mailto on
the website and we shall send you a pint of good English beer, free!
There is a certain amount of mathematics in this book. The target audience is third or fourth
year students in BSc/BEng/MEng courses in electrical or electronic engineering, software engi-
neering and computer science, or in mathematics or physics, and this is the level of mathematical
analysis here. Computer vision can be thought of as a branch of applied mathematics, although
this does not really apply to some areas within its remit, but certainly applies to the material
herein. The mathematics essentially concerns mainly calculus and geometry, although some of
it is rather more detailed than the constraints of a conventional lecture course might allow. Cer-
tainly, not all of the material here is covered in detail in undergraduate courses at Southampton.
The book starts with an overview of computer vision hardware, software and established
material, with reference to the most sophisticated vision system yet ‘developed’: the human
vision system. Although the precise details of the nature of processing that allows us to see
have yet to be determined, there is a considerable range of hardware and software that allow
us to give a computer system the capability to acquire, process and reason with imagery, the
function of ‘sight’. The first chapter also provides a comprehensive bibliography of material
on the subject, including not only textbooks, but also available software and other material. As
this will no doubt be subject to change, it might well be worth consulting the website for more
up-to-date information. The preferred journal references are those that are likely to be found
in local university libraries or on the web, IEEE Transactions in particular. These are often
subscribed to as they are relatively low cost, and are often of very high quality.
The next chapter concerns the basics of signal processing theory for use in computer vision.
It introduces the Fourier transform, which allows you to look at a signal in a new way, in terms
of its frequency content. It also allows us to work out the minimum size of a picture to conserve
information and to analyse the content in terms of frequency, and even helps to speed up some
of the later vision algorithms. Unfortunately, it does involve a few equations, but it is a new
way of looking at data and signals, and proves to be a rewarding topic of study in its own right.
We then start to look at basic image-processing techniques, where image points are mapped
into a new value first by considering a single point in an original image, and then by considering
groups of points. We see not only common operations to make a picture’s appearance better,
especially for human vision, but also how to reduce the effects of different types of commonly
encountered image noise. This is where the techniques are implemented as algorithms in Mathcad
and Matlab to show precisely how the equations work. We shall see some of the modern ways
to remove noise and thus clean images, and we shall also look at techniques which process an
image using notions of shape, rather than mapping processes.

Preface xiii
 The following chapter concerns low-level features, which are the techniques that describe
the content of an image, at the level of a whole image rather than in distinct regions of it. One
of the most important processes is edge detection. Essentially, this reduces an image to a form
of a caricaturist’s sketch, but without a caricaturist’s exaggerations. The major techniques are
presented in detail, together with descriptions of their implementation. Other image properties
we can derive include measures of curvature and measures of movement. These also are covered
in this chapter.
These edges, the curvature or the motion need to be grouped in some way so that we can
find shapes in an image. Our first approach to shape extraction concerns analysing the match
of low-level information to a known template of a target shape. As this can be computationally
very cumbersome, we then progress to a technique that improves computational performance,
while maintaining an optimal performance. The technique is known as the Hough transform,
and it has long been a popular target for researchers in computer vision who have sought to
clarify its basis, improve its speed, and increase its accuracy and robustness. Essentially, by
the Hough transform we estimate the parameters that govern a shape’s appearance, where the
shapes range from lines to ellipses and even to unknown shapes.
Some applications of shape extraction require the determination of rather more than the
parameters that control appearance, but require the ability to deform or flex to match the image
template. For this reason, the chapter on shape extraction by matching is followed by one on
flexible shape analysis. This is a topic that has shown considerable progress of late, especially
with the introduction of snakes (active contours). The newer material is the formulation by level
set methods, and brings new power to shape-extraction techniques. These seek to match a shape
to an image by analysing local properties. Further, we shall see how we can describe a shape by
its skeleton, although with practical difficulty which can be alleviated by symmetry (though this
can be slow), and also how global constraints concerning the statistics of a shape’s appearance
can be used to guide final extraction.
Up to this point, we have not considered techniques that can be used to describe the shape
found in an image. We shall find that the two major approaches concern techniques that describe
a shape’s perimeter and those that describe its area. Some of the perimeter description techniques,
the Fourier descriptors, are even couched using Fourier transform theory, which allows analysis
of their frequency content. One of the major approaches to area description, statistical moments,
also has a form of access to frequency components, but is of a very different nature to the Fourier
analysis. One advantage is that insight into descriptive ability can be achieved by reconstruction,
which should get back to the original shape.
The final chapter describes texture analysis, before some introductory material on pattern
classification. Texture describes patterns with no known analytical description and has been the
target of considerable research in computer vision and image processing. It is used here more
as a vehicle for material that precedes it, such as the Fourier transform and area descriptions,
although references are provided for access to other generic material. There is also introductory
material on how to classify these patterns against known data, but again this is a window on a
much larger area, to which appropriate pointers are given.
The appendices include a printout of abbreviated versions of the Mathcad and Matlab work-
sheets. The other appendices include material that is germane to the text, such as camera
models and coordinate geometry, the method of least squares and a topic known as principal
components analysis. These are aimed to be short introductions, and are appendices since they
are germane to much of the material. Other related, especially online, material is referenced
throughout the text.

xiv Preface
 In this way, the text covers all major areas of feature extraction in image processing and
computer vision. There is considerably more material on the subject than is presented here;
for example, there is an enormous volume of material on 3D computer vision and 2D signal
processing which is only alluded to here. Topics that are specifically not included are colour,
3D processing and image coding. But to include all that would lead to a monstrous book that
no one could afford, or even pick up! So we admit that we give a snapshot, but we hope that it
is considered to open another window on a fascinating and rewarding subject.

In gratitude

We are immensely grateful to the input of our colleagues, in particular to Prof. Steve Gunn and to
Dr John Carter. The family who put up with it are Maria Eugenia and Caz and the nippers. We are
also very grateful to past and present researchers in computer vision at the Information, Signals,
Images, Systems (ISIS) Research Group under (or who have survived?) Mark’s supervision at
the School of Electronics and Computer Science, University of Southampton. As well as Alberto
and Steve, these include Dr Hani Muammar, Prof. Xiaoguang Jia, Prof. Yan Qiu Chen, Dr
Adrian Evans, Dr Colin Davies, Dr David Cunado, Dr Jason Nash, Dr Ping Huang, Dr Liang Ng,
Dr Hugh Lewis, Dr David Benn, Dr Douglas Bradshaw, Dr David Hurley, Dr John Manslow, Dr
Mike Grant, Bob Roddis, Dr Andrew Tatem, Dr Karl Sharman, Dr Jamie Shutler, Dr Jun Chen,
Dr Andy Tatem, Dr Chew Yam, Dr James Hayfron-Acquah, Dr Yalin Zheng, Dr Jeff Foster, Dr
Peter Myerscough, Dr David Wagg, Dr Ahmad Al-Mazeed, Dr Jang-Hee Yoo, Dr Nick Spencer,
Stuart Mowbray, Dr Stuart Prismall, Dr Peter Gething, Dr Mike Jewell, Dr David Wagg, Dr
Alex Bazin, Hidayah Rahmalan, Xin Liu, Imed Bouchrika, Banafshe Arbab-Zavar, Dan Thorpe,
Cem Direkoglu (the latter two especially for the new active contour material), John Evans (for
the great hippo photo) and to Jamie Hutton, Ben Dowling and Sina Samangooei (for the Java
demonstrations site). We are also very grateful to other past Southampton students of BEng
and MEng Electronic Engineering, MEng Information Engineering, BEng and MEng Computer
Engineering, MEng Software Engineering and BSc Computer Science who have pointed our
earlier mistakes, noted areas for clarification and in some cases volunteered some of the material
herein. Beyond Southampton, we remain grateful to the reviewers of the two editions and to Prof.
Daniel Cremers, Dr Timor Kadir and Prof. Tim Cootes for observations on and improvements
to the text and for permission to use images. To all of you, our very grateful thanks.

Final message

We ourselves have already benefited much by writing this book, and this second edition. As
we already know, previous students have also benefited, and contributed to it as well. But it
remains our hope that it will inspire people to join in this fascinating and rewarding subject that
has proved to be such a source of pleasure and inspiration to its many workers.

Mark S. Nixon Alberto S. Aguado

University of Southampton University of Surrey

Preface xv
This page intentionally left blank
.
1 .

Introduction

1.1 Overview

This is where we start, by looking at the human visual system to investigate what is meant
by vision, then on to how a computer can be made to sense pictorial data and then how we
can process an image. The overview of this chapter is shown in Table 1.1; you will find a
similar overview at the start of each chapter. There are no references (citations) in the overview,
citations are made in the text and are collected at the end of each chapter.

Table 1.1 Overview of Chapter 1

Main topic Sub topics Main points

Human vision How the eye works, how visual Sight, lens, retina, image, colour,
system information is processed and how it can monochrome, processing, brain, visual
fail. illusions.
Computer How electronic images are formed, how Picture elements, pixels, video standard,
vision systems video is fed into a computer and how we camera technologies, pixel technology,
can process the information using a performance effects, specialist cameras,
computer. video conversion, computer languages,
processing packages. Demonstrations of
working techniques.
Mathematical How we can process images using Ease, consistency, support, visualization
systems mathematical packages; introduction to of results, availability, introductory use,
the Matlab and Mathcad systems. example worksheets.
Literature Other textbooks and other places to find Magazines, textbooks, websites and this
information on image processing, book’s website.
computer vision and feature extraction.

1.2 Human and computer vision

A computer vision system processes images acquired from an electronic camera, which is like the
human vision system where the brain processes images derived from the eyes. Computer vision
is a rich and rewarding topic for study and research for electronic engineers, computer scientists
and many others. Increasingly, it has a commercial future. There are now many vision systems
in routine industrial use: cameras inspect mechanical parts to check size, food is inspected

1
for quality, and images used in astronomy benefit from computer vision techniques. Forensic
studies and biometrics (ways to recognize people) using computer vision include automatic face
recognition and recognizing people by the ‘texture’ of their irises. These studies are paralleled
by biologists and psychologists who continue to study how our human vision system works,
and how we see and recognize objects (and people).
A selection of (computer) images is given in Figure 1.1; these images comprise a set of points
or picture elements (usually concatenated to pixels) stored as an array of numbers in a computer.
To recognize faces, based on an image such as Figure 1.1(a), we need to be able to analyse
constituent shapes, such as the shape of the nose, the eyes and the eyebrows, to make some
measurements to describe, and then recognize, a face. (Figure 1.1a is perhaps one of the most
famous images in image processing. It is called the Lena image, and is derived from a picture
of Lena Sjööblom in Playboy in 1972.) Figure 1.1(b) is an ultrasound image of the carotid
artery (which is near the side of the neck and supplies blood to the brain and the face), taken
as a cross-section through it. The top region of the image is near the skin; the bottom is inside
the neck. The image arises from combinations of the reflections of the ultrasound radiation by
tissue. This image comes from a study that aimed to produce three-dimensional (3D) models of
arteries, to aid vascular surgery. Note that the image is very noisy, and this obscures the shape
of the (elliptical) artery. Remotely sensed images are often analysed by their texture content.
The perceived texture is different between the road junction and the different types of foliage
seen in Figure 1.1(c). Finally, Figure 1.1(d) is a magnetic resonance imaging (MRI) image of a
cross-section near the middle of a human body. The chest is at the top of the image, the lungs
and blood vessels are the dark areas, and the internal organs and the fat appear grey. Nowadays,
MRI images are in routine medical use, owing to their ability to provide high-quality images.

(a) Face from a camera (b) Artery from (c) Ground by remote (d) Body by magnetic
ultrasound sensing resonance

Figure 1.1 Real images from different sources

There are many different image sources. In medical studies, MRI is good for imaging soft
tissue, but does not reveal the bone structure (the spine cannot be seen in Figure 1.1d); this
can be achieved by using computed tomography (CT), which is better at imaging bone, as
opposed to soft tissue. Remotely sensed images can be derived from infrared (thermal) sensors
or synthetic-aperture radar, rather than by cameras, as in Figure 1.1(c). Spatial information can
be provided by two-dimensional arrays of sensors, including sonar arrays. There are perhaps
more varieties of sources of spatial data in medical studies than in any other area. But computer
vision techniques are used to analyse any form of data, not just the images from cameras.
Synthesized images are good for evaluating techniques and finding out how they work, and
some of the bounds on performance. Two synthetic images are shown in Figure 1.2. Figure 1.2(a)

2 Feature Extraction and Image Processing

is an image of circles that were specified mathematically. The image is an ideal case: the circles
are perfectly defined and the brightness levels have been specified to be constant. This type of
synthetic image is good for evaluating techniques which find the borders of the shape (its edges)
and the shape itself, and even for making a description of the shape. Figure 1.2(b) is a synthetic
image made up of sections of real image data. The borders between the regions of image data are
exact, again specified by a program. The image data comes from a well-known texture database,
the Brodatz album of textures. This was scanned and stored as a computer image. This image can
be used to analyse how well computer vision algorithms can identify regions of differing texture.

(a) Circles (b) Textures

Figure 1.2 Examples of synthesized images

This chapter will show you how basic computer vision systems work, in the context of the
human vision system. It covers the main elements of human vision, showing you how your eyes
work (and how they can be deceived). For computer vision, this chapter covers the hardware
and the software used for image analysis, giving an introduction to Mathcad and Matlab, the
software tools used throughout this text to implement computer vision algorithms. Finally, a
selection of pointers to other material is provided, especially those for more detail on the topics
covered in this chapter.

1.3 The human vision system

Human vision is a sophisticated system that senses and acts on visual stimuli. It has evolved
for millions of years, primarily for defence or survival. Intuitively, computer and human vision
appear to have the same function. The purpose of both systems is to interpret spatial data,
data that is indexed by more than one dimension. Even though computer and human vision
are functionally similar, you cannot expect a computer vision system to replicate exactly the
function of the human eye. This is partly because we do not understand fully how the vision
system of the eye and brain works, as we shall see in this section. Accordingly, we cannot
design a system to replicate its function exactly. In fact, some of the properties of the human eye
are useful when developing computer vision techniques, whereas others are actually undesirable
in a computer vision system. But we shall see computer vision techniques which can, to some
extent, replicate, and in some cases even improve upon, the human vision system.

Introduction 3
Discovering Diverse Content Through
Random Scribd Documents
R. JAMES. F. WOOD.

E. P. BURNHAM.
 Mr. Ducker is essentially an originator. Whatever tends to make a successful race meeting when
traced back, nine times out of ten, will be found to have its impetus from him. The arranging of
programs, track building, timing, scoring, novelty races, all bear his stamp. Everybody concedes that the
Springfield tournaments were models; everything was managed with clockwork precision, and rarely
was there a hitch in the program. So great was their reputation that Mr. Ducker has often been called
upon to furnish details and even personal assistance for other meetings, and he has received letters
asking advice from Switzerland, Germany, and even Australia. His motto has always been: “The best is
none too good,” and as a result of strict adherence to that rule, the Springfield track holds to-day a
large proportion of the existing records.
His ideas on track building were the result of personal observation and study. Good side-paths in the
country were the means of awakening and guiding his attention. It occurred to him that if a path could
be built of nearly the same materials, the problem of good tracks would be solved. That he successfully
followed up this idea as well as the accuracy of his reasoning, the Springfield track, and, more recently,
the Buffalo track indisputably prove.
In 1885 and 1886, Mr. Ducker was chief consul of the Massachusetts division, L. A. W., and his work
in that office speaks for itself. He was also for two years a member of the racing board of the L. A. W.,
and representative for Massachusetts. He was for five years president of the Springfield Bicycle Club, of
which he is a life member; he is a member of the Massachusetts Bicycle Club of Boston, the Ixion Club
of New York City, the Ramblers of Buffalo and the N. C. U. of England. In connection with the
Springfield tournaments, Mr. Ducker founded the Springfield Wheelmen’s Gazette. It was intended at
first only as a tournament “boomer,” but it made such a hit, that he yielded to the public demand for its
permanent publication. Upon his removal to Buffalo, the Gazette was sold to Darrow Brothers, of
Indianapolis. While in Mr. Ducker’s control it was a crisp, sparkling sheet, and commendable from a
literary standpoint. He was also the publisher and editor, in connection with Henry Goodman, of “The
Wheelmen’s Reference Book.”
Mr. Ducker’s cycling correspondence is simply enormous. His private office is the headquarters for
cycling information of every kind, and in Springfield it was constantly besieged by newspaper men.
Until within a few months, Mr. Ducker has worked regularly at his business, consequently his cycling
work has been done after business hours. He is of medium height and inclined to stoutness. He is of
light complexion, with sandy, curly hair and heavy imperial and mustache. Nature has not endowed Mr.
Ducker with a very good voice, having oversupplied him with tones of the upper, entirely to the neglect
of those of the lower register. But his voice is no handicap to his ability to talk. He is an enthusiastic
conversationalist, and can convert the most skeptical to his optimistic way of thinking.
For the past few months, Mr. Ducker has given his entire attention to the World’s Tournament at
Buffalo, which is his latest project. The management of the Buffalo International Fair Association,
recognizing Mr. Ducker’s abilities, secured his services by most liberal offers of support. And Mr. Ducker’s
first official act was to appoint his friend G. M. Hendee as starter.
A full report of events as they shall become a matter of record in connection with the Buffalo meet,
will appear in later issues of OUTING.
It now remains for us to recall a few of the names of the noted cyclers who, under the management
of Mr. Ducker, visited Springfield during his prominent connection with the cycling history of that most
noted of American cycling clubs.
In the year 1886, W. A. Rowe defeated George M. Hendee and Fred Wood, of England, for the
world’s championship. Rowe is, of course, very well known to the cycling world by his wonderful record,
holding as he does all from a 1⁄4 mile to 22 miles. These have been, however, made at record trials,
i. e., against time and not in races. Recently Rowe visited England, but he has twice been unsuccessful
in holding the title of the world’s champion as against Richard Howell.
M. V. J. Webber, or “Alphabet” Webber, was one of the fast English amateurs who raced at
Springfield in ’85. He made 21 miles within the hour during a race. It was a 10-mile race, but he was
anxious to keep on, and was allowed to do so with the result above mentioned. He has been off the
path since his return to England.

G. M. HENDEE.

George Weber was America’s champion Star rider, but he died in ’85. He was a plucky rider, and
though he did not secure many first places in track riding, he was unconquerable in road racing and hill
climbing. He won the great 100-mile road race in the spring of ’85.
 HAMPDEN PARK, IN SPRINGFIELD, MASS.

Richard Howell, of England, professional, is undoubtedly the world’s champion. Indeed, he has for a
long time been called “King of the Wheel.” His recent defeats of Rowe have put his right to the title
beyond dispute. He has rarely been beaten and is a marvelous rider, having a spurt that cannot be
approached. He was the first to do a mile in 2m. 31 1-5s. It was a trial against time and was made just
after the ’85 tournament at Springfield.
Percy Furnivall, while on the path, was England’s fastest amateur rider, holding the amateur
championship of England for two years. He raced at the ’85 Springfield tournament and won every
event in which he started. He was to have raced against Hendee, at that time America’s champion
amateur, but Hendee was “spilled” and prevented from racing.
R. A. Cripps was another English amateur who raced at Springfield in ’85. He was first-class as a
tricycle rider.
Another English professional of note who has appeared on the Springfield track is Fred Wood. He
was formerly Howell’s great rival. In ’86, Wood was the only scratch man in a mile handicap at Hartford,
and won, his time being 2m. 33s., the fastest mile ever made in a race in America. The race was run on
a trotting track, and if it had been the Springfield track the time would have been nearer 2m. 31s. Wood
made 2m. 35s. at Springfield the following week.
E. P. Burnham is what is known as a “luck” rider, for in several races he has been first through
accidents to others. He is, however, a good rider, and very hard to beat on a tricycle. He has been off
the track for two years. H. G. Crocker is a protégé of Burnham, and is one of America’s best riders.
William M. Woodside is known as the Irish champion, and is a member of W. J. Morgan’s American
Racing Team. Woodside has sometimes been styled the champion of America, but has never really held
the title. He is best known by his having done so much “donkey work” in races, i. e., he has set the
pace for others and thus sacrificed his own chances for a position. He is a professional rider.
John Shillington Prince is also a professional. He was the first to put the mile record down to 2m.
39s., which performance was shortly afterwards equaled by Sanders Sellers, the fast English amateur,
who defeated Hendee in 1884. Prince has also posed as America’s champion rider. He formerly gained
much prominence when he was racing against John Keen, England’s old war-horse.
Of course, numerous other prominent riders have taken part in the Springfield tournaments. Lewis
B. Hamilton was a very popular amateur, and was known as the Yale College rider. Robert James,
professional, and Reuben Chambers, amateur, are Englishmen who have appeared several times. In ’85,
R. H. English performed as an amateur, but is now a professional, while at the same time W. A. and G.
H. Illston, both amateurs, were in America for the Springfield tournament. Space fails us to mention all
the prominent riders whose names have been on the programs of the Springfield tournaments, but the
few we have mentioned will convince the unprejudiced reader of the omnipotence in the bicycling world
of Henry E. Ducker.

[1] An article on this club appeared in OUTING, Vol. II., page 337. Another is now in preparation.--ED.
 WILD DUCK SHOOTING.
BY W. G. BEERS.

MONG the memorable events of my youth I can scarcely recall any rival to the
days spent on foot and in canoe hunting wild duck. It was the master passion
of the boyhood of many I know, becoming in later years a passion to master. It
was the acme of enjoyment in the days when one was light-hearted and
débonnaire, and went whistling through birthdays with that enviable serenity
so few of us manage to retain.
Wild duck! With the last fall of leaves and the first fall of snow, their quack
was music to the ear. Steeped to the lips in classics, one wondered if there
were no duck on the coast of Campania, that Tiberius tired of the pleasures around him and sighed in
vain for more; or if there were none in Assyria, that Sardanapalus sought to have new amusements
invented; or if there were no real ones where Loelius and Scipio made them on water with flat stones.
The first wild duck one kills, like first love, or one’s first proof-sheet, causes a sensation that is never
duplicated. The history of its mysterious and ecstatic thrill through the veins, its wild rush through the
soul, never knows a repetition. The duck may be in the “sere and yellow,” stricken in years, scraggy on
the crown, weak in the wings, tough to your teeth as parchment—aye, indeed, with one foot in the
grave and the other shot off, and have long ago ceased to scud between earth and sky for mere fun—
just as the first love may have been nearly old enough to have been your mother, and with no more
love in her eyes than an oyster; or as the first proof-sheet may have been an immature production to
which you are now thankful you did not append your name. But in the heyday of life a vivid imagination
throws a halo around our achievements, and though other duck, like other love, may turn out more
“tender and true,” yet there lingers about the memory of the first experience an inexpressible charm
which no gross soul can know.
I do not think I shall ever forget the first wild duck I shot. It was impressed upon me in a manner
too striking. During the school holidays a few of us undertook to dispose of our superfluous energy by a
pedestrian pilgrimage around the Island of Montreal, and as a dose for the game we might encounter,
we managed, by coaxing a big brother, to muster a single-barreled gun and liberal supply of
ammunition. There was a strong suspicion of rust down the barrel, and a disabled look about the
hammer, but the owner declared it was good enough for boys, with that sublime faith manifested by
watermen who let boats to inexperienced lads, that Providence takes special care of people who cannot
take care of themselves. A well-worn inscription on the butt was ominously deciphered as “Memento
mori.” I’ve seen more defective guns since—but they had burst.
 MALLARD DUCK (ANAS BOSCHAS).

We started from the Place d’Armes, and when we reached “the Cross,” at Hochelega, held a council
of war about loading the gun, as a scared squirrel had just darted under a fence and roused our thirst
for blood. Opinions conflicted as to whether the powder or shot should be put in first, as one dogmatic
adventurer, whose experience in squibs and fire-crackers entitled him to respect, declared with the
positiveness of error that the shot should have the preference. Better reasoning, however, prevailed,
and to make assurance doubly sure, down went a double charge of powder. “It’s not near full yet,”
sneered young Dogmatism. I hoped not; but to make assurance trebly sure, up came the flask again
and down went more powder. I remember one of the group, whose characteristic caution provoked us
throughout the trip, suggested mounting the gun in an embrasure in the fence, laying a train of powder
to the nipple, and testing its safety at discreet distance; but there was a display of fear in the proposal
that we, as of Saxon blood, could never countenance, and so we strangled it at birth. It is a memorable
fact, that may go some way to sustain the belief that I have mentioned above, that, as if prompted by
instinct, the gun refused to go off on several occasions, in spite of repeated cleanings of the nipple,
coaxing with grains of powder and fresh caps. We were unable to “distill the soul of goodness” in this
apparently evil and obdurate circumstance; so the charge was withdrawn, the barrel cleaned, and to
make assurance quadruply sure, the powder was poured down with even more liberality than before.
The third day we reached the upper end of Ste. Anne’s, near the old French fort. At that time the
village was even a quieter spot than now, where never a speculator had looked with greed upon the
soil; its greatest stir made by the visits and voices of the boisterous voyageurs; its rapids sacred to the
memory of the poet Moore, and the soft refrain of his “Canadian Boat Song.” Moreover, its surroundings
made it a perfect paradise for wild duck.
We were marching along, when some one’s sharp eyes espied a solitary black duck feeding close to
the shore, about thirty yards away. Suddenly it rose with a frightened flutter. With considerable difficulty
I had managed to cock my gun. I raised it to my shoulder, with a strong fear that it would go off, and
an inward prayer that it wouldn’t, took accurate aim by pointing in the direction of the bird, and shutting
my eyes—with the Latin inscription brought at that moment vividly before me, as if the letters had
elongated from the butt to the barrel—I thought of my past sins and pulled the trigger.

EIDER DUCK (SOMATERIA MOLLISSIMA).

Once I participated in a railroad accident when a locomotive almost telescoped our car; but it was an
insignificant impression to the condensed crash and astonishing concussion that followed the snapping
of the cap. As if weary of well-doing, the old gun went off with a vengeance, blowing the stock off the
barrel with a retrograde movement that met my shoulder on the way with a deliberate intention to
dislocate, sent the hammer into the air, singed the hair from around my eyes closer and more speedily
than I have ever been professionally shaved on my chin, and gave the trusting hand that was
supporting the barrel a shake of extreme familiarity—a left-handed compliment—that was reflected up
my arm and down the spinal column until it bred my deepest and most heartfelt contempt. Like Richard,
when about to fight for his kingdom, I was depressed, and
 “Had not that alacrity of spirit
And cheer of mind that I was wont to have.”

After having carried that gun round the island for three days, sparing no pains to keep it dry, to oil
its rusty barrel and wash its musty stock, I felt it had been an ungrateful companion, undeserving of the
personality with which we had almost invested it, and, to use a modern metaphor, that it “had gone
back on me.” It evoked on my part an et tu, Brute! sort of feeling. As I looked at it in silent woe, lock,
stock and barrel lying in bits, I felt sore enough at its conduct to have given it a retributive kick, and
sent it into the river, but the kicking capacity of my legs had been too materially weakened by the last
kick of the gun.
Gun gone to glory, vision of some one’s big brother with possible heavy fist and inevitable “good,
round, mouth-filling oath,” hand, head, and, indeed, all my anatomy aching, there was a consolation
that poured metaphorical oil on my wounds and alleviated the pangs of pain—I had shot the duck!
You won’t find wild duck at Ste. Anne’s to-day, except some stray ones of over-curious trait, who
refuse to be advised by their experienced friends. You’ll be lucky if you hit upon a spot within thirty
miles of Montreal where you do not find “pothunters” by the dozen—that New World species of the
genus homo who should have lived in Arcadia, where they would certainly have utilized their propensity
to good purpose by driving away the birds which haunted Lake Stymphalus, without the brazen clappers
of Vulcan or the arrows of Hercules.
For short holidays, one of the most popular localities, and therefore one which has been well spoiled,
was in the vicinity of Carillon Bay. You may enjoy a varied autumn vacation by taking the steamer Prince
of Wales at Lachine, landing at Carillon, and staging about twenty minutes to the beautifully situated
village of St. Andrews. There beg, buy or borrow a dug-out canoe, small enough to be concealed in
cover, and paddle down the charming North River, with its picturesque rocks and pretty shadows, until
you cast anchor at the portage of the Presqu’ Isle. Here you will find remnants of old camp-fires, plenty
of free fuel, hay-stacks in the vicinity to make your bed, and elderberries ripe in September, luscious in
October, waiting in thick and tempting clusters to be eaten on the spot, or taken home and made into
wine. Pitch your tent at this point, and portage your canoe through the narrow strip of loose soil and
water to some convenient slip in what is called “The Bay.” You fasten a stout stick through a rope or
chain on the nose of the boat, and two getting abreast of it where the portage is heavy, or at each end
with outstretched arms where the water is deep, you have quite an enjoyable tug, while the novelty of
being up to your knees in mud and water, without getting wet if you wear “beef” moccasins, or a
delicious indifference to wet feet if you do not, gives you a sensation of “roughing it,” that not even the
pain you’ll get across your shoulders can make you impugn.
The Bay, which is two miles across, is picturesque, and, were it not getting too well known, a
glorious place for duck. From it you see St. Placide, about seven miles away, its church spires gleaming
in the sunshine; and nearer, Presqu’ Isle Point, Borwash Point, Point de Roche, Coon’s Point, Jones’
Island, and Green Island—between which and the end of the Presqu’ Isle you can see any vessels that
pass up and down the Ottawa River. Mount Rigaud—mysterious hill, with its “Lake of Stones”—rises to
the west, while the few farms and houses of the Bay settlement lie on the uplands to the north. Over
the islands the smoke of steamers miles away may be seen, and the plash of the paddle-wheels heard
like the distant “rat-tat” of kettledrums.
The most unique echo I know in Canada follows your shot in this Bay, and is one of the “lions”—a
roaring lion at that—of the place. It travels in tremulous waves of sound across the water, lurks for a
moment in the bush of the Presqu’ Isle, then shoots out abruptly on the other side and flies over the
Ottawa to strike Mount Rigaud, where it reverberates from hill and dale, now to the right, now to the
left, in a mysterious prolonged monotone, as if at hide-and-seek in the “Lake of Stones.” Then it returns
with a scared suddenness, only to fly back in broken flutterings of sound, from crag to crag, from haunt
to haunt, again to be repeated, like frightened deer, chased and cooped up on every side, with no
escape, till, after several such re-echoes, it calms to a lullaby, and dies away on the distant hills. A
marsh fringes the Presqu’ Isle, and on its borders are many good feeding spots for the duck. The grass
of the marsh is mowed with scythes and heaped in large stacks, which you can mount to spy for duck
that may be feeding among the lily stalks—though, if your experience is limited, or your vision none of
the best, you will often be puzzled to know whether the moving objects are lily stalks or duck.
For many years, a few Canadians of French descent, the inheritors of the old voyageur-sportsman
spirit of the ancien régime, who dread legitimate labor with all their hearts, but love harder work that
smacks of adventure, have camped in the vicinity of the Bay, trapping musk-rats, catching fish or
shooting duck and snipe. The veritable chief of the clan bears the martial name of “Victor,” and is a
character in his way. I first saw him with his breeches rolled above his knees, loading his gun in the
marsh. Nature evidently made him in haste, for there is an unfinished look about his face, and enough
indentations around his head to give a phrenologist the blues. His nose is mostly nostril, and fiery
enough to make the nose of Bardolph look pale, while his eyes are black as a sloe and piercing as a
falcon’s. Though he can neither read nor rhyme, he has a taste in common with Byron—he hates pork
and loves gin. When he swears—and then he best pronounces English—spiders feign death, and his dog
turns his tail between his legs and moans. He is said, like sheep, to undress only once a year. When he
changes his clothes the very pores of his skin open themselves in mute astonishment. If you can hire
him by the day as your “Man Friday,” it will add very much to your sport, for he is a walking map of the
haunts of duck, and has a perfect genius for waking them up. He will steal with his canoe through the
marsh wherever they can go, quietly as a snake in the grass, until he is within gunshot of his game. To
crown all, he is the presiding genius of bouillon; and I canonize him for this, if for nothing more.
Have you ever tasted bouillon made in camp? It is not “fricasseed nightmare,” mon ami. It is more
savory than tongue of lark or peacocks’ brains, or other rarest dish that epicures of ancient Rome ever
compounded. Yes, it even throws the wild boar of Apicius or the roast pig of Charles Lamb into the
shades of unpalatableness. You take water, fish, musk-rat or squirrel (in lieu of beef), potatoes, onions,
butter, pepper and salt, and boil them all together in a pot, in the open air, over a glowing wood fire.
Pour off the soup, and you have the nectar of the gods; the balance is a dish I would not be ashamed
to set before a hungry king. I would not give one sip of bouillon made by Victor for a bottle of the wine
in which Cleopatra dissolved her precious pearl.
But where are the wild duck?—for this seems all digression. Ah! there they come, with the flutter of
wings which starts something of the same sort in your heart, their long necks stretched out, following
their leader in Indian file, or wedged together like the Macedonian phalanx, or spreading out when they
come nearer in échelon or like skirmishers, as if knowing the risk of receiving your shot in close column.
You lie low, concealed by the long stalks of the marsh grass—the point of your canoe hidden by the
house of a musk-rat. What a quiet few moments as they come within range! You can almost hear your
heart beat. Gun at full cock, nerves steady as a rock, ducks coming straight to their fate—look out!
Forty yards off, up goes gun to shoulder in a twinkling, eye following the game, a gentle pressure of the
trigger—deftly, as if all your care and coolness had been concentrated for that instant in your right
forefinger—down drop the legs of a duck, denoting mortal wound, off goes your dog at a plunge, back
in boisterous haste and trembling, with a frothy mouthful which he drops at your feet with an almost
human sense of importance, and an expressive wag of his tail that quivers delicious delight from every
hair! If a “fellow feeling” does not make you “wondrous kind” to that dog—if you do not realize the
touch of nature that Darwin declares makes you kin—if, after his companionship, you are not sparing in
your chastisement, generous with your pats, and loath to treat him like a dog, you must be a brute,
beneath the stature of a trained retriever, and unworthy to have the meanest and most mongrel cur
whine at your grave.
Education has ennobled your dog. His senses have gained a keenness you may envy, while more
eloquence and gratitude is gestured from his tail than can be uttered by many a human tongue and
eye. I will not question the propriety of Solomon’s instructions in training a child, but I protest against
its applicability to a dog. A dog that has been bullied into obedience possesses the same sort of training
as a boy who has been whipped into morality. They both become white-livered; the dog carries his tail
between his legs, and so would the boy if he had one. You may have seen a hot-tempered drover beat
an obstinate cow in unsuccessful attempts to make it move, while another simply twisted its tail, and at
once stimulated its muscles of locomotion. If you have to chastise a dumb brute at all, you may as well
do it mercifully, and on the Italian system of penmanship—the heavy strokes upward and the light ones
down; specially so with a dog you wish to be your companion in hunting duck or partridge.
If you have done much duck-hunting you will have discovered that within rifle-range of civilization
the instinct of duck is surpassingly keener than outside the pale. In spite of the “blue unclouded
weather,” soft calm on the water, and stillness in the air, you cannot catch them asleep any more than a
weasel. If you would get within range of them at their feeding-ground you must slip slyly and softly.
They sniff gunpowder in the air, and know it from the smell of burning bush. Victor vows they know an
empty cartridge-case or gun-wad a mile away. You cannot make them believe your canoe is a musk-rat
house, however you try. You cannot put an empty calabash on your head as they do in China, and wade
among them, so as to pull them under the water and secure them by a strap. You may fool a Chinese or
a Hindoo duck in that way, but not a Canadian. They will play in the water twenty yards away when you
have not a gun; but they know the difference between the barrels of one peeping from a marsh and the
grass stalks or lilies, better than many people know the difference between a duck and a crow.

WOOD DUCK (AIX SPONSA).

There is at least one virtue displayed by enthusiastic hunters of duck—it is that of patience. You may
not get a shot for days, or even catch a glimpse of a bird, except your tame decoys, and be tempted to
waste a cartridge for change on a stump or a branch; but it is not all monotony, sitting quietly in your
camp or in your canoe, or paddling through the marsh, and, Micawber-like, waiting for “something to
turn up.” There is a physical and intellectual enjoyment, if you have the capacity to take it in—a
pleasant antithesis to the excitement of a shot. If you’re in camp it is expended in a hundred ways. If
you do nothing more than lie on your back, with your arms under your head for a pillow, and look up
through spreading branches of trees, gorgeous with autumnal tints, into “the witchery of the soft blue
sky”—if you only let your mind lie fallow, and your hard-worked body feel the luxury of a genuine rest, it
is not time misspent. Toward the close of day the duck exercise their wings and take their supper, and
you may then get some good shots. If you are in your canoe waiting for their appearance, I commend
to you the magnificent sunset for which the Bay is famed.
 Flocks of blackbirds whiz and whir over your head in wild abandon, as if conscious they were not in
danger; the melancholy “too, too, too, to-o-t” of the owl is heard in the woods, as if it were mourning
for Minerva; kingfishers flutter in one narrow compass of mid-air over their prey, as if trembling with
apprehensive joy, and shoot down suddenly like meteors to seize the unsuspecting minnow below; the
“schayich” of the “ritualistic” snipe is heard as it rises from the bog in graceful evolutions and gyrations
a danseuse might envy; the incense of autumn is borne to your nostrils; a conversazione of swallows is
going on throughout the bush near by, while a perfect tempest of twitter rages on a tree-top. Is it love,
jealousy or scandal, is it an Œcumenical Council to proclaim the infallibility of the kingfisher or the
peacock, or are they only scolding their young ones to bed?
To complete the delight of your senses, you will be sure to add to your knowledge of entomology the
penetrating fact that, though the black flies have absconded, the marsh in autumn is “the last ditch” of
the mosquito. Here it conjugates the verb “to bite,” in all its moods and tenses, until the frost-king
subdues its ardor, or the dragonfly saves the frost the trouble. It does not interest you to know that its
wings vibrate three thousand times a minute, and that with these and the rapid vibrations of the
muscles of its chest it produces its soothing sound. Its sting is certainly very complex and attractive
under the microscope—not so under your skin. You may be ever so gallant, and yet be unable to pardon
the fact that only the female mosquitoes bite. You may be reduced to believe with Gay’s fable of the
man and the flea, “that men were made for fleas (or mosquitoes) to eat.” The mosquito is far too
insinuating in its manner. It depresses one’s mind, but it elevates one’s body. When you’re sitting in your
canoe on the qui vive for a shot, its familiar evening hymn is heard in a halo of buzzing around your
head. Sting first, like a sapper with his heel on his spade in the trenches in the face of the enemy, it
digs into you with a perseverance worthy of a nobler aim. A summer’s sucking has not satiated the
thirst of the seniors, while the junior cannibals are eager to try their stings; but the weather has curbed
their power if not their desire, and you may slap them into eternity with comparative ease. If there is no
food for powder in the air, You can live in hope and wish there was, or you can meditate on your sins,
or, what is more popular and pleasant, the sins of your friends and enemies; but it somewhat disturbs
the equanimity of your thought and humiliates your dignity to find a corduroy road of mosquito bites on
the back of your neck, and suddenly to realize that the last of the Mohicans is determined to “play tag”
with the tip of your nose, or to say its vespers vigorously in the hollow warmth of your ear.
If you’ve never shot wild duck, at least you’ve eaten them. Charles Lamb may extol roast pig, but, as
Victor says, “Pigs can’t lay eggs, nor can dey fly.” I doubt if the genial essayist ever ate wild roast duck,
done to a turn, with sage dressing, plump bellies and legs trussed, hung for a day or two before being
dressed, well basted while cooking, and sent to table hot, with apple sauce. Plutarch says that Cato
kept his household in health, when the plague was rife, by dieting them on roast duck. Can anything be
finer than the mellow sniff that steals up the nostrils from a tender roasted one, that you’ve shot
yourself?
The end of the hunting season is the ducks’ Thanksgiving Day. What tales they must hiss and stories
they must quack of shots escaped; and of nervous marksmen down whose very gun-barrels they stared
and quacked out defiance. How the veterans of the season must brag, and the Gascons of two put on
airs, and be envied as the heroes of many battles! How they must raise their wings and show their
scars, and be looked up to as ducks of valor and experience!
 PADDLES AND PALETTES.
BY EDWARD L. CHICHESTER.

Concluded from page 510.

FEW miles below Seneca Falls the river forks. One branch, flowing in a
northeasterly direction, is used as the canal; the other, probably at one time
the only course of the river, turns southeast toward Cayuga Lake. A loose pile
of rocks, forming an irregular wall, keeps the water from entirely forsaking the
commercial channel, but enough gushes over and through the barrier to form a
very respectable stream that eddies off between its own banks with a kind of
jolly flow of freedom, like a boy escaped from school.
On reaching this fork, we lifted the canoes over the obstruction and joined
our fortunes with the runaway, much preferring its adventurous course to the one laid down by the
State.
Large trees hung over the water, and an occasional rock or snag, crowned with a matted mass of
eel-grass that floated back on the surface like a mermaid’s hair, lifted its head in front of our bows and
seemed to rush toward us. The stream, though far from being rapid, was at first swift enough to give us
plenty of occupation to avoid obstructions, but, like some people, gained both breadth and repose as it
neared its end.
The village of Cayuga is built on a gentle slope near the foot of the lake by that name. A railroad
passes through the place and turns abruptly west, carried over a mile or so of water on a trestle. North
of the trestle extends the foot of the lake, very shallow here, and full of weeds that end in a bank of
cat-tails, stretching away toward Montezuma. The outlet cuts a broad swath in the flags and winds
slowly northward, now widening into a reedy lake and again narrowing, till the current becomes
perceptible enough to bend the rushes at its sides.
As we glided quietly along our course through the outlet, an occasional duck darted among the
rushes, or a big blue heron lifted himself from the water and flew slowly overhead, preserving his air of
dignity in spite of the long, bare legs sticking out behind. Bass and sunfish, lying close to the surface,
shot away from our bows, streaking the water with little wakes. As the day advanced, we looked
anxiously about for a place to camp, and at last came to an island that lifted itself like a whale’s back
from the surrounding swamp.
To be sure, it was rather bare—a stony ridge, growing mullen stalks and teasels, and inhabited by
some retired army mules, whose gaunt forms stood black against the sky; but it was a relief to see
something higher than the flags, and we gladly landed at the first opening and pulled the boats well up
on the shore.
We had a visit here from a genuine son of the soil, if such a country could be said to possess a soil.
He sauntered down to the camp before we were well settled for the night, and frankly gave us his
opinion of the boats and our other belongings.
He was a queer youngster, not more than fourteen years old, with innocent blue eyes and the
modest air of a little child when he asked questions, but changing instantly to the most reckless
braggadocio when he referred to his own experiences. He was born, he said, at Montezuma, pointing to
a distant spire, and hoped some day to jump from the Brooklyn Bridge. It has been a query in our
minds ever since, whether the mere fact of being born on a flat would gender such ambitions.
Below this island the stream flows under the aqueduct of the Erie Canal, and putting waterproof
blankets over our heads we shot under a dripping arch, coming out dry, but with decks glistening with
the shower-bath. The river widens here, becomes very shallow, and at last spreads out in all directions
like a huge Delta. It was often difficult to find the current, and the air seemed loaded with the
heaviness of the swamp.
Acres of water-lilies spread before us, small flowers of a waxy whiteness gleamed among patches of
sagittaria, and the interminable walls of reeds were weighted down with a plant resembling the hop-
vine, and bearing clusters of pink blossoms, that added their perfume to the heaviness of the air.

Slowly we worked our way through this strange region, the paddles after every stroke coming up
laden with dripping plants, while we were kept anxiously alert lest we should lose our way in the
labyrinth. We occasionally stood up in the boats in vain efforts to see where we were. At one spot the
Sybaris moored herself in a lush mass of lily-pads and grasses, from which the soft mud oozed as her
keel pressed it down, while Simpson, who had been exerting himself manfully, ceased his efforts in
disgust. I took advantage of his experience to avoid the slough, and as I paddled past, heard him
remark, as if to himself: “Query, is this land or water?”
 But, like Bunyan’s pilgrims on the enchanted ground, we “made a good shift and wagged along,” and
before night struck a State ditch—not a canal, but a broad channel dug to drain the region—a channel
with a current that bore us along with scarcely an effort on our part.
We were glad enough to escape, even through a ditch. This was our last day spent in a swamp, for
the country soon became more broken, the water clearer, and the air lost its malarial heaviness and
blew fresh over green hills. Even the mosquito stayed behind.
One evening Simpson was sitting by the fire, having arrived at a good camping-place and put the
Sybaris in order for the night before I had come up. He was frying potatoes, holding the spider in one
hand and running his eye over a letter that had reached him through the Weedsport post-office. He had
laid a stone on the letter to prevent its being blown away, and occasionally his eye would wander from
the closely-written page to the graceful lines of the canoe, whose jauntily striped tent was flapping back
and forth in the breeze.
In addition to these occupations he was singing something about his “Bonny over the Ocean,” and
his voice, which is not unmusical, came floating up to where I had moored the Rena, and was trying to
catch a sunset effect. The musical cadence fell in with the place and hour, and I found myself humming
the air while I worked; but suddenly it stopped, and I paused a moment in my drawing, thinking I heard
thunder.
Certainly there was a roar, though there was no sign of a storm overhead. I put my sketch under the
deck, pushed off the boat, and paddled down toward the camp.
On rounding a point I caught sight of Simpson, running toward the water with the Sybaris clasped in
his arms. She would weigh fully ninety pounds with her tent and bedding, and I was astonished to see
him lug her along in that reckless manner; but in a moment a bull tore through a hedge and bore down
upon him. The canoeist had a good start, and in another moment had run into the river, plunged head-
first into the boat, leaving his heels sticking out from under a torn tent-flap as he floated away, while
the bull stopped short on the shore, pawing and bellowing.
 When my friend’s head emerged from the cockpit the boat was some rods away, and the bull had
turned his attention to the potatoes. It was only by means of a red Jersey flaunted on the end of a
paddle that the animal’s attention was diverted from the camp long enough to rescue the duffle. I
diverted him, as Simpson flatly refused to again assume that rôle.
Nothing was injured but the letter, which had been trampled in the mud.
I naturally felt elated at escaping with so little loss, but Simpson was grumpy all the rest of the
evening.
From Weedsport to Cross Lake the Seneca River winds through a rich, rolling country, and we were
delighted with views of farm-yards with weather-beaten barns and stacks of grain. Fine cattle stood in
shallow places in the stream, chewing their cuds and lazily switching of the flies, and herds of colts
tossed their heads and galloped away as we came suddenly upon them. A settlement of old houses
clustered about the end of a bridge bore the name of Mosquito Point. Though the place provided us
with excellent bread and butter, we did not want to remain there, notwithstanding the inhabitants
stoutly asserted that the village bore a misnomer. “It’s nawthin’ to Montezumy,” remarked one gray-
bearded citizen, whom we took for the oldest inhabitant, and we believed him. They told us a legend
here of the Great Swamp.
The story ran, that a single pair of mosquitoes had their abode there, and these specimens were so
large they would devour an Indian without taking the trouble to peel off the canoe, much as a pig
would eat a beech-nut. In time, the tribes grew restive under this annoyance, and organized a grand
hunt, which resulted in the destruction of their enemies; but while rejoicing over the victory, myriads of
a smaller breed rose from the carcasses, and have infested the country ever since.
One of the pleasantest spots along the whole course of the Seneca River is Cross Lake, a beautiful
sheet of water crossed by the stream. Here we remained some time. The camp was made on a gravelly
beach not far from the village of Jordan. The scenery had that peculiar quality found in an uneven,
partially cleared country.
It composed well.
Some buttonwood grew near us on a side hill. A strip of swampy shore stretched away to the south,
and above us some bars, opening through a rickety fence overhung with bushes, led into a pasture
beyond.

“ASTRIDE THE DECK.”

The owner was going to fix the fence, but had not “got round to it.” We were glad he had not. Early
in the mornings we were awakened by the shrill cries of the tip-ups that fed in the marshy spots with
the woodcocks and schytepokes, the last-mentioned a brown-backed, wading bird, resembling at a
distance a crook-necked squash on stilts. Simpson was fond of shooting at this fowl with his revolver,
for, though holding the views promulgated by the Audubon Society, he said he had not signed the
pledge to abstain from wearing the feathers of non-edible birds—“besides,” he argued, ignoring this
point to make another, “we could eat a schytepoke.” We did not try it, however, mainly because he
never hit one.
On the last night of our stay here we neglected to button down the tents and were well-nigh
drowned out by a storm; but the rain ceased with the first streak of dawn, and the grand panorama
that was disclosed as we stepped out into the fresh wind was worth hours of discomfort to witness. The
clouds, though still black and threatening, were whirling off in ragged masses, and the lake stretched a
steely gray plain, seamed with the dark lines of its waves, and reflecting the first dull glow of the
morning.
The freshness of the air and the sense of conflict felt in a storm made one want to shout, while the
wild grandeur awed one to silence. It did not clear until late that afternoon, and the wind that blew all
day in wet gusts carried us swiftly down the river.
We found the current more rapid as we advanced, and the stream wound between rocky and, at
times, precipitous banks.
At one point a blasted oak stood white against the forest behind, and then flashes of sunlight lit up
stretches of stony pasture or revealed the wet roof of a barn hidden among the trees. As we bowled
along under full sail, I let out the trolling-line and captured some fine black bass and a pike before we
reached Baldwinsville, eight miles away.
Onondaga Lake empties into the Seneca River through a narrow outlet, scarcely a mile long, and
when we reached the mouth of this stream we turned and paddled against the current. As we entered
the lake the city of Syracuse loomed in sight, looking a smoky purple in the distance.
On the left rose the high chimneys of the salt-works of Liverpool, making the village look like a huge
burying-ground dotted with the monuments of a former industry. We secured supplies at this place, and
wandered through some of the buildings, now falling to decay.
In some places nature had tried to soften the outlines of ruin with grass and creeping vines; but tall
brick chimneys do not readily lend themselves to decoration, and there is something in rusting
machinery that reminds one of unburied bones, a kind of skeleton in chains doomed to be a blot on the
landscape so long as the gallows stands.
Half a day’s paddle from the lake brought us to the village of Clay, or New Bridge, as it is commonly
called. This place was old and ruinous, but presented a most picturesque aspect as we came suddenly
upon it, perched on the hillsides on either side of the river.
The unpainted houses, stained a dingy gray by the weather, were embowered in thick masses of
apple and plum trees, and down by the water stood a forsaken warehouse with a sunken canal-boat
before its doors. We spent a Sunday within a mile of the town, and rainy weather kept us some days
longer in the vicinity, so that we had a fine opportunity to study the old place. “God forsaken,” the
farmers called it. It was a sort of supply depot for passing canalers and certainly not a flourishing port,
but perhaps possessed an artistic interest in proportion to its ruin.
“If you want any good eatin’ apples, you’ll find ’em under them trees, an’ there’s green-corn in the
garden beyond; help yourselves.” This hospitable remark was made by a farmer who came to see our
sketches, and it was accompanied with a handful of ripe tomatoes and cucumbers.
 “LANDED FOR SUPPLIES.”

This sort of open-handedness had become a feature of the cruise, and on our last day on the river
we gave a lock-tender a goodly supply of superfluous vegetables. In fact, our living expenses were
made so small by the bounty of the people on whose land we camped, that we felt like distinguished
foreigners who had been given, not the liberty of the town, but of the whole country.
A few miles below Clay the Seneca unites with the Oneida River, the two forming the Oswego at
Three River Point, and by following this broad stream we reached the milling town of Phoenix. We were
delayed here by a short portage, but again in the canoes the stream carried us on, now heaving under
the boats as its deep volume eddied over hidden rocks, or spreading out into placid stretches that
seemed to have no perceptible current.
At one point we were whirled through an eel-weir rift and well spattered with spray; and again,
while passing under a bridge, a sunken pier caught one of the canoes as a submerged monster might
snatch a fly, but fortunately with no damage to the boat. A muskrat, drawing a long line across the
stream, ended it suddenly with the quotation mark of his tail as our bows came almost on him. Then
the river grew broad and still, and paddling on we entered the canal at Fulton. I had an embarrassing
adventure here. I had landed for supplies, and was again getting into the boat that lay some four feet
below, when the uneasy craft slipped under the docking, carrying my feet with her, leaving me hanging
by the elbows and shouting for Simpson, who was some distance away.
The muddy water of the canal never seemed less inviting than during those anxious moments, as I
hung with my arms gradually slipping, certain, if the Sybaris did not come quickly, of going in head
foremost. But fortunately she came quickly and I was rescued dry.
Below Fulton lies the historical spot known as Battle Island, the theatre of some exciting events of
the war of 1812. Near this island the river is obstructed by a dam, and here we lowered the boats over
with ropes.
The Sybaris went first, and, once over, shot off through a stretch of rapid water.
Simpson, in his efforts to guide her, broke his paddle, and was obliged to jump overboard in order to
keep her off the rocks. He came back dripping to help me with the Rena, and told me exactly how to
steer when I was cast adrift; but in rapids a little experience is certainly worth more than a good many
directions; and once started I found it useless to try to recall a word he had said. The sensation of
being carried through a rift is certainly peculiar. With the attention so closely exerted to avoid danger,
the boatman has no opportunity to watch the shores, and, as the Irishman expressed it, “see himself go
by.” On the contrary, he must fix his gaze forward, and soon has the feeling of standing quite still, while
the rocks bob up in front of him and rush at his boat. As I whirled along, a formidable line of boulders
rose at my left and swung steadily around to embrace me. Work as I would, they came nearer and
nearer, then there was an ominous grating, a rattle of iron (I carried the pots and kettles), and the Rena
stuck fast, with the water surging and boiling round her. I expected she would roll over, but she lay
wedged just where she struck, and observing there was no change, I pulled off my shoes, and, taking
hold of the combing, raised myself out, and sat down astride the deck just back of the cockpit.

“NOT EXACTLY A PADDLE.”

I had not calculated the effect of this change of position on the boat, for her stern dropped instantly,
and rearing like an impatient sea-horse she dashed forward, while I clung on as well as I could, feeling
like an amateur Neptune, or “a water imp,” as Simpson said. But I was really a little nervous at the time
and much relieved to reach still water in safety.
Lower down we landed, and my friend mended his paddle, and then stretched himself out in the sun
and read “Lorna Doone” till his clothes were dry. Then we went on—gliding under overhanging trees,
passing bare sand-banks crowned with sumac, and catching glimpses of little gullies full of poplars, and
fence corners yellow with golden-rod. Some houses and barns strung along the hill-top marked the
outskirts of Bundy’s Corners, and later we heard the roar of a fall, down at Minetto.
When we reached this village we found another high dam with a wooden apron below.
We inquired particularly about the channel: Was it deep under the dam? Did boats ever go over?—
Questions the people who came down to see the canoes answered readily. It was deep on the other
side, and flat-bottom boats had gone over. “Then we can go,” said Simpson, and pushed off with his
paddle.
I followed, and we skirted the upper edge of the dam, cautiously working across the river. The water
overflowed the obstruction in one thin sheet, and fell spattering among piles of ugly-looking stones,
until we reached the extreme east end; here a breach had been made and a heavy stream poured itself
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookfinal.com