Digital Image Processing with C++: Implementing Reference Algorithms with the CImg Library

309 Pages·2023·53.822 MB·English

by  David Tschumperle, Christophe Tilmant, Vincent Barra

Digital Image Processing
with C++
Digital Image Processing with C++: Implementing Reference Algorithms with the Clmg Library
presents the theory of digital image processing and implementations of algorithms using a dedi-
cated library. Processing a digital image means transforming its content (denoising, stylizing,
etc.) or extracting information to solve a given problem (object recognition, measurement, mo-
tion estimation, etc.). This book presents the mathematical theories underlying digital image
processing as well as their practical implementation through examples of algorithms implement-
ed in the C++ language using the free and easy-to-use CImg library.
Chapters cover the field of digital image processing in a broad way and propose practical and
functional implementations of each method theoretically described. The main topics covered
include filtering in spatial and frequency domains, mathematical morphology, feature extraction
and applications to segmentation, motion estimation, multispectral image processing and 3D
visualization.
Students or developers wishing to discover or specialize in this discipline and teachers and re-
searchers hoping to quickly prototype new algorithms or develop courses will all find in this
book material to discover image processing or deepen their knowledge in this field.
David Tschumperlé is a permanent CNRS research scientist heading the IMAGE team at the
GREYC Laboratory in Caen, France. He’s particularly interested in partial differential equations
and variational methods for processing multi-valued images in a local or non-local way. He has
authored more than 40 papers in journals or conferences and is the project leader of CImg and
G’MIC, two open-source software/libraries.
Christophe Tilmant is associate professor in computer science at Clermont-Auvergne Univer-
sity. His research activities include image processing and artificial intelligence, where he has au-
thored more than 30 papers. His teaching includes deep learning, image processing and network
security. He participates or leads several French research programs.
Vincent Barra is a full professor in computer science at Clermont-Auvergne University and
associate director of the LIMOS Lab. He teaches artificial intelligence and image processing in
engineering schools and master’s programs. His research activities focus on n-dimensional data
analysis with methodological and application aspects in various fields. He has authored more
than 90 papers in journals or conferences and participates or leads several French and European
research programs.
Taylor & Francis
Taylor & Francis Group
http://taylorandfrancis.com
Digital Image Processing
with C++
Implementing Reference Algorithms
with the CImg Library
David Tschumperlé
Christophe Tilmant
Vincent Barra
First edition published 2023
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742
and by CRC Press
4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
© 2023 David Tschumperlé, Christophe Tilmant and Vincent Barra
CRC Press is an imprint of Taylor & Francis Group, LLC
Title of the original French edition, Le traitement numerique des images en C++. Implementation
d’algorithmes avec la bibliotheque CIMG- published by Ellipses - Copyright 2021, Edition Marketing S. A.
Reasonable efforts have been made to publish reliable data and information, but the author and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The authors
and publishers have attempted to trace the copyright holders of all material reproduced in this publica-
tion and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any future
reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, trans-
mitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.com or
contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-
8400. For works that are not available on CCC please contact [e-mail]
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used
only for identification and explanation without intent to infringe.
ISBN: 978-1-032-34752-3 (hbk)
ISBN: 978-1-032-34753-0 (pbk)
ISBN: 978-1-003-32369-3 (ebk)
DOI: 10.1201/9781003323693
Typeset in Nimbus Roman
by KnowledgeWorks Global Ltd.
Publisher’s note: This book has been prepared from camera-ready copy provided by the author.
Contents
Preface ............................................. xi
Preamble ........................................... xv
I INTRODUCTION TO CImg
1 Introduction .......................................... 3
2 Getting Started with the CImg Library ................. 17
2.1 Objective: subdivide an image into blocks 17
2.2 Setup and first program 18
2.3 Computing the variations 19
2.4 Computing the block decomposition 24
2.5 Rendering of the decomposition 26
2.6 Interactive visualization 31
2.7 Final source code 36
II IMAGE PROCESSING USING CImg
3 Point Processing Transformations ...................... 43
3.1 Image operations 43
3.1.1 Mathematicaltransformations . . . . . . . . . . . . . . . . . . . . . . . . 43
3.1.2 Bitwisetransformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.3 Contrastenhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
vi Table of contents
3.2 Histogram operations 48
3.2.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.2.2 Histogramspecification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.2.3 Localhistogramspecification . . . . . . . . . . . . . . . . . . . . . . . . . 50
4 Mathematical Morphology ........................... 53
4.1 Binary images 54
4.1.1 Dilationanderosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.1.2 Openingandclosing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Gray-level images 58
4.3 Some applications 59
4.3.1 Kramer-Brucknerfilter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.3.2 Alternatingsequentialfilters . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.3.3 Morphologicalgradients . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3.4 Skeletonization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
5 Filtering ............................................. 69
5.1 Spatial filtering 69
5.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.1.2 Low-passfilters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5.1.3 High-passfilters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.1.4 Adaptivefilters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.1.5 Adaptivewindowfilters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2 Recursive filtering 84
5.2.1 Optimaledgedetection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.2.2 Derichefilter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.3 Frequency filtering 94
5.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.3.2 TheFouriertransform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.3.3 Frequencyfiltering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.3.4 ProcessingaMoiréimage . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.4 Diffusion filtering 110
5.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.4.2 Physicalbasisofdiffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
5.4.3 Lineardiffusionfilter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Table of contents vii
5.4.4 Non-lineardiffusionfilterintwodimensions. . . . . . . . . . . . . . 114
5.4.5 Non-lineardiffusionfilteronavideosequence . . . . . . . . . . 117
6 Feature Extraction .................................. 121
6.1 Points of interest 121
6.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.1.2 HarrisandStephensdetector . . . . . . . . . . . . . . . . . . . . . . . . 122
6.1.3 ShiandTomasialgorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
6.1.4 Pointsofinterestwithsub-pixelaccuracy . . . . . . . . . . . . . . . 127
6.2 Hough transform 128
6.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.2.2 Linedetection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.2.3 Circleandellipsedetection . . . . . . . . . . . . . . . . . . . . . . . . . 134
6.3 Texture features 137
6.3.1 Texturespectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
6.3.2 Tamuracoefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6.3.3 Localbinarypattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
6.3.4 Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
7 Segmentation ...................................... 151
7.1 Edge-based approaches 151
7.1.1 Introductiontoimplicitactivecontours . . . . . . . . . . . . . . . . 151
7.1.2 Implicitrepresentationofacontour . . . . . . . . . . . . . . . . . . . 156
7.1.3 Evolutionequation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
7.1.4 Discretizationoftheevolutionequation . . . . . . . . . . . . . . . . 160
7.1.5 Geodesicmodelpropagationalgorithm . . . . . . . . . . . . . . . 161
7.2 Region-based approaches 163
7.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
7.2.2 Histogram-basedmethods . . . . . . . . . . . . . . . . . . . . . . . . . . 163
7.2.3 Thresholdingbyclustering . . . . . . . . . . . . . . . . . . . . . . . . . . 167
7.2.4 Transformationofregions . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
7.2.5 Super-pixelspartitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
viii Table of contents
8 Motion Estimation ................................... 183
8.1 Optical flow: dense motion estimation 183
8.1.1 Variationalmethods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
8.1.2 LucasandKanadedifferentialmethod . . . . . . . . . . . . . . . . 189
8.1.3 Affineflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
8.2 Sparse estimation 195
8.2.1 Displacementfieldusingspatialcorrelation . . . . . . . . . . . . . 196
8.2.2 Displacementfieldusingphasecorrelation . . . . . . . . . . . . . 198
8.2.3 Kalmanfiltering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
9 Multispectral Approaches ........................... 209
9.1 Dimension reduction 209
9.1.1 Principalcomponentanalysis. . . . . . . . . . . . . . . . . . . . . . . . 210
9.1.2 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
9.2 Color imaging 213
9.2.1 Colorimetricspaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
9.2.2 Medianfilteringincolorimaging . . . . . . . . . . . . . . . . . . . . . 218
9.2.3 Edgedetectionincolorimaging . . . . . . . . . . . . . . . . . . . . . 220
10 3D Visualisation .................................... 227
10.1 Structuring of 3D mesh objects 227
10.2 3D plot of a function z= f(x,y) 229
10.3 Creating complex 3D objects 233
10.3.1 Detailsonvertexstructuring . . . . . . . . . . . . . . . . . . . . . . . . . 233
10.3.2 Detailsonprimitivestructuring . . . . . . . . . . . . . . . . . . . . . . . 234
10.3.3 Detailsonmaterialstructuring . . . . . . . . . . . . . . . . . . . . . . . 235
10.3.4 Detailsonopacitystructuring . . . . . . . . . . . . . . . . . . . . . . . . 235
10.4 Visualization of a cardiac segmentation in MRI 236
10.4.1 Descriptionoftheinputdata . . . . . . . . . . . . . . . . . . . . . . . . 236
10.4.2 Extractionofthe3Dsurfaceoftheventricle . . . . . . . . . . . . 237
10.4.3 Adding3Dmotionvectors . . . . . . . . . . . . . . . . . . . . . . . . . . 238
10.4.4 Addingcuttingplanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
10.4.5 Finalresult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Table of contents ix
11 And So Many Other Things... ....................... 243
11.1 Compression by transform (JPEG) 243
11.1.1 Introduction-Compressionbytransform . . . . . . . . . . . . . . . 243
11.1.2 JPEGAlgorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
11.1.3 Discretecosinetransformandquantization . . . . . . . . . . . . . 245
11.1.4 SimplifiedJPEGalgorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
11.2 Tomographic reconstruction 252
11.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
11.2.2 Analyticaltomographicreconstruction . . . . . . . . . . . . . . . . 254
11.2.3 Algebraictomographicreconstruction . . . . . . . . . . . . . . . . 259
11.3 Stereovision 264
11.3.1 Epipolargeometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
11.3.2 Depthestimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
11.4 Interactive deformation using RBF 273
11.4.1 Goaloftheapplication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
11.4.2 TheRBF interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
11.4.3 RBF forimagewarping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
11.4.4 Userinterfaceforkeypointmanagement. . . . . . . . . . . . . . . 277
List of CImg Codes ................................. 281
Bibliography ....................................... 285
Index .............................................. 289