A Practical Introduction to Computer Vision with OpenCV

Kenneth Dawson-Howe, School of Computer Science and Statistics, Trinity College Dublin, Ireland
Dr. Dawson-Howe is a Lecturer in the School of Computer Science and Statistics and part of the Graphics, Vision and Visualisation (GV2) Research Group at Trinity College Dublin. He currently teaches the course Computer Vision/Vision Systems to final year undergraduate and Masters students. He has been teaching courses in the area of computer vision for over 20 years. He is co-author of the Dictionary of Computer Vision & Image Processing published by Wiley in 2005 (2nd Edition to publish December 2013).

Explains the theory behind basic computer vision and provides a bridge from the theory to practical implementation using the industry standard OpenCV libraries
 
Computer Vision is a rapidly expanding area and it is becoming progressively easier for developers to make use of this field due to the ready availability of high quality libraries (such as OpenCV 2). This text is intended to facilitate the practical use of computer vision with the goal being to bridge the gap between the theory and the practical implementation of computer vision. The book will explain how to use the relevant OpenCV library routines and will be accompanied by a full working program including the code snippets from the text. This textbook is a heavily illustrated, practical introduction to an exciting field, the applications of which are becoming almost ubiquitous. We are now surrounded by cameras, for example cameras on computers & tablets/ cameras built into our mobile phones/ cameras in games consoles; cameras imaging difficult modalities (such as ultrasound, X-ray, MRI) in hospitals, and surveillance cameras. This book is concerned with helping the next generation of computer developers to make use of all these images in order to develop systems which are more intuitive and interact with us in more intelligent ways.
* Explains the theory behind basic computer vision and provides a bridge from the theory to practical implementation using the industry standard OpenCV libraries
* Offers an introduction to computer vision, with enough theory to make clear how the various algorithms work but with an emphasis on practical programming issues
* Provides enough material for a one semester course in computer vision at senior undergraduate and Masters levels
* Includes the basics of cameras and images and image processing to remove noise, before moving on to topics such as image histogramming; binary imaging; video processing to detect and model moving objects; geometric operations & camera models; edge detection; features detection; recognition in images
* Contains a large number of vision application problems to provide students with the opportunity to solve real problems. Images or videos for these problems are provided in the resources associated with this book which include an enhanced eBook

Preface xiii
 
1 Introduction 1
 
1.1 A Difficult Problem 1
 
1.2 The Human Vision System 2
 
1.3 Practical Applications of Computer Vision 3
 
1.4 The Future of Computer Vision 5
 
1.5 Material in This Textbook 6
 
1.6 Going Further with Computer Vision 7
 
2 Images 9
 
2.1 Cameras 9
 
2.1.1 The Simple Pinhole Camera Model 9
 
2.2 Images 10
 
2.2.1 Sampling 11
 
2.2.2 Quantisation 11
 
2.3 Colour Images 13
 
2.3.1 Red-Green-Blue (RGB) Images 14
 
2.3.2 Cyan-Magenta-Yellow (CMY) Images 17
 

2.3.3 YUV Images 17
 
2.3.4 Hue Luminance Saturation (HLS) Images 18
 
2.3.5 Other Colour Spaces 20
 
2.3.6 Some Colour Applications 20
 
2.4 Noise 22
 
2.4.1 Types of Noise 23
 
2.4.2 Noise Models 25
 
2.4.3 Noise Generation 26
 
2.4.4 Noise Evaluation 26
 
2.5 Smoothing 27
 
2.5.1 Image Averaging 27
 
2.5.2 Local Averaging and Gaussian Smoothing 28
 
2.5.3 Rotating Mask 30
 
2.5.4 Median Filter 31
 
3 Histograms 35
 
3.1 1D Histograms 35
 
3.1.1 Histogram Smoothing 36
 
3.1.2 Colour Histograms 37
 
3.2 3D Histograms 39
 
3.3 Histogram/Image Equalisation 40
 
3.4 Histogram Comparison 41
 
3.5 Back-projection 43
 
3.6 k-means Clustering 44
 
4 Binary Vision 49
 
4.1 Thresholding 49
 
4.1.1 Thresholding Problems 50
 
4.2 Threshold Detection Methods 51
 
4.2.1 Bimodal Histogram Analysis 52
 
4.2.2 Optimal Thresholding 52
 
4.2.3 Otsu Thresholding 54
 
4.3 Variations on Thresholding 56
 
4.3.1 Adaptive Thresholding 56
 
4.3.2 Band Thresholding 57
 
4.3.3 Semi-thresholding 58
 
4.3.4 Multispectral Thresholding 58
 
4.4 Mathematical Morphology 59
 
4.4.1 Dilation 60
 
4.4.2 Erosion 62
 
4.4.3 Opening and Closing 63
 
4.4.4 Grey-scale and Colour Morphology 65
 
4.5 Connectivity 66
 
4.5.1 Connectedness: Paradoxes and Solutions 66
 
4.5.2 Connected Components Analysis 67
 
5 Geometric Transformations 71
 
5.1 Problem Specification and Algorithm 71
 
5.2 Affine Transformations 73
 
5.2.1 Known Affine Transformations 74
 
5.2.2 Unknown Affine Transformations 75
 
5.3 Perspective Transformations 76
 
5.4 Specification of More Complex Transformations 78
 
5.5 Interpolation 78
 
5.5.1 Nearest Neighbour Interpolation 79
 
5.5.2 Bilinear Interpolation 79
 
5.5.3 Bi-Cubic Interpolation 80
 
5.6 Modelling and Removing Distortion from Cameras 80
 
5.6.1 Camera Distortions 81
 
5.6.2 Camera Calibration and Removing Distortion 82
 
6 Edges 83
 
6.1 Edge Detection 83
 
6.1.1 First Derivative Edge Detectors 85
 
6.1.2 Second Derivative Edge Detectors 92
 
6.1.3 Multispectral Edge Detection 97
 
6.1.4 Image Sharpening 98
 
6.2 Contour Segmentation 99
 
6.2.1 Basic Representations of Edge Data 99
 
6.2.2 Border Detection 102
 
6.2.3 Extracting Line Segment Representations of Edge Contours 105
 
6.3 Hough Transform 108
 
6.3.1 Hough for Lines 109
 
6.3.2 Hough for Circles 111
 
6.3.3 Generalised Hough 112
 
7 Features 115
 
7.1 Moravec Corner Detection 117
 
7.2 Harris Corner Detection 118
 
7.3 FAST Corner Detection 121
 
7.4 SIFT 122
 
7.4.1 Scale Space E