Muutke küpsiste eelistusi

Feature Extraction and Image Processing for Computer Vision 4th edition [Pehme köide]

3.85/5 (13 hinnangut Goodreads-ist)
(Principal Programmer, Sportradar, Brighton, UK), (Professor of Electronics and Computer Science, University of Southampton, UK)
  • Formaat: Paperback / softback, 650 pages, kõrgus x laius: 235x191 mm, kaal: 1340 g
  • Ilmumisaeg: 18-Nov-2019
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0128149760
  • ISBN-13: 9780128149768
Teised raamatud teemal:
Feature Extraction and Image Processing for Computer Vision 4th editionSuurem pilt
  • Formaat: Paperback / softback, 650 pages, kõrgus x laius: 235x191 mm, kaal: 1340 g
  • Ilmumisaeg: 18-Nov-2019
  • Kirjastus: Academic Press Inc
  • ISBN-10: 0128149760
  • ISBN-13: 9780128149768
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780128149768.html
Märksõnad:

Feature Extraction for Image Processing and Computer Vision is an essential guide to the implementation of image processing and computer vision techniques, with tutorial introductions and sample code in MATLAB and Python. Algorithms are presented and fully explained to enable complete understanding of the methods and techniques demonstrated. As one reviewer noted, "The main strength of the proposed book is the link between theory and exemplar code of the algorithms." Essential background theory is carefully explained.

This text gives students and researchers in image processing and computer vision a complete introduction to classic and state-of-the art methods in feature extraction together with practical guidance on their implementation.

  • The only text to concentrate on feature extraction with working implementation and worked through mathematical derivations and algorithmic methods
  • A thorough overview of available feature extraction methods including essential background theory, shape methods, texture and deep learning
  • Up to date coverage of interest point detection, feature extraction and description and image representation (including frequency domain and colour)
  • Good balance between providing a mathematical background and practical implementation
  • Detailed and explanatory of algorithms in MATLAB and Python
Preface xix
1 Introduction
1(1)
1.1 Overview
1(1)
1.2 Human and computer vision
1(2)
1.3 The human vision system
3(4)
1.3.1 The eye
4(3)
1.3.2 The neural system
7(1)
1.3.3 Processing
8(2)
1.4 Computer vision systems
10(5)
1.4.1 Cameras
11(3)
1.4.2 Computer interfaces
14(1)
1.5 Processing images
15(12)
1.5.1 Processing
15(2)
1.5.2 Hello Python, hello images!
17(4)
1.5.3 Mathematical tools
21(2)
1.5.4 Hello Matlab
23(4)
1.6 Associated literature
27(4)
1.6.1 Journals, magazines and conferences
27(1)
1.6.2 Textbooks
28(3)
1.6.3 The web
31(1)
1.7 Conclusions
31(4)
References
31(4)
2 Images, sampling and frequency domain processing
35(1)
2.1 Overview
35(1)
2.2 Image formation
35(4)
2.3 The Fourier Transform
39(5)
2.4 The sampling criterion
44(6)
2.5 The discrete Fourier Transform
50(11)
2.5.1 One-dimensional transform
50(3)
2.5.2 Two-dimensional transform
53(8)
2.6 Properties of the Fourier Transform
61(5)
2.6.1 Shift invariance
61(1)
2.6.2 Rotation
62(2)
2.6.3 Frequency scaling
64(1)
2.6.4 Superposition (linearity)
64(2)
2.6.5 The importance of phase
66(1)
2.7 Transforms other than Fourier
66(10)
2.7.1 Discrete cosine transform
66(2)
2.7.2 Discrete Hartley Transform
68(2)
2.7.3 Introductory wavelets
70(1)
2.7.3.1 Gabor Wavelet
70(2)
2.7.3.2 Haar Wavelet
72(4)
2.7.4 Other transforms
76(1)
2.8 Applications using frequency domain properties
76(3)
2.9 Further reading
79(4)
References
79(4)
3 Image processing
83(1)
3.1 Overview
83(1)
3.2 Histograms
84(1)
3.3 Point operators
84(12)
3.3.1 Basic Point operations
84(3)
3.3.2 Histogram normalisation
87(1)
3.3.3 Histogram equalisation
88(2)
3.3.4 Thresholding
90(6)
3.4 Group operations
96(10)
3.4.1 Template convolution
96(3)
3.4.2 Averaging operator
99(1)
3.4.3 On different template size
99(1)
3.4.4 Template convolution via the Fourier transform
100(3)
3.4.5 Gaussian averaging operator
103(3)
3.4.6 More on averaging
106(1)
3.5 Other image processing operators
106(20)
3.5.1 Median filter
106(2)
3.5.2 Mode filter
108(3)
3.5.3 Nonlocal means
111(3)
3.5.4 Bilateral filtering
114(1)
3.5.5 Anisotropic diffusion
115(7)
3.5.6 Comparison of smoothing operators
122(1)
3.5.7 Force field transform
123(2)
3.5.8 Image ray transform
125(1)
3.6 Mathematical morphology
126(10)
3.6.1 Morphological operators
128(2)
3.6.2 Grey level morphology
130(2)
3.6.3 Grey level erosion and dilation
132(1)
3.6.4 Minkowski operators
133(3)
3.7 Further reading
136(5)
References
136(5)
4 Low-level feature extraction (including edge detection)
141(1)
4.1 Overview
141(2)
4.2 Edge detection
143(1)
4.2.1 First-order edge detection operators
143(1)
4.2.1.1 Basic operators
143(1)
4.2.1.2 Analysis of the basic operators
144(2)
4.2.1.3 Prewitt edge detection operator
146(2)
4.2.1.4 Sobel edge detection operator
148(6)
4.2.1.5 The Canny edge detector
154(8)
4.2.2 Second-order edge detection operators
162(1)
4.2.2.1 Motivation
162(1)
4.2.2.2 Basic operators: The Laplacian
162(2)
4.2.2.3 The Marr-Hildreth operator
164(4)
4.2.3 Other edge detection operators
168(2)
4.2.4 Comparison of edge detection operators
170(1)
4.2.5 Further reading on edge detection
171(1)
4.3 Phase congruency
172(6)
4.4 Localised feature extraction
178(25)
4.4.1 Detecting image curvature (corner extraction)
178(1)
4.4.1.1 Definition of curvature
178(2)
4.4.1.2 Computing differences in edge direction
180(2)
4.4.1.3 Measuring curvature by changes in intensity (differentiation)
182(3)
4.4.1.4 Moravec and Harris detectors
185(4)
4.4.1.5 Further reading on curvature
189(1)
4.4.2 Feature point detection; region/patch analysis
190(1)
4.4.2.1 Scale invariant feature transform
190(3)
4.4.2.2 Speeded up robust features
193(1)
4.4.2.3 FAST, ORB, FREAK, LOCKY and other keypoint detectors
194(4)
4.4.2.4 Other techniques and performance issues
198(1)
4.4.3 Saliency
198(1)
4.4.3.1 Basic saliency
198(1)
4.4.3.2 Context aware saliency
199(3)
4.4.3.3 Other saliency operators
202(1)
4.5 Describing image motion
203(14)
4.5.1 Area-based approach
204(3)
4.5.2 Differential approach
207(7)
4.5.3 Recent developments: deep flow, epic flow and extensions
214(1)
4.5.4 Analysis of optical flow
215(2)
4.6 Further reading
217(6)
References
217(6)
5 High-level feature extraction: fixed shape matching
223(1)
5.1 Overview
223(2)
5.2 Thresholding and subtraction
225(2)
5.3 Template matching
227(13)
5.3.1 Definition
227(7)
5.3.2 Fourier transform implementation
234(5)
5.3.3 Discussion of template matching
239(1)
5.4 Feature extraction by low-level features
240(7)
5.4.1 Appearance-based approaches
240(1)
5.4.1.1 Object detection by templates
240(1)
5.4.1.2 Object detection by combinations of parts
241(1)
5.4.2 Distribution-based descriptors
242(1)
5.4.2.1 Description by interest points (SIFT, SURF, BRIEF)
242(4)
5.4.2.2 Characterising object appearance and shape
246(1)
5.5 Hough transform
247(38)
5.5.1 Overview
247(1)
5.5.2 Lines
248(6)
5.5.3 HT for circles
254(4)
5.5.4 HT for ellipses
258(2)
5.5.5 Parameter space decomposition
260(1)
5.5.5.1 Parameter space reduction for lines
261(2)
5.5.5.2 Parameter space reduction for circles
263(5)
5.5.5.3 Parameter space reduction for ellipses
268(3)
5.5.6 Generalised Hough transform
271(2)
5.5.6.1 Formal definition of the GHT
273(2)
5.5.6.2 Polar definition
275(1)
5.5.6.3 The GHT technique
275(4)
5.5.6.4 Invariant GHT
279(5)
5.5.7 Other extensions to the HT
284(1)
5.6 Further reading
285(2)
References
287(4)
6 High-level feature extraction: deformable shape analysis
291(1)
6.1 Overview
291(1)
6.2 Deformable shape analysis
292(1)
6.2.1 Deformable templates
292(2)
6.2.2 Parts-based shape analysis
294(2)
6.3 Active contours (snakes)
296(23)
6.3.1 Basics
296(2)
6.3.2 The Greedy Algorithm for snakes
298(5)
6.3.3 Complete (Kass) Snake implementation
303(5)
6.3.4 Other Snake approaches
308(1)
6.3.5 Further Snake developments
309(4)
6.3.6 Geometric active contours (Level Set-Based Approaches)
313(6)
6.4 Shape Skeletonisation
319(9)
6.4.1 Distance transforms
319(2)
6.4.2 Symmetry
321(7)
6.5 Flexible shape models -- active shape and active appearance
328(4)
6.6 Further reading
332(7)
References
339
7 Object description
339(1)
7.1 Overview and invariance requirements
339(1)
7.2 Boundary descriptions
340(28)
7.2.1 Boundary and region
340(1)
7.2.2 Chain codes
341(3)
7.2.3 Fourier descriptors
344(1)
7.2.3.1 Basis of Fourier descriptors
344(2)
7.2.3.2 Fourier expansion
346(2)
7.2.3.3 Shift invariance
348(1)
7.2.3.4 Discrete computation
349(2)
7.2.3.5 Cumulative angular function
351(9)
7.2.3.6 Elliptic Fourier descriptors
360(3)
7.2.3.7 Invariance
363(5)
7.3 Region descriptors
368(28)
7.3.1 Basic region descriptors
368(4)
7.3.2 Moments
372(1)
7.3.2.1 Definition and properties
372(1)
7.3.2.2 Geometric moments
373(2)
7.3.2.3 Geometric complex moments and centralised moments
375(2)
7.3.2.4 Rotation and scale invariant moments
377(6)
7.3.2.5 Zernike moments
383(4)
7.3.2.6 Tchebichef moments
387(1)
7.3.2.7 Krawtchouk moments
387(7)
7.3.2.8 Other moments
394(2)
7.4 Further reading
396(3)
References
396(3)
8 Region-based analysis
399(1)
8.1 Overview
399(1)
8.2 Region-based analysis
400(11)
8.2.1 Watershed transform
400(4)
8.2.2 Maximally stable extremal regions
404(3)
8.2.3 Superprxels
407(1)
8.2.3.1 Basic techniques and normalised cuts
407(1)
8.2.3.2 Simple linear iterative clustering
407(4)
8.3 Texture description and analysis
411(18)
8.3.1 What is texture?
411(2)
8.3.2 Performance requirements
413(3)
8.3.3 Structural approaches
416(1)
8.3.4 Statistical approaches
416(2)
8.3.4.1 Co-occurrence matrix
418(1)
8.3.4.2 Learning-based approaches
418(1)
8.3.5 Combination approaches
419(2)
8.3.6 Local binary patterns
421(6)
8.3.7 Other approaches
427(1)
8.3.8 Segmentation by texture
427(2)
8.4 Further reading
429(4)
References
429(4)
9 Moving object detection and description
433(1)
9.1 Overview
433(1)
9.2 Moving object detection
434(1)
9.2.1 Basic approaches
434(1)
9.2.1.1 Detection by subtracting the background
434(4)
9.2.1.2 Improving quality by morphology
438(1)
9.2.2 Modelling and adapting to the (static) background
439(5)
9.2.3 Background segmentation by thresholding
444(2)
9.2.4 Problems and advances
446(1)
9.3 Tracking moving features
447(21)
9.3.1 Tracking moving objects
447(1)
9.3.2 Tracking by local search
448(3)
9.3.3 Problems in tracking
451(1)
9.3.4 Approaches to tracking
451(1)
9.3.5 MeanShift and Camshift
452(1)
9.3.5.1 Kernel-based density estimation
453(3)
9.3.5.2 MeanShift tracking
456(5)
9.3.5.3 Camshift technique
461(4)
9.3.6 Other approaches
465(3)
9.4 Moving feature extraction and description
468(9)
9.4.1 Moving (biological) shape analysis
468(2)
9.4.2 Space--time interest points
470(1)
9.4.3 Detecting moving shapes by shape matching in image sequences
470(4)
9.4.4 Moving shape description
474(3)
9.5 Further reading
477(6)
References
478(5)
10 Camera geometry fundamentals
483(1)
10.1 Overview
483(1)
10.2 Projective space
483(1)
10.2.1 Homogeneous co-ordinates and projective geometry
484(1)
10.2.2 Representation of a line, duality and ideal points
485(2)
10.2.3 Transformations in the projective space
487(3)
10.2.4 Computing a planar homography
490(3)
10.3 The perspective camera
493(9)
10.3.1 Perspective camera model
494(4)
10.3.2 Parameters of the perspective camera model
498(1)
10.3.3 Computing a projection from an image
498(4)
10.4 Affine camera
502(5)
10.4.1 Affine camera model
503(1)
10.4.2 Affine camera model and the perspective projection
504(2)
10.4.3 Parameters of the affine camera model
506(1)
10.5 Weak perspective model
507(1)
10.6 Discussion
508(1)
10.7 Further reading
509(2)
References
510(1)
11 Colour images
511(1)
11.1 Overview
511(1)
11.2 Colour image theory
512(1)
11.2.1 Colour images
512(1)
11.2.2 Tristimulus theory
513(2)
11.2.3 The colourimetric equation
515(1)
11.2.4 Luminosity function
516(1)
11.3 Perception-based colour models: CIE RGB and CIE XYZ
517(21)
11.3.1 CIE RGB colour model: Wright--Guild data
518(1)
11.3.2 CIE RGB colour matching functions
519(3)
11.3.3 CIE RGB chromaticity diagram and chromaticity co-ordinates
522(2)
11.3.4 CIE XYZ colour model
524(5)
11.3.5 CIE XYZ colour matching functions
529(3)
11.3.6 XYZ chromaticity diagram
532(1)
11.3.7 Uniform colour spaces: CIE LUV and CIE LAB
533(5)
11.4 Additive and subtractive colour models
538(7)
11.4.1 RGB and CMY
538(2)
11.4.2 Transformation between RGB models
540(3)
11.4.3 Transformation between RGB and CMY models
543(2)
11.5 Luminance and chrominance colour models
545(8)
11.5.1 YUV, YIQ and YCbCr models
545(1)
11.5.2 Luminance and gamma correction
546(3)
11.5.3 Chrominance
549(1)
11.5.4 Transformations between YUV, YIQ and RGB colour models
550(1)
11.5.5 Colour model for component video: YPbPr
550(1)
11.5.6 Colour model for digital video: YCbCr
551(2)
11.6 Additive perceptual colour models
553(15)
11.6.1 The HSV and HLS colour models
553(1)
11.6.2 The hexagonal model: HSV
554(6)
11.6.3 The triangular model: HLS
560(4)
11.6.4 Transformation between HLS and RGB
564(4)
11.7 More colour models
568(3)
References
569(2)
12 Distance, classification and learning
571(1)
12.1 Overview
571(1)
12.2 Basis of classification and learning
571(3)
12.3 Distance and classification
574(1)
12.3.1 Distance measures
574(1)
12.3.1.1 Manhattan and Euclidean Ln norms
575(1)
12.3.1.2 Mahalanobis, Bhattacharrya and Matusita
576(5)
12.3.1.3 Histogram intersection, Chi2 (Χ2) and the Earth Mover's distance
581(3)
12.3.2 The k-nearest neighbour for classification
584(4)
12.4 Neural networks and Support Vector Machines
588(3)
12.5 Deep learning
591(10)
12.5.1 Basis of deep learning
591(5)
12.5.2 Major deep learning architectures
596(2)
12.5.3 Deep learning for feature extraction
598(3)
12.5.4 Deep learning performance evaluation
601(1)
12.6 Further reading
601(4)
References
602(3)
Index 605
Mark Nixon is the Professor in Computer Vision at the University of Southampton UK. His research interests are in image processing and computer vision. His team develops new techniques for static and moving shape extraction which have found application in biometrics and in medical image analysis. His team were early workers in automatic face recognition, later came to pioneer gait recognition and more recently joined the pioneers of ear biometrics. With Tieniu Tan and Rama Chellappa, their book Human ID based on Gait is part of the Springer Series on Biometrics and was published in 2005. He has chaired/ program chaired many conferences (BMVC 98, AVBPA 03, IEEE Face and Gesture FG06, ICPR 04, ICB 09, IEEE BTAS 2010) and given many invited talks. Dr. Nixon is a Fellow IET and a Fellow IAPR. Alberto Aguado is a principal algorithm researcher and developer at Foundry London were he works developing Image Processing, Computer Vision and rendering technologies for video production. Previously, he was head of research on animation technologies at Natural Motion. He developed image processing technologies for sport tracking at Sportradar. He worked as developer for Electronic Arts and for Black Rock Disney Game Studios. He gained academic experience as a Lecturer in the Centre for Vision, Speech and Signal Processing in the University of Surrey. He pursued a postdoctoral fellowship in Computer Vision at INRIA Rhône-Alpes (Marie Curie fellowship) and he received his PhD in Computer Vision /Image Processing from the University of Southampton.