Muutke küpsiste eelistusi

Image Restoration: Fundamentals and Advances [Kõva köide]

Edited by (West Virginia University, Morgantown, USA), Edited by (Louisiana State University, Baton Rouge, USA)
  • Formaat: Hardback, 378 pages, kõrgus x laius: 254x178 mm, kaal: 907 g, 13 Tables, black and white; 45 Illustrations, color; 118 Illustrations, black and white
  • Sari: Digital Imaging and Computer Vision
  • Ilmumisaeg: 11-Sep-2012
  • Kirjastus: CRC Press Inc
  • ISBN-10: 1439869553
  • ISBN-13: 9781439869550
Teised raamatud teemal:
Image Restoration: Fundamentals and AdvancesSuurem pilt
  • Formaat: Hardback, 378 pages, kõrgus x laius: 254x178 mm, kaal: 907 g, 13 Tables, black and white; 45 Illustrations, color; 118 Illustrations, black and white
  • Sari: Digital Imaging and Computer Vision
  • Ilmumisaeg: 11-Sep-2012
  • Kirjastus: CRC Press Inc
  • ISBN-10: 1439869553
  • ISBN-13: 9781439869550
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9781439869550.html
Märksõnad:
Image Restoration: Fundamentals and Advances responds to the need to update most existing references on the subject, many of which were published decades ago. Providing a broad overview of image restoration, this book explores breakthroughs in related algorithm development and their role in supporting real-world applications associated with various scientific and engineering fields. These include astronomical imaging, photo editing, and medical imaging, to name just a few. The book examines how such advances can also lead to novel insights into the fundamental properties of image sources.



Addressing the many advances in imaging, computing, and communications technologies, this reference strikes just the right balance of coverage between core fundamental principles and the latest developments in this area. Its content was designed based on the idea that the reproducibility of published works on algorithms makes it easier for researchers to build on each others work, which often benefits the vitality of the technical community as a whole. For that reason, this book is as experimentally reproducible as possible.

Topics covered include:











Image denoising and deblurring Different image restoration methods and recent advances such as nonlocality and sparsity Blind restoration under space-varying blur Super-resolution restoration Learning-based methods Multi-spectral and color image restoration New possibilities using hybrid imaging systems

Many existing references are scattered throughout the literature, and there is a significant gap between the cutting edge in image restoration and what we can learn from standard image processing textbooks. To fill that need but avoid a rehash of the many fine existing books on this subject, this reference focuses on algorithms rather than theories or applications. Giving readers access to a large amount of downloadable source code, the book illustrates fundamental techniques, key ideas developed over the years, and the state of the art in image restoration. It is a valuable resource for readers at all levels of understanding.
Preface xi
Editors xiii
Contributors xv
1 Image Denoising: Past, Present, and Future
1(24)
Xin Li
1.1 Introduction
1(1)
1.2 Historical Review of Image Denoising
2(3)
1.3 First Episode: Local Wiener Filtering
5(3)
1.4 Second Episode: Understanding Transient Events
8(5)
1.4.1 Local Wiener Filtering in the Wavelet Space
8(1)
1.4.2 Wavelet vs. DCT Denoising
9(4)
1.5 Third Generation: Understanding Nonlocal Similarity
13(4)
1.6 Conclusions and Perspectives
17(8)
1.6.1 Representation versus Optimization
17(1)
1.6.2 Is Image Denoising Dead?
18(1)
Bibliography
18(7)
2 Fundamentals of Image Restoration
25(38)
Bahadir K. Gunturk
2.1 Introduction
25(1)
2.2 Linear Shift-Invariant Degradation Model
26(3)
2.3 Image Restoration Methods
29(17)
2.3.1 Least Squares Estimation
29(4)
2.3.2 Steepest Descent Approach
33(1)
2.3.3 Regularization Models
34(1)
2.3.4 Robust Estimation
35(1)
2.3.5 Regularization with lp Norm, O < p ≤ 1
36(3)
2.3.6 Wiener Filter
39(1)
2.3.7 Bayesian Approach
40(2)
2.3.8 Projection onto Convex Sets
42(2)
2.3.9 Learning-Based Image Restoration
44(2)
2.4 Blind Image Restoration
46(2)
2.4.1 Alternating Minimization
47(1)
2.4.2 Iterative Blind Deconvolution
48(1)
2.5 Other Methods of Image Restoration
48(2)
2.6 Super Resolution Image Restoration
50(1)
2.7 Regularization Parameter Estimation
51(1)
2.8 Beyond Linear Shift-Invariant Imaging Model
52(1)
2.9 Summary
53(10)
Bibliography
53(10)
3 Restoration in the Presence of Unknown Spatially Varying Blur
63(26)
Michal Sorel
Filip Sroubek
3.1 Introduction
63(1)
3.2 Blur models
64(11)
3.2.1 Camera Motion Blur
70(2)
3.2.2 Scene Motion Blur
72(1)
3.2.3 Defocus and Aberrations
73(2)
3.3 Space-Variant Super Resolution
75(9)
3.3.1 Algorithm
75(2)
3.3.2 Splitting
77(1)
3.3.3 PSF Estimation
78(2)
3.3.4 PSF Refinement
80(1)
3.3.5 Deconvolution and Super Resolution
80(3)
3.3.6 Experiments
83(1)
3.4 Summary
84(5)
Bibliography
84(5)
4 Image Denoising and Restoration Based on Nonlocal Means
89(26)
Peter van Beek
Yeping Su
Junlan Yang
4.1 Introduction
89(3)
4.2 Image Denoising Based on the Nonlocal Means
92(7)
4.2.1 NLM Filter
92(5)
4.2.2 Iterative NLM Denoising
97(2)
4.3 Image Deblurring Using Nonlocal Means Regularization
99(2)
4.3.1 Iterative Deblurring
99(1)
4.3.2 Iterative Deblurring with Nonlocal Means Regularization
100(1)
4.4 Recent Nonlocal and Sparse Modeling Methods
101(6)
4.5 Reducing Computational Cost of NLM-Based Methods
107(2)
4.6 Conclusions
109(6)
Bibliography
111(4)
5 Sparsity-Regularized Image Restoration: Locality and Convexity Revisited
115(26)
Weisheng Dong
Xin Li
5.1 Introduction
115(2)
5.2 Historical Review of Sparse Representations
117(1)
5.3 From Local to Nonlocal Sparse Representations
118(6)
5.3.1 Local Variations: Wavelets and Beyond
118(2)
5.3.2 Nonlocal Similarity: From Manifold Learning to Subspace Constraint Exploitation
120(4)
5.4 From Convex to Nonconvex Optimization Algorithms
124(3)
5.5 Reproducible Experimental Results
127(4)
5.5.1 Image Deblurring
127(1)
5.5.2 Super Resolution
128(1)
5.5.3 Compressed Sensing
129(2)
5.6 Conclusions and Connections
131(10)
Bibliography
133(8)
6 Resolution Enhancement Using Prior Information
141(34)
Hsin M. Shieh
Charles L. Byrne
Michael A. Fiddy
6.1 Introduction
141(2)
6.2 Fourier Transform Estimation and Minimum L2-Norm Solution
143(3)
6.2.1 Hilbert Space Reconstruction Methods
143(1)
6.2.2 Minimum L2-Norm Solutions
144(1)
6.2.3 Case of Fourier-Transform Data
144(1)
6.2.4 Case of Under-Determined Systems of Linear Equations
145(1)
6.3 Minimum Weighted L2-Norm Solution
146(11)
6.3.1 Class of Inner Products
147(1)
6.3.2 Minimum T-Norm Solutions
148(1)
6.3.3 Case of Fourier-Transform Data
148(1)
6.3.4 Case of p(x) = Xx (x)
149(1)
6.3.5 Regularization
150(3)
6.3.6 Multidimensional Problem
153(1)
6.3.7 Case of Radon-Transform Data: Tomographic Data
154(1)
6.3.8 Under-Determined Systems of Linear Equations
154(1)
6.3.9 Discrete PDFT
155(2)
6.4 Solution Sparsity and Data Sampling
157(4)
6.4.1 Compressed Sensing
157(1)
6.4.2 Sparse Solutions
158(1)
6.4.3 Why Sparseness?
158(2)
6.4.4 Tomographic Imaging
160(1)
6.4.5 Compressed Sampling
161(1)
6.5 Minimum L1-Norm and Minimum Weighted L1-Norm Solutions
161(3)
6.5.1 Minimum L1-Norm Solutions
161(1)
6.5.2 Why the One-Norm?
162(1)
6.5.3 Comparison with the PDFT
163(1)
6.5.4 Iterative Reweighting
163(1)
6.6 Modification with Nonuniform Weights
164(5)
6.6.1 Selection of Windows
164(1)
6.6.2 Multidimensional Case
165(1)
6.6.3 Challenge of the Modified PDFT for Realistic Applications
165(2)
6.6.4 Modified Strategy in the Choice of Weighted Windows
167(2)
6.7 Summary and Conclusions
169(6)
Bibliography
171(4)
7 Transform Domain-Based Learning for Super Resolution Restoration
175(42)
Prakash P. Gajjar
Manjunath V. Joshi
Kishor P. Upla
7.1 Introduction to Super Resolution
175(3)
7.1.1 Limitations of Imaging Systems
176(1)
7.1.2 Super Resolution Concept
176(1)
7.1.3 Super Resolution: Ill-Posed Inverse Problem
177(1)
7.2 Related Work
178(5)
7.2.1 Motion-Based Super Resolution
178(2)
7.2.2 Motion-Free Super Resolution
180(1)
7.2.3 Learning-Based Super Resolution
181(2)
7.3 Description of the Proposed Approach
183(7)
7.3.1 Image Acquisition Model
184(1)
7.3.2 Learning the Initial HR Estimation
185(1)
7.3.3 Degradation Estimation
185(1)
7.3.4 Image Field Model and MAP Estimation
186(4)
7.3.5 Applying the Algorithm to Color Images
190(1)
7.4 Transform Domain-Based Learning of the Initial HR Estimate
190(10)
7.4.1 Learning the Initial HR Estimate Using DWT
191(2)
7.4.2 Initial Estimate Using Discrete Cosine Transform
193(4)
7.4.3 Learning the Initial HR Estimate Using Contourlet Transform
197(3)
7.5 Experimental Results
200(7)
7.5.1 Construction of the Training Database
200(1)
7.5.2 Results on Gray-Scale Images
201(3)
7.5.3 Results on Color Images
204(3)
7.6 Conclusions and Future Research Work
207(10)
7.6.1 Conclusions
207(2)
7.6.2 Future Research Work
209(1)
Bibliography
210(7)
8 Super Resolution for Multispectral Image Classification
217(32)
Feng Li
Xiuping Jia
Donald Fraser
Andrew Lambert
8.1 Introduction
217(3)
8.2 Methodology
220(10)
8.2.1 Background
220(2)
8.2.2 Super Resolution Based on a Universal Hidden Markov Tree Model
222(6)
8.2.3 MAP-uHMT on Multispectral Images
228(2)
8.3 Experimental Results
230(15)
8.3.1 Testing with MODIS data
230(8)
8.3.2 Testing with ETM+ data
238(7)
8.4 Conclusion
245(4)
Bibliography
246(3)
9 Color Image Restoration Using Vector Filtering Operators
249(36)
Rastislav Lukac
9.1 Introduction
249(1)
9.2 Color Imaging Basics
250(8)
9.2.1 Numeral Representation
251(1)
9.2.2 Image Formation
252(1)
9.2.3 Noise Modeling
253(3)
9.2.4 Distance and Similarity Measures
256(2)
9.3 Color Space Conversions
258(4)
9.3.1 Standardized Representations
258(1)
9.3.2 Luminance-Chrominance Representations
259(1)
9.3.3 Cylindrical Representations
260(2)
9.3.4 Perceptual Representations
262(1)
9.4 Color Image Filtering
262(12)
9.4.1 Order-Statistic Methods
263(7)
9.4.2 Combination Methods
270(4)
9.5 Color Image Quality Evaluation
274(3)
9.5.1 Subjective Assessment
274(1)
9.5.2 Objective Assessment
275(2)
9.6 Conclusion
277(8)
Bibliography
277(8)
10 Document Image Restoration and Analysis as Separation of Mixtures of Patterns: From Linear to Nonlinear Models
285(26)
Anna Tonazzini
Ivan Gerace
Francesca Martinelli
10.1 Introduction
285(4)
10.1.1 Related Work
286(1)
10.1.2 Blind Source Separation Approach
287(1)
10.1.3
Chapter Outline
288(1)
10.2 Linear Instantaneous Data Model
289(7)
10.2.1 Single-Side Document Case
289(1)
10.2.2 Recto-Verso Document Case
290(1)
10.2.3 Solution through Independent Component Analysis
291(1)
10.2.4 Solution through Data Decorrelation
292(1)
10.2.5 Discussion of the Experimental Results
293(3)
10.3 Linear Convolutional Data Model
296(6)
10.3.1 Solution through Regularization
299(2)
10.3.2 Discussion of the Experimental Results
301(1)
10.4 Nonlinear Convolutional Data Model for the Recto-Verso Case
302(3)
10.4.1 Solution through Regularization
304(1)
10.4.2 Discussion of the Experimental Results
305(1)
10.5 Conclusions and Future Prospects
305(6)
Bibliography
307(4)
11 Correction of Spatially Varying Image and Video Motion Blur Using a Hybrid Camera
311(25)
Yu-Wing Tai
Michael S. Brown
11.1 Introduction
311(2)
11.2 Related Work
313(2)
11.2.1 Traditional Deblurring
313(1)
11.2.2 PSF Estimation and Priors
313(1)
11.2.3 Super Resolution and Upsampling
314(1)
11.3 Hybrid Camera System
315(4)
11.3.1 Camera Construction
316(1)
11.3.2 Blur Kernel Approximation Using Optical Flows
317(1)
11.3.3 Back-Projection Constraints
318(1)
11.4 Optimization Framework
319(6)
11.4.1 Richardson-Lucy Image Deconvolution
319(1)
11.4.2 Optimization for Global Kernels
320(1)
11.4.3 Spatially Varying Kernels
321(3)
11.4.4 Discussion
324(1)
11.5 Deblurring of Moving Objects
325(1)
11.6 Temporal Upsampling
326(2)
11.7 Results and Comparisons
328(7)
11.8 Conclusion
335(1)
Bibliography 336(5)
Index 341
Bahadir K. Gunturk received his B.S. degree from Bilkent University, Turkey, and his Ph.D. degree from the Georgia Institute of Technology in 1999 and 2003, respectively, both in electrical engineering. Since 2003, he has been with the Department of Electrical and Computer Engineering at Louisiana State University, where he is an associate professor. His research interests are in image processing and computer vision. Dr. Gunturk was a visiting scholar at the Air Force Research Lab in Dayton, Ohio, and at Columbia University in New York City. He is the recipient of the Outstanding Research Award at the Center of Signal and Image Processing at Georgia Tech in 2001, the Air Force Summer Faculty Fellowship Program (SFFP) Award in 2011 and 2012, and named as a Flagship Faculty at Louisiana State University in 2009.

Xin Li received his B.S. degree with highest honors in electronic engineering and information science from the University of Science and Technology of China, Hefei, in 1996, and his Ph.D. degree in electrical engineering from Princeton University, Princeton, New Jersey, in 2000. He was a member of the technical staff with Sharp Laboratories of America, Camas, Washington, from August 2000 to December 2002. Since January 2003, he has been a faculty member in the Lane Department of Computer Science and Electrical Engineering at West Virginia University. He is currently a tenured associate professor at that school. His research interests include image/video coding and processing. Dr. Li received a Best Student Paper Award at the Visual Communications and Image Processing Conference in 2001; a runner-up prize of Best Student Paper Award at the IEEE Asilomar Conference on Signals, Systems and Computers in 2006; and a Best Paper Award at the Visual Communications and Image Processing Conference in 2010.