Multidimensional Signal, Image, and Video Processing and Coding

Digital images have become mainstream of late notably within HDTV, cell phones, personal cameras, and many medical applications. The processing of digital images and video includes adjusting illumination, manufacturing enlargements/reductions, and creating contrast. This development has made it possible to take long forgotten, badly damaged photos and make them new again with image estimation. It can also help snapshot photographers with image restoration, a method of reducing the influence of an unsteady hand. Dr. Woods has constructed a book for professionals and graduate students that will give them the thorough understanding of image and video processing that they need in order to contribute to this hot technology's future advances. Examples and problems at the end of each chapter help the reader digest what has just been read. Forged from a theoretical base, this exceptional book develops into an essential guide to hands-on endeavors in signal processing. FOR INSTRUCTORS: To obtain access to the solutions manual for this title simply register on our textbook website (textbooks.elsevier.com)and request access to the Computer Science or Electronics and Electrical Engineering subject area. Once approved (usually within one business day) you will be able to access all of the instructor-only materials through the'Instructor Manual'link on this book's academic web page at textbooks.elsevier.com.•Overflowing with over 150 digital images •Brimming with productive examples and challenging problems •Written by celebrated MIT graduate who has authored four other exceptional books Front cover Title page Date-line Contents Preface 1 Two-Dimensional Signals and Systems 1.1 Two-Dimensional Signals 1.1.1 Separable Signals 1.1.2 Periodic signals 1.1.3 2-D Discrete-Space Systems 1.1.4 Two-Dimensional Convolution 1.1.5 Stability of 2-D Systems 1.2 2-D Discrete-Space Fourier Transform 1.2.1 Inverse 2-D Fourier Transform 1.2.2 Fourier Transform of 2-D or Spatial Convolution 1.2.3 Symmetry Properties of Fourier Transform 1.2.4 Continuous-Space Fourier Transform 1.3 Conclusions 1.4 Problems References 2 Sampling in Two Dimensions 2.1 Sampling Theorem—Rectangular Case 2.1.1 Reconstruction Formula 2.1.2 Ideal Rectangular Sampling 2.2 Sampling Theorem—General Regular Case 2.2.1 Hexagonal Reconstruction Formula 2.3 Change of Sample Rate 2.3.1 Downsampling by Integers $M_1 \times M_2$ 2.3.2 Ideal Decimation 2.3.3 Upsampling by Integers $L_1 \times L_2$ 2.3.4 Ideal Interpolation 2.4 Sample-Rate Change—General Case 2.4.1 General Downsampling 2.5 Conclusions 2.6 Problems References 3 Two-Dimensional Systems and Z-Transforms 3.1 Linear Spatial or 2-D Systems 3.2 Z-Transforms 3.3 Regions of Convergence 3.3.1 More General Case 3.4 Some Z-Transform Properties 3.4.1 Linear Mapping of Variables 3.4.2 Inverse Z-Transform 3.5 2-D Filter Stability 3.5.1 First-Quadrant Support 3.5.2 Second-Quadrant Support 3.5.3 Root Maps 3.5.4 Stability Criteria for NSHP Support Filters 3.6 Conclusions 3.7 Problems References 4 Two-Dimensional Discrete Transforms 4.1 Discrete Fourier Series 4.1.1 Properties of the DFS Transform 4.1.2 Periodic Convolution 4.1.3 Shifting or Delay Property 4.2 Discrete Fourier Transform 4.2.1 DFT Properties 4.2.2 Relation of DFT to Fourier Transform 4.2.3 Effect of Sampling in Frequency 4.2.4 Interpolating the DFT 4.3 2-D Discrete Cosine Transform 4.3.1 Review of 1-D DCT 4.3.2 Some 1-D DCT Properties 4.3.3 Symmetric Extension in 2-D DCT 4.4 Subband/Wavelet Transform (SWT) 4.4.1 Ideal Filter Case 4.4.2 1-D SWT with Finite-Order Filter 4.4.3 2-D SWT with FIR Filters 4.4.4 Relation of SWT to DCT 4.4.5 Relation of SWT to Wavelets 4.5 Fast Transform Algorithms 4.5.1 Fast DFT Algorithm 4.5.2 Fast DCT Methods 4.6 Sectioned Convolution Methods 4.7 Conclusions 4.8 Problems References 5 Two-Dimensional Filter Design 5.1 FIR Filter Design 5.1.1 FIR Window Function Design 5.1.2 Design by Transformation of 1-D Filter 5.1.3 Projection-Onto-Convex-Sets Method 5.2 IIR Filter Design 5.2.1 2-D Recursive Filter Design 5.2.2 Fully Recursive Filter Design 5.3 Subband/Wavelet Filter Design 5.3.1 Wavelet (Biorthogonal) Filter Design Method 5.4 Conclusions 5.5 Problems References 6 Introductory Image Processing 6.1 Light and Luminance 6.2 Still Image Visual Properties 6.2.1 Weber's Law 6.2.2 Contrast Sensitivity Function 6.2.3 Local Contrast Adaptation 6.3 Time-Variant Human Visual System Properties 6.4 Image Sensors 6.4.1 Electronic 6.4.2 Film 6.5 Image and Video Display 6.5.1 Gamma 6.6 Simple Image Processing Filters 6.6.1 Box Filter 6.6.2 Gaussian Filter 6.6.3 Prewitt Operator 6.6.4 Sobel Operator 6.6.5 Laplacian Filter 6.7 Conclusions 6.8 Problems References 7 Image Estimation and Restoration 7.1 2-D Random Fields 7.1.1 Filtering a 2-D Random Field 7.1.2 Autoregressive Random Signal Models 7.2 Estimation for Random Fields 7.2.1 Infinite Observation Domain 7.3 2-D Recursive Estimation 7.3.1 1-D Kalman Filter 7.3.2 2-D Kalman Filtering 7.3.3 Reduced Update Kalman Filter 7.3.4 Approximate RUKF 7.3.5 Steady-State RUKF 7.3.6 LSI Estimation and Restoration Examples with RUKF 7.4 Inhomogeneous Gaussian Estimation 7.4.1 Inhomogeneous Estimation with RUKF 7.5 Estimation in the Subband/Wavelet Domain 7.6 Bayesian and MAP Estimation 7.6.1 Gauss Markov Image Models 7.6.2 Simulated Annealing 7.7 Image Identification and Restoration 7.7.1 Expectation-Maximization Algorithm Approach 7.7.2 EM Method in the Subband/Wavelet Domain 7.8 Color Image Processing 7.9 Conclusions 7.10 Problems References 8 Digital Image Compression 8.1 Introduction 8.2 Transformation 8.2.1 DCT 8.2.2 SWT 8.2.3 DPCM 8.3 Quantization 8.3.1 Uniform Quantization 8.3.2 Optimal MSE Quantization 8.3.3 Vector Quantization 8.3.4 LBG Algorithm [7] 8.4 Entropy Coding 8.4.1 Huffman Coding 8.4.2 Arithmetic Coding 8.4.3 ECSQ and ECVQ 8.5 DCT Coder 8.6 SWT Coder 8.6.1 Multiresolution SWT Coding 8.6.2 Nondyadic SWT Decompositions 8.6.3 Fully Embedded SWT Coders 8.6.4 Embedded Zero-Tree Wavelet (EZW) Coder 8.6.5 Set Partitioning in Hierarchical Trees (SPIHT) Coder 8.6.6 Embedded Zero Block Coder (EZBC) 8.7 JPEG 2000 8.8 Color Image Coding 8.8.1 Scalable Coder Results Comparison 8.9 Robustness Considerations 8.10 Conclusions 8.11 Problems References 9 Three-Dimensional and Spatiotemporal Processing 9.1 3-D Signals and Systems 9.1.1 Properties of 3-D Fourier Transform 9.1.2 3-D Filters 9.2 3-D Sampling and Reconstruction 9.2.1 General 3-D Sampling 9.3 Spatiotemporal Signal Processing 9.3.1 Spatiotemporal Sampling 9.3.2 Spatiotemporal Filters 9.3.3 Intraframe Filtering 9.3.4 Intraframe Wiener Filter 9.3.5 Interframe Filtering 9.3.6 Interframe Wiener Filter 9.4 Spatiotemporal Markov Models 9.4.1 Causal and Semicausal 3-D Field Sequences 9.4.2 Reduced Update Spatiotemporal Kalman Filter 9.5 Conclusions 9.6 Problems References 10 Digital Video Processing 10.1 Interframe Processing 10.2 Motion Estimation and Motion Compensation 10.2.1 Block Matching Method 10.2.2 Hierarchical Block Matching 10.2.3 Overlapped Block Motion Compensation 10.2.4 Pel-Recursive Motion Estimation 10.2.5 Optical flow methods 10.3 Motion-Compensated Filtering 10.3.1 MC-Wiener Filter 10.3.2 MC-Kalman Filter 10.3.3 Frame-Rate Conversion 10.3.4 Deinterlacing 10.4 Bayesian Method for Estimating Motion 10.4.1 Joint Motion Estimation and Segmentation 10.5 Conclusions 10.6 Problems References 10.7 Appendix: Digital Video Formats SIF CIF ITU 601 Digital TV (aka SMPTE D1 and D5) ATSC Formats 11 Digital Video Compression 11.1 Intraframe Coding 11.1.1 M-JPEG Pseudo Algorithm 11.1.2 DV Codec 11.1.3 Intraframe SWT Coding 11.1.4 M-JPEG 2000 11.2 Interframe Coding 11.2.1 Generalizing 1-D DPCM to Interframe Coding 11.2.2 MC Spatiotemporal Prediction 11.3 Interframe Coding Standards 11.3.1 MPEG1 11.3.2 MPEG 2—"a Generic Standard" 11.3.3 The Missing MPEG 3—High-Definition Television 11.3.4 MPEG 4—Natural and Synthetic Combined 11.3.5 Video Processing of MPEG-Coded Bitstreams 11.3.6 H.263 Coder for Visual Conferencing 11.3.7 H.264/AVC 11.3.8 Video Coder Mode Control 11.3.9 Network Adaptation 11.4 Interframe SWT Coders 11.4.1 Motion-Compensated SWT Hybrid Coding 11.4.2 3-D or Spatiotemporal Transform Coding 11.5 Scalable Video Coders 11.5.1 MoreonMCTF 11.5.2 Detection of Covered Pixels 11.5.3 Bidirectional MCTF 11.6 Object-Based Video Coding 11.7 Comments on the Sensitivity of Compressed Video 11.8 Conclusions 11.9 Problems References 12 Video Transmission over Networks 12.1 Video on IP Networks 12.1.1 Overview of IP Networks 12.1.2 Error-Resilient Coding 12.1.3 Transport-Level Error Control 12.1.4 Wireless Networks 12.1.5 Joint Source-Channel Coding 12.1.6 Error Concealment 12.2 Robust SWT Video Coding (Bajic) 12.2.1 Dispersive Packetization 12.2.2 Multiple Description FEC 12.3 Error-Resilience Features of H.264/AVC 12.3.1 Syntax 12.3.2 Data Partitioning 12.3.3 Slice Interleaving and Flexible Macroblock Ordering 12.3.4 Switching Frames 12.3.5 Reference Frame Selection 12.3.6 Intrablock Refreshing 12.3.7 Error Concealment in H.264/AVC 12.4 Joint Source-Network Coding 12.4.1 Digital Item Adaptation (DIA) in MPEG 21 12.4.2 Fine-Grain Adaptive FEC 12.5 Conclusions 12.6 Problems References Index Separate plates (color in original) Back cover Digital images and video are everywhere, notably within television, Internet, cell phones, personal cameras, and many medical and industrial applications. The manipulation and processing of these data, image, and video elements is vital to the advancement of these and many other wired and wireless communication and entertainment devices. The science of compression, filter design, and restoration needed to improve current applications and design "the next big thing" are just some of the many topics discussed in this book.Dr. Woods has delivered an exceptional book for professionals and graduate students that will give them the thorough understanding of signal, image, and video processing that they require in order to contribute to the future of this omnipresent technology. This reference is ideal as a comprehensive self-teaching tool or as a text for a graduate level multimedia signal processing course. Divided into two sections, the first covers multidimensional signal processing theory and the second delves into image and video processing and coding applications. Examples and problems at the end of each chapter help the reader continue the learning process. Digital images have become mainstream of late notably within HDTV, cell phones, personal cameras, and many medical applications. This book offers an understanding of image and video processing. It includes over 150 digital images and productive examples and challenging problems