Computer Vision - Linda Shapiro.pdf

所需积分/C币:50 2018-02-06 18:03:16 19.16MB PDF

Computer Vision - Linda Shapiro.pdf Computer Vision - Linda Shapiro.pdf Computer Vision - Linda Shapiro.pdf Computer Vision - Linda Shapiro.pdf Computer Vision - Linda Shapiro.pdf
Contents 1 Introductie 13 1.1 Machines that see? 14 2 Application problems 1. 3 Operations on Images 23 1.4 The the Bad, and the ugly 26 1.5 Use of Computers and Software 27 1.6R 28 1.7 The rest of the bool .28 1.8 References 29 1.9 Additional Exercises 30 2 Imaging and Image Representation 33 2.⊥ 2.2 Imaging Devices 34 2.3 x ProbleIns in Digilal Images ...39 2. 4 Picture Functions and Digital I 41 2.5 X Digital Image Formats ..46 2.6 Richness and Problems of Real Imagery 52 2.7 3D St om 2D Images 2.8 Five Frames of Reference 54 2.9 * Other Tvpes of Sensors 56 2.10 Ref 60 3 Binary Image Analysis 63 3.1 Pixels and Neighborhoods 63 3.2 Applying g 64 3.3 Counting the Objects in an Image 3.4 Connected Components Labeling 3.5 Binary lmage Morphology 3.5.1 Structuring elements 3.5.2 Basic Operat 3.5.3 Some Applications of Binary Morphology 3.5.4 Conditional Dilation 3.6 Region Properties 3.7 Region Ad 3.8 Thresholding Gray-Scale Images 97 专注人工智能 Computer vision: Mar 2000 3.8.1 The use of histograms for Threshold Selection 3.8.2 Automatic Thresholding: the Otsu Method 3.9 References .103 4 Pattern Recognition Concepts 107 4.1 Pattern Recognition Problems 107 4.2 Common model for classification ....109 4.3 Precision versuS reca. ll 112 4.4 Features used for representation 113 4.5 feature Vector Representation 114 4.6 Implementing the Classifier 116 4.7 Structural Techniques 119 4. 8 The Confusion Matrix 122 4.9 Decision trees 122 4.10 Bayesian decision-making 128 4.11 Decisions using Multidimensional Dat. .132 4.12 Machines that learn 135 4.13 Artificial Neural Nets 136 4.14 References 5 Filtering and Enhancing Images 145 5.1 What needs fixing? 5.1.1 An Image needs Improvement 146 5.1.2 Low-level features must be detected ..146 5.2 Grey level mapping 147 5.2.1 Histogram equalization 149 5. 3 Removal of Simlall linage Re .3.1R of Salt-and-Pepper 15I 5.3.2 Removal of Small Components 152 5.4 Iimage Smoothing 153 5.5 Median Filtering 154 5.5.1 Computing an Output Image from an Input Image .156 5.6 Detecting Edge Differencing masks 156 5.6.⊥Diff ID Signals 158 5.6.2 Diffe Operators for 2D I 160 5.7 Gaussian Filtering and LOG Edge Detection 166 5.7.1 Detecting Edges with the LOG Filter 170 5.7.2 On Human Edge-Detection 173 5. 7. 3 Marr-Hildreth Theory 174 5. 8 The Canny Edge Detector 176 5.9 *Masks as Matched Filters 176 5.9.1 The Vector Space of all Signals of n Samples 177 5.9.2 Using an Ort hogonal Basis 179 5.9.3 Cauchy-Schwartz inequality 181 5.9.4 The Vector Space of m x n images 181 5.9.5 A Roberts basis for 2 2 neighborhoods 81 5.9.6 The Frei-Chen basis for 33 neighborhoods 5.10 *Convolution and Cross Correlation 186 专注人工智能 Shapiro and stockman 5. 10.1 Defining operations via Masks 186 5.10.2 The Convolution Operation .188 5.10.3 Possible parallel implementations ..192 5.11 Analysis of spatial frequency using sinusoids ..192 5.11.1 A Fourier basis 193 5.11.2 2D Picture Functions .197 5.11.3 Discrete Fourier Transform 199 5.11.1 Bandpass Filtering 201 5.11.5 Discussion of the Fourier Transform 203 5.11. 6 *The Con volution Theorem 203 5. 12 Summary and discussion 205 5.13 References 205 6 Color and Shading 209 6.1 Some Physics of color 210 6.1.1 Sensing Illuminated Objects 211 6.1.2 Additional factors 211 6.1.3 Sensitivity of Receptors 212 6.2 The rgb basis for colo 213 6.3 Other Color bases 215 6.3.1 The CMY Subtractive Color System ...217 6.3.2 HSI: Hue-Saturation-Intensity ...217 6.3.3 YIQ and YUV for TV sig 219 6.3.4 Using Color for Classification 6.4 Color Histograms 221 6.5 Color Se 223 6.6 Shading .225 6.6.1 Radiation from One Light s 225 6.6.2 Diffuse Reflecti 226 6.6.3 Specular reflecti 6.6.4 Dark ith distance 6.6.5 Complications 230 6.6.6* Phong Model of Shading 231 6.6.7 Human Perception using Shading 231 6.7 x Related Topics 231 6.7.1Ap 231 6.7.2 Human Color P G. 7. 3 Multispectral linages 232 6.7.4 Thematic Imag 6.8 References 7 exture 255 7.1 Texture. Texels. and St atistics 7.2 Texel-Based Texture Descriptions 7.3 Quantitative Texture Measures 7.3.1 Edge densit y and directi 239 7.3.2 Local Binary Partition 210 7.3.3 Co-occurrence Matrices and Features 240 专注人工智能 Computer vision: Mar 2000 7.3.4 Laws'Text ure Energy Measures 243 7.3.5 Autocorrelation and Power Spectrum 211 7.4 Texture Segmentation 245 7.5 References 247 8 Content-Based Image Retrieval 249 8.1 Image Database Examples 249 8.2 Image Database Queries 251 8.3 Query-by-Example 252 8.4 Image Distance Measures 254 8.4.1 Color Similarity measures ...254 8. 1.2 Texture Similarity Measures 257 8.4.3 Shape Similarity Measures .259 8.4.4 Object Presence and relational similarit y Measures 264 8.5 Database Organization 8.5.1 Standard Indexes 8.5.2 Spatial Indexing ..271 8.5. 3 Indexing for Content-Based Image Retrieval with Multiple Distance Measures 271 8. 6 References 273 9 Motion from 2D Image Sequences 275 9. 1 Motion Phenomena and applications 9.2 Image Subtraction 27 9.3 Computing Motion Vectors 279 9.3.1 The Decathete Game 280 9.3.2 USing Point Correspondences .281 9.3.3 MPEG C Compression of video 285 9.3.4 Computing Image Flow ..287 9.3. The Image Flow Equation 288 9.3.6 Solving for Image Flow by Propagating Constraints 289 9.4 Computing the Paths of moving poin 9.4.1 Integrated Problem-Specific Trackin g 296 9.5 Detecting Significant Changes in Video 9.5.1 Segmenting Video Sequences 299 9.5.2 Ignoring Certain Camera Effects .301 9.5.3 Storing Video Subsequences 302 9.6 References ..303 10 Image Segmentation 305 10.1 Identifying ro 306 10.1.1 Clustering Methods 307 10.1.2 Region growing 315 10.2 Representing Regions 10.2.1 Overlays ...318 10.2.2 t led Image 10.2.3 Boundary coding 320 10.2. 4 Quad Trees 320 专注人工智能 Shapiro and stockman 10.2.5 Property Tables 322 10.3 Identifying Contours 10.3.1 Tracking Existing Region Boundaries ..323 10.3.2 The Canny Edge Detector and Linker 326 10.3. 3 Aggregating Consistent Neighboring Edgels into Curves 10.3.4 Hough Transform for Lines and Circular Arcs ..330 10.4 Fitting Models to Segments 341 10.5 Identifying Higher-level Structure 316 10.5.1 Ribbons 346 10.5.2 Detecting C corners 348 10.6 Segment ation using Motion Coherence 349 10.6.1 Boundaries in Space-TiIne ..350 10.6.2 Aggregrating Motion Trajectories .350 10.7 References ,,351 1 Matching in 2D 357 11.1 Registration of 2D Data 11.2R entation of point 359 11.3 Affine Mapping Functions 360 11.4* A Best 2D Affine Transformation 371 11.5 2D Object Recogllilion via Alline Mapping .372 11.6 2D Object Recognition via Relational Matching 11.7 Nonlinear Warping .398 Summary 402 11.g Referen ces 12 Perceiving 3D from 2D Images 405 12.1 Intrinsic lmages 405 12.2 Labeling of Line Drawings from Blocks World 410 12.3 3D Cues Available in 2D Images 417 12.1 Other phenomen a 422 12.4.1 Shape from X 422 12.4.2 Vanishing point 426 12.4.3 Depth from Focus 427 12.4.4 Motion phenormena ..428 12.4.5 Boundaries and virtual Lines .428 12.4.6 Alignments are Non-Accidenta 12.5 The Perspective Imaging model .429 12.6 Depth p 2.6.1 Establishing Correspondences 434 12. 7 The Thin Lens Equation 12.8 Concluding Discussion 441 12.9 References 442 专注人工智能 Computer vision: Mar 2000 13 3D Sensing and Object Pose Computation 445 13. 1 General Stereo Configuration 116 13.2 3D Affine Transformations .448 13.2.1 Coordinate Frames ...448 13.2.2 Translation 150 13.2.3 Scaling 13.2.4 Rotation 450 13.2.5 Arbitrary rotation 453 13.2.6 Alignment via Transformation Calculus .454 13.3 Camera model 454 13.3.1 Perspective Transformation Matrix 458 13.3.2 Orthographic and Weak perspective projections 461 13.3.3 Computing 3D Points Using Multiple Cameras 13.4 Best Affine Calibration Matrix .465 13.4. 1 Calibration jig ..466 13.4.2 Defining the Least-Squares Problem .466 13.4.3 Discussion of the affine method 471 13.5 Using structured Light 472 13.6 A Simple Pose estimation Procedure .474 13.7 An Improved Camera Calibration Method 479 13.7.1 Intrinsic Camera Parameters ..480 13013.7.3 Calibration Example ameters 13.7.2 Extrinsic Camera Pa 485 8 Pose Estimation 489 13.8.1 Pose from 2D-3D Point Correspondences 491 13.8.2 Constrained Linear Optimization 492 13.8. 3 Computing the Transformation Tr=R. TH ...,,,,...193 13.8.4 Verification and Optimization of Pose 495 13. 9 3D Object reconstruction 496 13.9.1 Data Ac 497 13.9.2 Registration of views ..499 13. 9. 3 Surface reconstruction ,..,500 C 500 13.10Computing Shape from Shading ..505 3.10.1 Photometric stereo 13.10. 2Integrating Spatial Constraints .......,..509 13.11Structure from Motion 511 13.12References 514 14 3D Models and Matching 517 14.1 Survey of Common Representation Methods 518 14.1.1 3D Mesh Model 14.1.2 Surface-Edge-Vertex Models 518 11.1.3 Generalized-Cylinder Models 521 14.1.4 Octrees ..524 14.1.5S dr 11.2 True 3D Models versus view-Class Models 526 14.3 Physics-based and Deformable Models .528 专注人工智能 Shapiro and stockman 14.3.1 Snakes: Active contour models 528 11.3.2 Balloon Models for 3D .531 14.4 3D Object Recognition Paradigm an Heart 14.3.3 Modeling Motion of the hu .532 14.4.1 Matching geometric Models via Alignment 36 14.4.2 Matching Relational models .542 14.4.3 Matching functional models 554 14.4. 4 Recognition by Appearance 560 14.5 Referen ces .567 15 Virtual Reality 571 15. 1 Features of Virtual Reality Systems ·. ·· 572 152 Applications ofⅤR 5. 3 Augmented Reality(AR) 15.4 Teleoperation 6 5.5 Virtual Reality Devices 15.6 Summary of Sensing Devices for VR 15.7 Rendering Simple 3D Models 587 15.8 Composing Real and Synthetic Imagery 589 15. 9 HCl and Psychological Issues 593 15.10Referen ces ..593 16 Case studies 595 16.1 Veggie Vision: A System for Checking out Vegetables 16.1.1 Application Domain and requirements 597 16.1.2 System Design 16.1. 3 Identification procedure 598 16.1.4 More Details on the process 598 16.1.5 Performance ..60 16.2 IdenlLilying Humans via the iris of anl Eye 602 16.2.I Requirements for identification systems 602 16.2.2 System Design 604 16.2.3 Performance .607 16.2. 4 References 609 专注人工智能 Computer vision: Mar 2000 Preface This book is intended as an introduction to computer vision for a broad audience. It provides necessary theory and examples for students and practicioners who will work in fields where significant, information must be ext racted aut omatically from images. The book should be a useful resource book for professionals, a text for both undergraduate and beginning graduate courses, and a resource for enrichment of college or even high school projects. Our goals were to provide a basic set of fundamental concepts and algorithms and also discuss some of the exciting evolving application areas. This book is unique in that it contains chapters on image databases(Ch 8)and on virtual and augmented reality(Ch 15 two exciting evolving application areas. A final chapter(Ch 16)gives a complete view of real world systems that use computer vision Due to recent progress in the computer field, economical and flexible use of computer im- ages is now pervasive. Computing with images is no longer just for the realm of the sciences but also for the arts and social sciences and even for hobbyists. The book should serve an established and growing a.dience including those interested in mu It imedia, art and design geographic information systems, and image databases, in addition to the traditional areas of automation. image science, medical imaging remote sensing and computer cartography a broad purpose at first seems impossible to achieve. However, there are other kinds of texts that already do this in other areas-calculus, physics, and general computing. We hope we have made at least a good beginning- we wanted a book that would be useful in the classroom and also to the independent reader. We find the chosen topics interesting and SOilletines exciting, anld hope that they are accessible lo a large audience. It is assuRed that use of the text in a graduate, or even senior level, computer vision course would be supplemented by papers from the archival literature. Coverage is not intended to be com prehensive; only a modest set of papers are cited at the end of each chapt The early chapters begin at an intuitive level and progress towards mathematical models with the goal of intuitive understanding before formal characterization. Sections marked by x, are more mathematical or more advanced and need not be covered in a less technical course. To strengthen the intuitive approach, we have stayed with the processing of iconic imagery for the first eleven chapters and have delayed 3d computer vision until the later chapters, but it should be easy for experienced instructors to resequence them to fit a partic ular course or teaching style. There are many viable applications that are entirely 2D, and Inlany concepls and algorithns are Inore simply laught in their 2D forIIl. We provide soine basics of pattern recognition in Chapter 4, so that students can consider complete recog- nition systems before the full coverage of image features and matching. A reader should have a good idea of 2D image processing applications after Chapter 4; Chapters 5, 6, and 7 add in gray-tone, color, and texture features. Chapter 8 treats image databases, a popular recent topic. Although some colleagues advised us to place this material near the end of the book, our goal of positioning it early in the chapter sequence is to reinforce the con cepts of Che prior chaplers and to provide inlalerial that can lead to all excellent halr-LerInl project Segmentation and matching are treated in their 2D forms in Chapters 10 and 11, so that the basic concepts are presented in a simple form, without introducing the complexities of 3D transformations 专注人工智能

...展开详情
试读 127P Computer Vision - Linda Shapiro.pdf
img
xpanxcom

关注 私信 TA的资源

上传资源赚积分,得勋章
    最新推荐
    Computer Vision - Linda Shapiro.pdf 50积分/C币 立即下载
    1/127
    Computer Vision - Linda Shapiro.pdf第1页
    Computer Vision - Linda Shapiro.pdf第2页
    Computer Vision - Linda Shapiro.pdf第3页
    Computer Vision - Linda Shapiro.pdf第4页
    Computer Vision - Linda Shapiro.pdf第5页
    Computer Vision - Linda Shapiro.pdf第6页
    Computer Vision - Linda Shapiro.pdf第7页
    Computer Vision - Linda Shapiro.pdf第8页
    Computer Vision - Linda Shapiro.pdf第9页
    Computer Vision - Linda Shapiro.pdf第10页
    Computer Vision - Linda Shapiro.pdf第11页
    Computer Vision - Linda Shapiro.pdf第12页
    Computer Vision - Linda Shapiro.pdf第13页
    Computer Vision - Linda Shapiro.pdf第14页
    Computer Vision - Linda Shapiro.pdf第15页
    Computer Vision - Linda Shapiro.pdf第16页
    Computer Vision - Linda Shapiro.pdf第17页
    Computer Vision - Linda Shapiro.pdf第18页
    Computer Vision - Linda Shapiro.pdf第19页
    Computer Vision - Linda Shapiro.pdf第20页

    试读已结束,剩余107页未读...

    50积分/C币 立即下载 >