Page 2
Related Titles Florczyk, S. Pratt, W. K. Robot Vision Digital Image Processing Video-based Indoor Exploration PIKS Inside with Autonomous 756 pages and Mobile Robots 2001 216 pages with 89 figures and 3 tables Hardcover 2005 ISBN 0-471-37407-5 Hardcover Nikolaidis, N., Pitas, I. ISBN 3-527-40544-5 3-D Image Processing Mather, P. M. Algorithms Computer Processing 190 pages of Remotely-Sensed Images 2001 An Introduction Hardcover ISBN 0-471-37736-8 442 pages 2004 Pitas, I. Softcover Digital Image Processing ISBN 0-470-84919-3 Algorithms and Applications Bose, T., Meyer, F. 432 pages Digital Signal 2000 and Image Processing Hardcover ISBN 0-471-37739-2 International Edition Vincze, M., Hager, G.D. (eds.) approx. 656 pages 2003 Robust Vision for Vision-Based Hardcover Control of Motion ISBN 0-471-45230-0 approx. 262 pages Mann, S. 2000 Hardcover Intelligent Image Processing ISBN 0-7803-5378-1 360 pages Petrou, M., Bosdogianni, P. 2002 Hardcover Image Processing ISBN 0-471-40637-6 The Fundamentals 354 pages 1999 Hardcover ISBN 0-471-99883-4
Page 4
The Editor All books published by Wiley-VCH are carefully produced. Nevertheless, editors, authors and Prof. Dr. Alexander Hornberg publisher do not warrant the information contained Hochschule Esslingen in these books to be free of errors. Readers are FB Mechatronik u. Eletrotechnik advised to keep in mind that statements, data, alexander.hornberg@fht-esslingen.de illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for Cover Individual controllable stack of red and IR British Library Cataloguing-in-Publication ring lights for presence check of circuits and Data: quality check of soldering points. A catalogue record for this book is available from ©c Vision & Control GmbH, Suhl/Germany the British Library. Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at http://dnb.ddb.de ©c 2006 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photocopying, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Printed in the Federal Republic of Germany Printed on acid-free and chlorine-free paper Cover Design: aktivcomm, Weinheim Typesetting: Steingraeber Satztechnik GmbH, Ladenburg Printing: betz-druck GmbH, Darmstadt Bookbinding: Litges & Dopf Buchbinderei GmbH, Heppenheim ISBN-13: 978-3-527-40584-8 ISBN-10: 3-527-40584-4
Page 5
V Contents 1 Processing of Information in the Human Visual System 1 Prof. Dr. F. Schaeffel, University of Tübingen 1.1 Preface 1 1.2 Design and Structure of the Eye 1 1.3 Optical Aberrations and Consequences for Visual Performance 5 1.4 Chromatic Aberration 9 1.5 Neural Adaptation to Monochromatic Aberrations 12 1.6 Optimizing Retinal Processing with Limited Cell Numbers, Space and Energy 13 1.7 Adaptation to Different Light Levels 13 1.8 Rod and Cone Responses 15 1.9 Spiking and Coding 18 1.10 Temporal and Spatial Performance 19 1.11 ON/OFF Structure, Division of the Whole Illuminance Amplitude in Two Segments 20 1.12 Consequences of the Rod and Cone Diversity on Retinal Wiring 21 1.13 Motion Sensitivity in the Retina 21 1.14 Visual Information Processing in Higher Centers 23 1.14.1 Morphology 23 1.14.2 Functional Aspects – Receptive Field Structures and Cortical Modules 24 1.15 Effects of Attention 26 1.16 Color Vision, Color Constancy, and Color Contrast 27 1.17 Depth Perception 28 1.18 Adaptation in the Visual System to Color, Spatial, and Temporal Contrast 29 1.19 Conclusions 30 References 32 Handbook of Machine Vision. Alexander Hornberg (Ed.) Copyright c© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-40584-7
Page 6
VI Contents 2 Introduction to Building a Machine Vision Inspection 35 Axel Telljohann, Consulting Team Machine Vision (CTMV) 2.1 Preface 35 2.2 Specifying a Machine Vision System 36 2.2.1 Task and Benefit 37 2.2.2 Parts 37 2.2.2.1 Different Part Types 38 2.2.3 Part Presentation 38 2.2.4 Performance Requirements 39 2.2.4.1 Accuracy 39 2.2.4.2 Time Performance 39 2.2.5 Information Interfaces 39 2.2.6 Installation Space 40 2.2.7 Environment 41 2.2.8 Checklist 41 2.3 Designing a Machine Vision System 41 2.3.1 Camera Type 41 2.3.2 Field of View 42 2.3.3 Resolution 43 2.3.3.1 Camera Sensor Resolution 44 2.3.3.2 Spatial Resolution 44 2.3.3.3 Measurement Accuracy 44 2.3.3.4 Calculation of Resolution 45 2.3.3.5 Resolution for a Line Scan Camera 45 2.3.4 Choice of Camera, Frame Grabber, and Hardware Platform 46 2.3.4.1 Camera Model 46 2.3.4.2 Frame Grabber 46 2.3.4.3 Pixel Rate 46 2.3.4.4 Hardware Platform 47 2.3.5 Lens Design 48 2.3.5.1 Focal Length 48 2.3.5.2 Lens Flange Focal Distance 49 2.3.5.3 Extension Tubes 49 2.3.5.4 Lens Diameter and Sensor Size 50 2.3.5.5 Sensor Resolution and Lens Quality 50 2.3.6 Choice of Illumination 50 2.3.6.1 Concept: Maximize Contrast 51 2.3.6.2 Illumination Setups 51 2.3.6.3 Light Sources 52 2.3.6.4 Approach to the Optimum Setup 52 2.3.6.5 Interfering Lighting 53 2.3.7 Mechanical Design 53
Page 7
Contents VII 2.3.8 Electrical Design 53 2.3.9 Software 54 2.3.9.1 Software Library 54 2.3.9.2 Software Structure 54 2.3.9.3 General Topics 55 2.4 Costs 56 2.5 Words on Project Realization 57 2.5.1 Development and Installation 57 2.5.2 Test Run and Acceptance Test 58 2.5.3 Training and Documentation 58 2.6 Examples 58 2.6.1 Diameter Inspection of Rivets 59 2.6.1.1 Task 59 2.6.1.2 Specification 59 2.6.1.3 Design 60 2.6.2 Tubing Inspection 65 2.6.2.1 Task 65 2.6.2.2 Specification 65 2.6.2.3 Design 66 3 Lighting in Machine Vision 73 I. Jahr, Vision & Control GmbH 3.1 Introduction 73 3.1.1 Prologue 73 3.1.2 The Involvement of Lighting in the Complex Machine Vision Solution 75 3.2 Demands on Machine Vision lighting 78 3.3 Light used in Machine Vision 82 3.3.1 What is Light? Axioms of Light 82 3.3.2 Light and Light Perception 84 3.3.3 Light Sources for Machine Vision 88 3.3.3.1 Incandescent Lamps / Halogen Lamps 89 3.3.3.2 Metal Vapor Lamps 91 3.3.3.3 Xenon Lamps 92 3.3.3.4 Fluorescent Lamps 93 3.3.3.5 LEDs (Light Emitting Diodes) 95 3.3.3.6 Lasers 99 3.3.4 The Light Sources in Comparison 100 3.3.5 Considerations for Light Sources: Lifetime, Aging, Drift 100 3.3.5.1 Lifetime 100 3.3.5.2 Aging and Drift 102
Page 8
VIII Contents 3.4 Interaction of Test Object and Light 105 3.4.1 Risk Factor Test Object 105 3.4.1.1 What Does the Test Object do With the Incoming Light? 106 3.4.1.2 Reflection/reflectance/scattering 107 3.4.1.3 Total Reflection 111 3.4.1.4 Transmission/transmittance 112 3.4.1.5 Absorption/absorbance 113 3.4.1.6 Diffraction 114 3.4.1.7 Refraction 115 3.4.2 Light Color and Part Color 116 3.4.2.1 Visible Light (VIS) – Monochromatic Light 116 3.4.2.2 Visible Light (VIS) – White Light 118 3.4.2.3 Infrared Light (IR) 119 3.4.2.4 Ultraviolet Light (UV) 121 3.4.2.5 Polarized Light 122 3.5 Basic Rules and Laws of Light Distribution 125 3.5.1 Basic Physical Quantities of Light 125 3.5.2 The Photometric Inverse Square Law 127 3.5.3 The Constancy of Luminance 128 3.5.4 What Light Arrives at the Sensor – Light Transmission Through the Lens 129 3.5.5 Light Distribution of Lighting Components 131 3.5.6 Contrast 134 3.5.7 Exposure 136 3.6 Light Filters 138 3.6.1 Characteristic Values of Light Filters 138 3.6.2 Influences of Light Filters on the Optical Path 140 3.6.3 Types of Light Filters 141 3.6.4 Anti-Reflective Coatings (AR) 143 3.6.5 Light Filters for Machine Vision 144 3.6.5.1 UV Blocking Filter 144 3.6.5.2 Daylight Suppression Filter 144 3.6.5.3 IR Suppression Filter 146 3.6.5.4 Neutral Filter/Neutral Density Filter/Gray Filter 146 3.6.5.5 Polarization Filter 147 3.6.5.6 Color Filters 148 3.6.5.7 Filter Combinations 149 3.7 Lighting Techniques and Their Use 150 3.7.1 How to Find a Suitable Lighting? 150 3.7.2 Planning the Lighting Solution – Influence Factors 151 3.7.3 Lighting Systematics 154 3.7.3.1 Directional Properties of the Light 155
Page 9
Contents IX 3.7.3.2 Arrangement of the Lighting 157 3.7.3.3 Properties of the Illuminated Field 158 3.7.4 The Lighting Techniques in Detail 160 3.7.4.1 Diffuse Bright Field Incident Light 160 3.7.4.2 Directed Bright Field Incident Light 162 3.7.4.3 Telecentric Bright Field Incident Light 163 3.7.4.4 Structured Bright Field Incident Light 165 3.7.4.5 Diffuse/Directed Partial Bright Field Incident Light 169 3.7.4.6 Diffuse/Directed Dark Field Incident Light 173 3.7.4.7 The Limits of the Incident Lighting 176 3.7.4.8 Diffuse Bright Field Transmitted Lighting 177 3.7.4.9 Directed Bright Field Transmitted Lighting 178 3.7.4.10 Telecentric Bright Field Transmitted Lighting 180 3.7.4.11 Diffuse/Directed Transmitted Dark Field Lighting 184 3.7.5 Combined Lighting Techniques 184 3.8 Lighting Control 186 3.8.1 Reasons for Light Control – the Environmental Industrial Conditions 186 3.8.2 Electrical Control 186 3.8.2.1 Stable Operation 186 3.8.2.2 Brightness Control 189 3.8.2.3 Temporal Control: Static-Pulse-Flash 189 3.8.2.4 Some Considerations for the Use of Flash Light 191 3.8.2.5 Temporal and Local Control: Adaptive Lighting 195 3.8.3 Geometrical Control 198 3.8.3.1 Lighting from Large Distances 198 3.8.3.2 Light Deflection 199 3.8.4 Suppression of Ambient and Extraneous Light – Measures for a Stable Lighting 200 3.9 Lighting Perspectives for the Future 201 References 202 4 Optical Systems in Machine Vision 205 Dr. Karl Lenhardt, Jos. Schneider Optische Werke GmbH 4.1 A Look on the Foundations of Geometrical Optics 205 4.1.1 From Electrodynamics to the Light Rays 205 4.1.2 The Basic Laws of Geometrical Optics 208 4.2 Gaussian Optics 210 4.2.1 Reflection and Refraction at the Boundary between two Media 210 4.2.2 Linearizing the Law of Refraction – the Paraxial Approximation 212
Page 10
X Contents 4.2.3 Basic Optical Conventions 213 4.2.4 The Cardinal Elements of a Lens in Gaussian Optics 216 4.2.5 The Thin Lens Approximation 221 4.2.6 Beam Converging and Beam Diverging Lenses 222 4.2.7 Graphical Image Constructions 223 4.2.7.1 Beam Converging Lenses 223 4.2.7.2 Beam Diverging Lenses 224 4.2.8 Imaging Equations and Their Related Coordinate Systems 224 4.2.8.1 Reciprocity Equation 225 4.2.8.2 Newton’s Equations 226 4.2.8.3 General Imaging Equation 227 4.2.8.4 The Axial Magnification Ratio 229 4.2.9 The Overlapping of Object and Image Space 229 4.2.10 Focal Length, Lateral Magnification and the Field of View 229 4.2.11 Systems of Lenses 231 4.2.12 Consequences of the Finite Extension of Ray Pencils 234 4.2.12.1 Effects of Limitations of the Ray Pencils 234 4.2.12.2 Several Limiting Openings 237 4.2.12.3 Characterizing the Limits of Ray Pencils 240 4.2.12.4 The Relation to the Linear Camera Model 243 4.2.13 Geometrical Depth of Field and Depth of Focus 244 4.2.13.1 Depth of Field as a Function of the Object Distance p 246 4.2.13.2 Depth of Field as a Function of β 247 4.2.13.3 The Hyperfocal Distance 248 4.2.13.4 The Permissible Size for the Circle of Confusion d′ 249 4.2.14 The Laws of Central Projection–Telecentric System 251 4.2.14.1 An Introduction to the Laws of Perspective 251 4.2.14.2 Central Projection from Infinity – Telecentric Perspective 260 4.3 The Wave Nature of Light 268 4.3.1 Introduction 268 4.3.2 The Rayleigh–Sommerfeld Diffraction Integral 269 4.3.3 Further Approximations to the Huygens–Fresnel Principle 272 4.3.3.1 Fresnel’s Approximation 273 4.3.4 The Impulse Response of an Aberration Free Optical System 275 4.3.4.1 The Case of Circular Aperture, Object Point on the Optical Axis 278 4.3.5 The Intensity Distribution in the Neighbourhood of the Geometrical Focus 278 4.3.6 The Extension of the Point Spread Function in a Defocused Image Plane 282 4.3.7 Consequences for the Depth of Field Considerations 285 4.4 Information Theoretical Treatment of Image Transfer and Storage 287
Page 11
Contents XI 4.4.1 Physical Systems as Linear, Invariant Filters 288 4.4.2 The Optical Transfer Function (OTF) and the meaning of spatial frequency 296 4.4.3 Extension to the Two-Dimensional Case 298 4.4.3.1 The Interpretation of Spatial Frequency Components (r, s) 298 4.4.3.2 Reduction to One-Dimensional Representations 300 4.4.4 Impulse Response and MTF for Semiconductor Imaging Devices 302 4.4.5 The Transmission Chain 304 4.4.6 The Aliasing Effect and the Space Variant Nature of Aliasing 305 4.4.6.1 The Space Variant Nature of Aliasing 312 4.5 Criteria for Image Quality 316 4.5.1 Gaussian Data 316 4.5.2 Overview on Aberrations of Third Order 316 4.5.3 Image Quality in the Space Domain: PSF, LSF, ESF and Distortion 317 4.5.4 Image Quality in Spatial Frequency Domain: MTF 321 4.5.5 Other Image Quality Parameters 323 4.5.5.1 Relative Illumination (Relative Irradiance ) [17] 323 4.5.5.2 Deviation from Telecentricity (for Telecentric Lenses only) 325 4.5.6 Manufacturing Tolerances and Image Quality 325 4.6 Practical Aspects 326 References 331 5 Camera Calibration 333 R. Godding, AICON 3D Systems GmbH 5.1 Introduction 333 5.2 Terminology 334 5.2.1 Camera, Camera system 334 5.2.2 Coordinate systems 335 5.2.3 Interior Orientation and Calibration 335 5.2.4 Exterior and Relative Orientation 335 5.2.5 System Calibration 336 5.3 Physical Effects 336 5.3.1 Optical System 336 5.3.2 Camera and Sensor Stability 336 5.3.3 Signal Processing and Transfer 337 5.4 Mathematical Calibration Model 338 5.4.1 Central Projection 338 5.4.2 Camera Model 339 5.4.3 Focal Length and Principal Point 340
Page 12
XII Contents 5.4.4 Distortion and Affinity 340 5.4.5 Radial symmetrical distortion 341 5.4.6 Radial Asymmetrical and Tangential Distortion 342 5.4.7 Affinity and Nonorthogonality 343 5.4.8 Variant Camera Parameters 343 5.4.9 Sensor Flatness 345 5.4.10 Other Parameters 345 5.5 Calibration and Orientation Techniques 346 5.5.1 In the Laboratory 346 5.5.2 Using Bundle Adjustment to Determine Camera Parameters 346 5.5.2.1 Calibration based Exclusively on Image Information 347 5.5.2.2 Calibration and Orientation with Additional Object Information 349 5.5.2.3 Extended System Calibration 351 5.5.3 Other Techniques 352 5.6 Verification of Calibration Results 354 5.7 Applications 355 5.7.1 Applications with Simultaneous Calibration 355 5.7.2 Applications with precalibrated cameras 356 5.7.2.1 Tube Measurement within a Measurement Cell 356 5.7.2.2 Online Measurements in the Field of Car Safety 357 5.7.2.3 Other Applications 358 References 358 6 Camera Systems in Machine Vision 361 Horst Mattfeldt, Allied Vision Technologies GmbH 6.1 Camera Technology 361 6.1.1 History in Brief 361 6.1.2 Machine Vision versus Closed Circuit Television (CCTV) 362 6.2 Sensor Technologies 363 6.2.1 Spatial Differentiation: 1D and 2D 364 6.2.2 CCD Technology 364 6.2.3 Full Frame Principle 365 6.2.4 Frame Transfer Principle 366 6.2.5 Interline Transfer 366 6.2.5.1 Interlaced Scan Interline Transfer 367 6.2.5.2 Frame Readout 369 6.2.5.3 Progressive Scan Interline Transfer 370 6.3 CCD Image Artifacts 372 6.3.1 Blooming 372 6.3.2 Smear 372
Page 13
Contents XIII 6.4 CMOS Image Sensor 373 6.4.1 Advantages of CMOS Sensors 374 6.4.2 CMOS Sensor Shutter Concepts 376 6.4.2.1 Comparison of CMOS versus CCD 378 6.4.2.2 Integration Complexity of CDD versus CMOS Camera Technology 379 6.4.2.3 Video Standards 379 6.4.2.4 Sensor Sizes and Dimensions 380 6.4.2.5 Sony HAD Technology 381 6.4.2.6 Sony SuperHAD Technology 381 6.4.2.7 Sony EXView HAD Technology 381 6.5 Block Diagrams and their Description 383 6.5.1 Block Diagram of a Progressive Scan Analog Camera 383 6.5.1.1 CCD Read-Out Clocks 383 6.5.1.2 Spectral Sensitivity 386 6.5.1.3 Analog Signal Processing 387 6.5.1.4 Camera and Frame Grabber 388 6.5.2 Block Diagram of Color Camera with Digital Image Processing 389 6.5.2.1 BayerTM Complementary Color Filter Array 389 6.5.2.2 Complementary Color Filters Spectral Sensitivity 390 6.5.2.3 Generation of Color Signals 390 6.6 Digital Cameras 393 6.6.1 Black and White Digital Cameras 393 6.6.1.1 B/W Sensor and Processing 394 6.6.2 Color Digital Cameras 396 6.6.2.1 Analog Processing 396 6.6.2.2 One-Chip Color Processing 396 6.6.2.3 Analog Front End (AFE) 398 6.6.2.4 One Push White Balance 399 6.6.2.5 Automatic White Balance 400 6.6.2.6 Manual Gain 400 6.6.2.7 Auto Shutter/Gain 400 6.6.2.8 A/D Conversion 400 6.6.2.9 Lookup Table (LUT ) and Gamma Function 401 6.6.2.10 Shading Correction 402 6.6.2.11 Horizontal Mirror Function 403 6.6.2.12 Frame Memory and Deferred Image Transport 404 6.6.2.13 Color Interpolation 404 6.6.2.14 Color Correction 406 6.6.2.15 RGB to YUV Conversion 406 6.6.2.16 Binning versus Area of Interest (AOI ) 406 6.7 Controlling Image Capture 407
Page 14
XIV Contents 6.7.1 Hardware Trigger Modes 408 6.7.1.1 Latency (Jitter ) Aspects 409 6.7.2 Pixel Data 410 6.7.3 Data Transmission 411 6.7.4 IEEE-1394 Port Pin Assignment 412 6.7.5 Operating the Camera 414 6.7.6 HiRose Jack Pin Assignment 414 6.7.7 Frame Rates and Bandwidth 415 6.8 Configuration of the Camera 418 6.8.1 Camera Status Register 418 6.9 Camera Noise1 420 6.9.1 Photon Noise 421 6.9.2 Dark Current Noise 421 6.9.3 Photo Response Nonuniformity (PRNU) 422 6.9.4 Reset Noise 422 6.9.5 1/f Noise (Amplifier Noise) 422 6.9.6 Quantization Noise 423 6.9.7 Noise Floor 423 6.9.8 Dynamic Range 423 6.9.9 Signal to Noise Ratio 424 6.10 Digital Interfaces 424 References 426 7 Camera Computer Interfaces 427 Tony Iglesias, Anita Salmon, Johann Scholtz, Robert Hedegore, Julianna Borgendale, Brent Runnels, Nathan McKimpson, National Instruments 7.1 Overview 427 7.2 Analog Camera Buses 429 7.2.1 Analog Video Signal 429 7.2.2 Interlaced Video 430 7.2.3 Progressive Scan Video 431 7.2.4 Timing Signals 431 7.2.5 Analog Image Acquisition 432 7.2.6 S-Video 432 7.2.7 RGB 433 7.2.8 Analog Connectors 433 7.3 Parallel Digital Camera Buses 434 7.3.1 Digital Video Transmission 434 7.3.2 Taps 434 7.3.3 Differential Signaling 436 7.3.4 Line Scan 436
Page 15
Contents XV 7.3.5 Parallel Digital Connectors 437 7.3.6 Camera Link 437 7.3.7 Camera Link Signals 439 7.3.7.1 Camera Link Connectors 440 7.4 Standard PC Buses 440 7.4.1 USB 440 7.4.1.1 USB for Machine Vision 442 7.4.2 IEEE 1394 (FireWire©R ) 442 7.4.2.1 IEEE 1394 for Machine Vision 445 7.4.3 Gigabit Ethernet (IEEE 802.3z) 450 7.4.3.1 Gigabit Ethernet for Machine Vision 451 7.5 Choosing a Camera Bus 453 7.5.1 Bandwidth 454 7.5.2 Resolution 455 7.5.3 Frame Rate 457 7.5.4 Cables 457 7.5.5 Line Scan 458 7.5.6 Reliability 458 7.5.7 Summary of Camera Bus Specifications 460 7.5.8 Sample Use Cases 460 7.5.8.1 Manufacturing inspection 460 7.5.8.2 LCD inspection 461 7.5.8.3 Security 461 7.6 Computer Buses 462 7.6.1 ISA/EISA 462 7.6.2 PCI/CompactPCI/PXI 464 7.6.3 PCI-X 465 7.6.4 PCI Express/CompactPCI Express/PXI Express 466 7.6.5 Throughput 469 7.6.6 Prevalence and Lifetime 471 7.6.7 Cost 472 7.7 Choosing a Computer Bus 472 7.7.1 Determine Throughput Requirements 472 7.7.2 Applying the Throughput Requirements 474 7.8 Driver Software 475 7.8.1 Application Programming Interface 476 7.8.2 Supported Platforms 479 7.8.3 Performance 479 7.8.4 Utility Functions 481 7.8.5 Acquisition Mode 481 7.8.5.1 Snap 481 7.8.5.2 Grab 482
Page 16
XVI Contents 7.8.5.3 Sequence 483 7.8.5.4 Ring 483 7.8.6 Image Representation 485 7.8.6.1 Image Representation in Memory 485 7.8.6.2 Bayer Color Encoding 489 7.8.6.3 Image Representation on Disk 491 7.8.7 Image Display 492 7.8.7.1 Understanding Display Modes 492 7.8.7.2 Palettes 494 7.8.7.3 Nondestructive Overlays 494 7.9 Features of a Machine Vision System 495 7.9.1 Image Reconstruction 496 7.9.2 Timing and Triggering 497 7.9.3 Memory Handling 500 7.9.4 Additional Features 502 7.9.4.1 Look-up Tables 502 7.9.4.2 Region of Interest 504 7.9.4.3 Color Space Conversion 505 7.9.4.4 Shading Correction 507 7.9.5 Summary 508 8 Machine Vision Algorithms 511 Dr. Carsten Steger, MVTec Software GmbH 8.1 Fundamental Data Structures 511 8.1.1 Images 511 8.1.2 Regions 513 8.1.3 Subpixel-Precise Contours 515 8.2 Image Enhancement 516 8.2.1 Gray Value Transformations 516 8.2.2 Radiometric Calibration 519 8.2.3 Image Smoothing 525 8.3 Geometric Transformations 536 8.3.1 Affine Transformations 537 8.3.2 Projective Transformations 538 8.3.3 Image Transformations 539 8.3.4 Polar Transformations 544 8.4 Image Segmentation 544 8.4.1 Thresholding 545 8.4.2 Extraction of Connected Components 551 8.4.3 Subpixel-Precise Thresholding 553 8.5 Feature Extraction 555
Page 17
Contents XVII 8.5.1 Region Features 556 8.5.2 Gray Value Features 560 8.5.3 Contour Features 565 8.6 Morphology 566 8.6.1 Region Morphology 566 8.6.2 Gray Value Morphology 583 8.7 Edge Extraction 587 8.7.1 Definition of Edges in 1D and 2D 587 8.7.2 1D Edge Extraction 591 8.7.3 2D Edge Extraction 597 8.7.4 Accuracy of Edges 605 8.8 Segmentation and Fitting of Geometric Primitives 612 8.8.1 Fitting Lines 613 8.8.2 Fitting Circles 617 8.8.3 Fitting Ellipses 619 8.8.4 Segmentation of Contours into Lines, Circles, and Ellipses 620 8.9 Template Matching 624 8.9.1 Gray-Value-Based Template Matching 626 8.9.2 Matching Using Image Pyramids 632 8.9.3 Subpixel-Accurate Gray-Value-Based Matching 636 8.9.4 Template Matching with Rotations and Scalings 637 8.9.5 Robust Template Matching 638 8.10 Stereo Reconstruction 653 8.10.1 Stereo Geometry 654 8.10.2 Stereo Matching 662 8.11 Optical Character Recognition 666 8.11.1 Character Segmentation 667 8.11.2 Feature Extraction 669 8.11.3 Classification 672 References 688 9 Machine Vision in Manufacturing 693 Dr.-Ing. Peter Waszkewitz, Robert Bosch GmbH 9.1 Introduction 693 9.2 Application Categories 694 9.2.1 Types of Tasks 694 9.2.2 Types of Production 696 9.2.2.1 Discrete Unit Production Versus Continuous Flow 697 9.2.3 Types of Evaluations 698 9.2.4 Value-Adding Machine Vision 699 9.3 System Categories 700
Page 18
XVIII Contents 9.3.1 Common Types of Systems 700 9.3.2 Sensors 701 9.3.3 Vision Sensors 702 9.3.4 Compact Systems 703 9.3.5 Vision Controllers 704 9.3.6 PC-Based Systems 704 9.3.7 Summary 707 9.4 Integration and Interfaces 708 9.5 Mechanical Interfaces 710 9.5.1 Dimensions and Fixation 711 9.5.2 Working Distances 711 9.5.3 Position Tolerances 712 9.5.4 Forced Constraints 713 9.5.5 Additional Sensor Requirements 713 9.5.6 Additional Motion Requirements 714 9.5.7 Environmental Conditions 715 9.5.8 Reproducibility 716 9.5.9 Gauge Capability 718 9.6 Electrical Interfaces 720 9.6.1 Wiring and Movement 721 9.6.2 Power Supply 721 9.6.3 Internal Data Connections 722 9.6.4 External Data Connections 724 9.7 Information Interfaces 725 9.7.1 Interfaces and Standardization 725 9.7.2 Traceability 726 9.7.3 Types of Data and Data Transport 727 9.7.4 Control Signals 727 9.7.5 Result and Parameter Data 728 9.7.6 Mass Data 729 9.7.7 Digital I/O 729 9.7.8 Field Bus 730 9.7.9 Serial Interfaces 730 9.7.10 Network 731 9.7.11 Files 732 9.7.12 Time and Integrity Considerations 732 9.8 Temporal Interfaces 734 9.8.1 Discrete Motion Production 735 9.8.2 Continuous Motion Production 737 9.8.3 Line-Scan Processing 740 9.9 Human–Machine Interfaces 743 9.9.1 Interfaces for Engineering Vision Systems 743
Page 19
Contents XIX 9.9.2 Runtime Interface 745 9.9.3 Remote Maintenance 749 9.9.4 Offline Setup 750 9.10 Industrial Case Studies 752 9.10.1 Glue Check under UV Light 752 9.10.1.1 Solution 752 9.10.2 Completeness Check 754 9.10.3 Multiple Position and Completeness Check 757 9.10.4 Pin Type Verification 760 9.10.5 Robot Guidance 763 9.10.6 Type and Result Data Management 766 9.11 Constraints and Conditions 769 9.11.1 Inspection Task Requirements 769 9.11.2 Circumstantial Requirements 770 9.11.3 Limits and Prospects 773 References 775 Index 781
Page 20
XXI Preface Why a Further Book on Machine Vision? The idea to write another book about machine vision (or image processing) seems to be unnecessary. That is only the first impression. The search of a book which describes the whole information processing chain was unsuccess- ful, because the most books deal predominantly with the algorithms of image processing which is an important part of a machine vision application. They do not take into account that first we have to acquire a digital image of the real part of the world, which have some important properties: • high contrast • high resolution The success of developing a machine vision system depends on the under- standing all parts of the imaging chain. The charm and the complexity of machine vision lies in the range of the specialized engineering fields involved in it, namely • mechanical engineering • electrical engineering • optical engineering • software engineering each of which struggles for a primary role. The interdisciplinary thought is the base for the successful development of a machine vision applications. Today we have a new term for this field of engineering called mechatronics. This fact of the case determine the difficulties and the possibilities of machine vision inspection. Handbook of Machine Vision. Alexander Hornberg (Ed.) Copyright c© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-40584-7
Showing the first 20 pages. Download the document to see it all.
Comments (0)
No comments yet. Be the first!