Table Of Content

�� Since the first edition of “Digital Image Processing” was published in 1991 it has found its way to many desks and J Bernd Jähne ä classrooms. The book offers an integral view of image pro- h n cessing from image acquisition to the extraction of the data e of interest. The discussion of the general concepts is supple- �� mented with examples from applications on PC-based im- �� age processing systems and ready-to-use implementations Digital of important algorithms. The sixth edition has been revised and extended. The whole textbook is now clearly partitioned into basic and advanced Image Processing material in order to cope with the ever-increasing field of digital image processing. In this way, you can first work your way through the basic principles of digital image processing without getting overwhelmed by the wealth of the material and then extend your studies to selected topics of inter- 1 6th revised and extended edition est. Each chapter now includes exercises that help you to test your understanding, train your skills, and introduce you to real-world image processing tasks. An important part of the exercises is a wealth of interactive computer exercises, which cover all topics of this textbook. These exercises are D performed with the image processing software heurisko, i g which is included on the accompanying CD-ROM. In this i way you can get own practical experience with almost all t a topics and algorithms covered by this book. The CD-ROM l also includes a large collection of images, image sequences, I and volumetric images that can be used together with the m computer exercises. The complete hyperlinked text of the a book is now available on the accompanying CD-ROM. It is g also possible to execute the computer exercises directly e from the PDF document. P r o c e s s i n g ISBN 3-540-24035-7 6 t h e d 9 783540 240358 it 123 io n springeronline.com Jaehne 24035-7 e.indd 1 07.02.2005 16:30:09 Uhr Bernd Jähne Digital Image Processing Bernd Jähne Digital Image Processing 6th revised and extended edition With 248 Figures ,155 Exercises, and CD-ROM 1 3 Professor Dr. Bernd Jähne Interdisciplinary Center for Scientific Computing University of Heidelberg Im Neuenheimer Feld 368 69120 Heidelberg Germany [email protected] www.bernd-jaehne.de http://klimt.uni-heidelberg.de Library of Congress Control Number: 2005920591 ISBN 3-540-24035-7 Springer Berlin Heidelberg New York This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in other ways, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution act under German Copyright Law. Springer is a part of Springer Science + Business Media springeronline.com © Springer-Verlag Berlin Heidelberg 2005 Printed in Germany The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting: Digital data supplied by author Cover-Design: Struve & Partner,Heidelberg Production: medionet AG, Berlin Printed on acid-free paper 62/3141 Rw 5 4 3 2 1 0 Preface The sixth edition of this worldwide used textbook was thoroughly re- visedandextended. Throughoutthewholetextyouwillfindnumerous improvements,extensions,andupdates. Aboveall,Iwouldliketodraw yourattentiontotwomajorchanges. Firstly, the whole textbook is now clearly partitioned into basic and advancedmaterialinordertocopewiththeever-increasingfieldofdigi- talimageprocessing. Themostimportantequationsareputintoframed boxes. The advanced sections are located in the second part of each chapter and are marked by italic headlines and by a smaller typeface. In this way, you can first work your way through the basic principles of digital image processing without getting overwhelmed by the wealth of the material. You can extend your studies later to selected topics of interest. The second most notable extension are exercises that are now included at the end of each chapter. These exercise help you to test your understanding, train your skills, and introduce you to real-world image processing tasks. The exercises are marked with one to three stars to indicate their difficulty. An important part of the exercises is a wealth of interactive computer exercises, which cover all topics of this textbook. These exercises are performed with the image processing software heurisko® (http://www.heurisko.de), which is included on the accompanying CD-ROM. In this way you can get own practical experience with almost all topics and algorithms covered by this book. The CD-ROM also includes a large collection of images, image sequences, andvolumetricimagesthatcanbeusedtogetherwiththecomputerex- ercises. Informationaboutthesolutionsoftheexercisesandupdatesof the computer exercises can be found on the homepage of the author at http://www.bernd-jaehne.de. Eachchaptercloseswithasection“FurtherReading”thatguidesthe interestedreadertofurtherreferences. Theappendixincludestwochap- ters. AppendixAgivesaquickaccesstoacollectionofoftenusedrefer- encematerialandAppendixBdetailsthenotationusedthroughoutthe book. Thecompletetextofthebookisnowavailableontheaccompany- ingCD-ROM.Itishyperlinkedsothatitcanbeusedinaveryflexibleway. V VI You can jump from the table of contents to the corresponding section, fromcitationstothebibliography,fromtheindextothecorresponding page,andtoanyothercross-references. Itisalsopossibletoexecutethe computerexercisesdirectlyfromthePDFdocument. Iwouldliketothankallindividualsandorganizationswhohavecon- tributedvisualmaterialforthisbook. Thecorrespondingacknowledge- ments can be found where the material is used. I would also like to express my sincere thanks to the staff of Springer-Verlag for their con- stantinterestinthisbookandtheirprofessionaladvice. Specialthanks areduetomyfriendsatAEONVerlag&Studio, Hanau, Germany. With- outtheirdedicationandprofessionalknowledgeitwouldnothavebeen possibletoproducethisbookand,inparticular,theaccompanyingCD- ROM. Finally, Iwelcomeanyconstructiveinputfromyou, thereader. Iam grateful for comments on improvements or additions and for hints on errors,omissions,ortypingerrors,which—despiteallthecaretaken— mayhaveslippedattention. Heidelberg,January2005 BerndJähne From the preface of the fifth edition Asthefourthedition,thefiftheditioniscompletelyrevisedandextended. The wholetextofthebookisnowarrangedin20insteadof16chapters. Aboutone thirdoftextismarkedasadvancedmaterial. Inthisway,youwillfindaquick andsystematicwaythroughthebasicmaterialandyoucanextendyourstudies latertospecialtopicsofinterest. Themostnotableextensionsincludeadetaileddiscussiononrandomvariables andfields(Chapter3),3-Dimagingtechniques(Chapter8)andanapproachto regularized parameter estimation unifying techniques including inverse problems, adaptive filter techniques such as anisotropic diffusion, and variational approachesforoptimalsolutionsinimagerestoration,tomographicreconstruc- tion,segmentation,andmotiondetermination(Chapter17). Eachchapternow closes with a section “Further Reading” that guides the interested reader to furtherreferences. ThecompletetextofthebookisnowavailableontheaccompanyingCD-ROM. It is hyperlinked so that it can be used in a very flexible way. You can jump fromthetableofcontentstothecorrespondingsection, fromcitationstothe bibliography,fromtheindextothecorrespondingpage,andtoanyothercross- references. Heidelberg,November2001 BerndJähne From the preface of the fourth edition Inafastdevelopingareasuchasdigitalimageprocessingabookthatappeared initsfirsteditionin1991requiredacompleterevisionjustsixyearslater. But whathasnotchangedistheprovenconcept,offeringasystematicapproachto VII digital image processing with the aid of concepts and general principles also used in other areas of natural science. In this way, a reader with a general backgroundinnaturalscienceoranengineeringdisciplineisgivenfastaccess to the complex subject of image processing. The book covers the basics of imageprocessing. Selectedareasaretreatedindetailinordertointroducethe reader both to the way of thinking in digital image processing and to some current research topics. Whenever possible, examples and image material are usedtoillustratebasicconcepts. Itisassumedthatthereaderisfamiliarwith elementarymatrixalgebraandtheFouriertransform. Theneweditioncontainsfourparts. Part1summarizesthebasicsrequiredfor understandingimageprocessing. Thusthereisnolongeramathematicalappen- dixasinthepreviouseditions. Part2onimageacquisitionandpreprocessing hasbeenextendedbyadetaileddiscussionofimageformation. Motionanalysis hasbeenintegratedintoPart3asonecomponentoffeatureextraction. Object detection,objectformanalysis,andobjectclassificationareputtogetherinPart 4onimageanalysis. Generally,thisbookisnotrestrictedto2-Dimageprocessing. Whereverpossi- ble,thesubjectsaretreatedinsuchamannerthattheyarealsovalidforhigher- dimensionalimagedata(volumetricimages,imagesequences). Likewise,color imagesareconsideredasaspecialcaseofmultichannelimages. Heidelberg,May1997 BerndJähne From the preface of the first edition Digitalimageprocessingisafascinatingsubjectinseveralaspects. Humanbe- ings perceive most of the information about their environment through their visual sense. While for a long time images could only be captured by photo- graphy, we are now at the edge of another technological revolution which al- lowsimagedatatobecaptured,manipulated,andevaluatedelectronicallywith computers. With breathtaking pace, computers are becoming more powerful andatthesametimelessexpensive,sothatwidespreadapplicationsfordigital imageprocessingemerge. Inthisway,imageprocessingisbecomingatremen- dous tool for analyzing image data in all areas of natural science. For more andmorescientistsdigitalimageprocessingwillbethekeytostudycomplex scientific problems they could not have dreamed of tackling only a few years ago. Adoorisopeningfornewinterdisciplinarycooperationmergingcomputer sciencewiththecorrespondingresearchareas. Manystudents,engineers,andresearchersinallnaturalsciencesarefacedwith the problem of needing to know more about digital image processing. This book is written to meet this need. The author — himself educated in physics —describesdigitalimageprocessingasanewtoolforscientificresearch. The book starts with the essentials of image processing and leads — in selected areas—tothestate-of-theart. Thisapproachgivesaninsightastohowimage processing really works. The selection of the material is guided by the needs of a researcher who wants to apply image-processing techniques in his or her field. Inthissense,thisbooktriestoofferanintegralviewofimageprocessing fromimageacquisitiontotheextractionofthedataofinterest. Manyconcepts andmathematicaltoolsthatfindwidespreadapplicationinnaturalsciencesare VIII alsoappliedindigitalimageprocessing. Suchanalogiesarepointedout,since theyprovideaneasyaccesstomanycomplexproblemsindigitalimageprocess- ingforreaderswithageneralbackgroundinnaturalsciences. Thediscussion of the general concepts is supplemented with examples from applications on PC-based image processing systems and ready-to-use implementations of im- portantalgorithms. Iamdeeplyindebtedtothemanyindividualswhohelpedmetowritethisbook. Idothisbytracingitshistory. Intheearly1980s,whenIworkedonthephysics ofsmall-scaleair-seainteractionattheInstituteofEnvironmentalPhysicsatHei- delbergUniversity,itbecameobviousthatthesecomplexphenomenacouldnot beadequatelytreatedwithpointmeasuringprobes. Consequently,anumberof areaextendedmeasuringtechniquesweredeveloped. ThenIsearchedfortech- niques to extract the physically relevant data from the images and sought for colleagueswithexperienceindigitalimageprocessing. Thefirstcontactswere establishedwiththeInstituteforAppliedPhysicsatHeidelbergUniversityand theGermanCancerResearchCenterinHeidelberg. IwouldliketothankProf. Dr. J. Bille, Dr. J. Dengler and Dr. M. Schmidt cordially for many eye-opening conversationsandtheircooperation. IwouldalsoliketothankProf.Dr.K.O.Münnich, directoroftheInstitutefor Environmental Physics. From the beginning, he was open-minded about new ideasontheapplicationofdigitalimageprocessingtechniquesinenvironmen- tal physics. It is due to his farsightedness and substantial support that the research group “Digital Image Processing in Environmental Physics” could de- velop so fruitfully at his institute. Many of the examples shown in this book aretakenfrommyresearchatHeidelbergUniversityandtheScrippsInstitution ofOceanography. Igratefullyacknowledgefinancialsupportforthisresearch fromtheGermanScienceFoundation,theEuropeanCommunity,theUSNational ScienceFoundation,andtheUSOfficeofNavalResearch. LaJolla,California,andHeidelberg,spring1991 BerndJähne Contents I Foundation 1 ApplicationsandTools 3 1.1 AToolforScienceandTechnique . . . . . . . . . . . . . 3 1.2 ExamplesofApplications . . . . . . . . . . . . . . . . . . 4 1.3 HierarchyofImageProcessingOperations . . . . . . . . 15 1.4 ImageProcessingandComputerGraphics . . . . . . . . 17 1.5 Cross-disciplinaryNatureofImageProcessing . . . . . 17 1.6 HumanandComputerVision . . . . . . . . . . . . . . . . 18 1.7 ComponentsofanImageProcessingSystem . . . . . . 21 1.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1.9 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 28 2 ImageRepresentation 31 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.2 SpatialRepresentationofDigitalImages . . . . . . . . . 31 2.3 WaveNumberSpaceandFourierTransform . . . . . . . 41 2.4 DiscreteUnitaryTransforms . . . . . . . . . . . . . . . . 63 2.5 FastAlgorithmsforUnitaryTransforms . . . . . . . . . 67 2.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 2.7 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 80 3 RandomVariablesandFields 81 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 81 3.2 RandomVariables . . . . . . . . . . . . . . . . . . . . . . . 83 3.3 MultipleRandomVariables . . . . . . . . . . . . . . . . . 87 3.4 ProbabilityDensityFunctions . . . . . . . . . . . . . . . . 91 3.5 StochasticProcessesandRandomFields . . . . . . . . . . 98 3.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 3.7 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 104 4 NeighborhoodOperations 105 4.1 BasicPropertiesandPurpose . . . . . . . . . . . . . . . . 105 4.2 LinearShift-InvariantFilters . . . . . . . . . . . . . . . . . 108 4.3 RankValueFilters . . . . . . . . . . . . . . . . . . . . . . . 119 4.4 LSI-Filters: FurtherProperties . . . . . . . . . . . . . . . . 120 4.5 RecursiveFilters . . . . . . . . . . . . . . . . . . . . . . . . 122 IX X Contents 4.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 4.7 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 134 5 MultiscaleRepresentation 135 5.1 Scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 5.2 MultigridRepresentations . . . . . . . . . . . . . . . . . . 138 5.3 ScaleSpaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 5.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 5.5 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 153 II Image Formation and Preprocessing 6 QuantitativeVisualization 157 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 157 6.2 Radiometry,Photometry,Spectroscopy,andColor . . . 159 6.3 WavesandParticles . . . . . . . . . . . . . . . . . . . . . . 168 6.4 InteractionsofRadiationwithMatter . . . . . . . . . . . 174 6.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 6.6 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 187 7 ImageFormation 189 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 189 7.2 WorldandCameraCoordinates. . . . . . . . . . . . . . . 189 7.3 IdealImaging: PerspectiveProjection . . . . . . . . . . . 192 7.4 RealImaging . . . . . . . . . . . . . . . . . . . . . . . . . . 195 7.5 RadiometryofImaging . . . . . . . . . . . . . . . . . . . . 201 7.6 LinearSystemTheoryofImaging. . . . . . . . . . . . . . 205 7.7 HomogeneousCoordinates . . . . . . . . . . . . . . . . . . 212 7.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 7.9 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 215 8 3-DImaging 217 8.1 Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 8.2 DepthfromTriangulation . . . . . . . . . . . . . . . . . . 221 8.3 DepthfromTime-of-Flight . . . . . . . . . . . . . . . . . . 228 8.4 DepthfromPhase: Interferometry . . . . . . . . . . . . . 229 8.5 ShapefromShading . . . . . . . . . . . . . . . . . . . . . . 229 8.6 DepthfromMultipleProjections: Tomography . . . . . 235 8.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 8.8 FurtherReadings . . . . . . . . . . . . . . . . . . . . . . . . 242 9 Digitization,Sampling,Quantization 243 9.1 DefinitionandEffectsofDigitization . . . . . . . . . . . 243 9.2 ImageFormation,Sampling,Windowing . . . . . . . . . 245 9.3 ReconstructionfromSamples . . . . . . . . . . . . . . . . 249 9.4 MultidimensionalSamplingonNonorthogonalGrids . . 251 9.5 Quantization . . . . . . . . . . . . . . . . . . . . . . . . . . 253 9.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254