版权信息书 名: 机器视觉理论、算法与
实践
作者：(英国)E.R.Davies
出版社：人民邮电出版社
出版时间： 2009
ISBN: 9787115195494
开本： 16
定价: 128.00 元
内容简介《机器视觉理论、算法与实践(英文版·第3版)》是机器视觉课程的理想教材，作者清晰、系统地阐述了机器视觉的基本概念，介绍理论的基本元素的同时强调算法和实用设计的约束。书中阐述各个主题时，既阐述了基本算法，又介绍了数学工具。此外，《机器视觉理论、算法与实践(英文版·第3版)》还使用案例演示具体技术的应用，并阐明设计现实机器视觉系统的关键约束。
《机器视觉理论、算法与实践(英文版·第3版)》适合作为高等院校计算机及电子工程相关专业研究生的教材，更是从事机器视觉、计算机视觉和机器人领域研究的人员不可多得的技术参考书。
作者简介E.R.Davies，著名机器视觉专家。英国物理学会会士、IEE会士、英国机器视觉协会的执行委员。毕业于牛津大学，现任伦敦大学皇家霍洛威学院机器视觉教授。在机器视觉、图像分析、自动视觉检测、噪声抑制技术等方面有丰富的教学和科研经验。
编辑推荐40年来，机器视觉在各行各业得到了广泛的应用，包括自动检测、机器人组装、行车导引、流量监控、签名验证、生物测量、遥感图像分析等。但是另一方面，面对大量新的研究成果，要充分理解相关的理论和应用，进行算法和系统的设计，却越来越困难。
《机器视觉理论、算法与实践(英文版·第3版)》能够满足广大读者学习和掌握机器视觉知识的需求。全书图文并茂，清晰、系统地阐述了基本概念，提供了丰富的应用案例和代码，强调了算法和实用设计的各种约束条件。新版做了全面的更新，反映了最新进展，内容更加全面。《机器视觉理论、算法与实践(英文版·第3版)》是机器视觉课程的理想教材，已经成为国内外很多名校的指定教学参考书。同时，《机器视觉理论、算法与实践(英文版·第3版)》也是工程技术人员不可或缺的权威参考书。
目录CHAPTER1Vision,theChallenge
1.1Introduction-TheSenses1
1.2TheNatureofVision2
1.2.1TheProcessofRecognition2
1.2.2TacklingtheRecognitionProblem4
1.2.3ObjectLocation7
1.2.4SceneAnalysis9
1.2.5VisionasInverseGraphics10
1.3FromAutomatedVisualInspectiontoSurveillance11
1.4WhatThisBookIsAbout12
1.5TheFollowingChapters14
1.6BibliographicalNotes15
PART1LOW-LEVELVISION17
CHAPTER2ImagesandImagingOperations
2.1Introduction19
2.1.1Gray-scaleversusColor21*
2.2ImageProcessingOperations24
2.2.1SomeBasicOperationsonGray-scaleImages25
2.2.2BasicOperationsonBinaryImages32
2.2.3NoiseSuppressionbyImageAccumulation37
2.3ConvolutionsandPointSpreadFunctions39
2.4SequentialversusParallelOperations41
2.5ConcludingRemarks43
2.6BibliographicalandHistoricalNotes44
2.7Problems44
CHAPTER3BasicImageFilteringOperations
3.1Introduction47
3.2NoiseSuppressionbyGaussianSmoothing49
3.3MedianFilters51
3.4ModeFilters54
3.5RankOrderFilters61
3.6ReducingComputationalLoad61
3.6.1ABit-basedMethodforFastMedianFiltering64
3.7Sharp-UnsharpMasking65
3.8ShiftsIntroducedbyMedianFilters66
3.8.1ContinuumModelofMedianShifts68
3.8.2GeneralizationtoGray-scaleImages72
3.8.3ShiftsArisingwithHybridMedianFilters75
3.8.4ProblemswithStatistics76
3.9DiscreteModelofMedianShifts78
3.9.1GeneralizationtoGray-scaleImages82
3.10ShiftsIntroducedbyModeFilters84
3.11ShiftsIntroducedbyMeanandGaussianFilters86
3.12ShiftsIntroducedbyRankOrderFilters86
3.12.1ShiftsinRectangularNeighborhoods87
3.12.2CaseofHighCurvature91
3.12.3TestoftheModelinaDiscreteCase91
3.13TheRoleofFiltersinIndustrialApplicationsofVision94
3.14ColorinImageFiltering94
3.15ConcludingRemarks96
3.16BibliographicalandHistoricalNotes96
3.17Problems98
CHAPTER4ThresholdingTechniques
4.1Introduction103
4.2Region-growingMethods104
4.3Thresholding105
4.3.1FindingaSuitableThreshold105
4.3.2TacklingtheProblemofBiasinThresholdSelection107
4.3.3AConvenientMathematicalModel111
4.3.4Summary114
4.4AdaptiveThresholding114
4.4.1TheChowandKanekoApproach118
4.4.2LocalThresholdingMethods119
4.5MoreThoroughgoingApproachestoThresholdSelection122
4.5.1Variance-basedThresholding122
4.5.2Entropy-basedThresholding123
4.5.3MaximumLikelihoodThresholding125
4.6ConcludingRemarks126
4.7BibliographicalandHistoricalNotes127
4.8Problems129
CHAPTER5EdgeDetection
5.1Introduction131
5.2BasicTheoryofEdgeDetection132
5.3TheTemplateMatchingApproach133
5.4Theoryof3×3TemplateOperators135
5.5Summary-DesignConstraintsandConclusions140
5.6TheDesignofDifferentialGradientOperators141
5.7TheConceptofaCircularOperator143
5.8DetailedImplementationofCircularOperators144
5.9StructuredBandsofPixelsinNeighborhoodsofVariousSizes146
5.10TheSystematicDesignofDifferentialEdgeOperators150
5.11ProblemswiththeaboveApproach-SomeAlternativeSchemes151
5.12ConcludingRemarks155
5.13BibliographicalandHistoricalNotes156
5.14Problems157
CHAPTER6BinaryShapeAnalysis
6.1Introduction159
6.2ConnectednessinBinaryImages160
6.3ObjectLabelingandCounting161
6.3.1SolvingtheLabelingProbleminaMoreComplexCase164
6.4MetricPropertiesinDigitalImages168
6.5SizeFiltering169
6.6TheConvexHullandItsComputation171
6.7DistanceFunctionsandTheirUses177
6.8SkeletonsandThinning181
6.8.1CrossingNumber183
6.8.2ParallelandSequentialImplementationsofThinning186
6.8.3GuidedThinning189
6.8.4ACommentontheNatureoftheSkeleton189
6.8.5SkeletonNodeAnalysis191
6.8.6ApplicationofSkeletonsforShapeRecognition192
6.9SomeSimpleMeasuresforShapeRecognition193
6.10ShapeDescriptionbyMoments194
6.11BoundaryTrackingProcedures195
6.12MoreDetailontheSigmaandChiFunctions196
6.13ConcludingRemarks197
6.14BibliographicalandHistoricalNotes199
6.15Problems200
CHAPTER7BoundaryPatternAnalysis
7.1Introduction207
7.1.1HysteresisThresholding209
7.2BoundaryTrackingProcedures212
7.3TemplateMatching-AReminder212
7.4CentroidalProfiles213
7.5ProblemswiththeCentroidalProfileApproach214
7.5.1SomeSolutions216
7.6The(s,ψ)Plot218
7.7TacklingtheProblemsofOcclusion220
7.8ChainCode223
7.9The(r,s)Plot224
7.10AccuracyofBoundaryLengthMeasures225
7.11ConcludingRemarks227
7.12BibliographicalandHistoricalNotes228
7.13Problems229
CHAPTER8MathematicalMorphology
8.1Introduction233
8.2DilationandErosioninBinaryImages234
8.2.1DilationandErosion234
8.2.2CancellationEffects234
8.2.3ModifiedDilationandErosionOperators235
8.3MathematicalMorphology235
8.3.1GeneralizedMorphologicalDilation235
8.3.2GeneralizedMorphologicalErosion237
8.3.3DualitybetweenDilationandErosion238
8.3.4PropertiesofDilationandErosionOperators239
8.3.5ClosingandOpening242
8.3.6SummaryofBasicMorphologicalOperations245
8.3.7Hit-and-MissTransform248
8.3.8TemplateMatching249
8.4Connectivity-basedAnalysisofImages249
8.4.1SkeletonsandThinning250
8.5Gray-scaleProcessing251
8.5.1MorphologicalEdgeEnhancement252
8.5.2FurtherRemarksontheGeneralizationtoGray-scaleProcessing252
8.6EffectofNoiseonMorphologicalGroupingOperations255
8.6.1DetailedAnalysis257
8.6.2Discussion259
8.7ConcludingRemarks259
8.8BibliographicalandHistoricalNotes260
8.9Problem261
PART2INTERMEDIATE-LEVELVISION263
CHAPTER9LineDetection
9.1Introduction265
9.2ApplicationoftheHoughTransformtoLineDetection265
9.3TheFoot-of-NormalMethod269
9.3.1ErrorAnalysis272
9.3.2QualityoftheResultingData274
9.3.3ApplicationoftheFoot-of-NormalMethod276
9.4LongitudinalLineLocalization276
9.5FinalLineFitting277
9.6ConcludingRemarks277
9.7BibliographicalandHistoricalNotes278
9.8Problems280
CHAPTER10CircleDetection
10.1Introduction283
10.2Hough-basedSchemesforCircularObjectDetection284
10.3TheProblemofUnknownCircleRadius288
10.3.1ExperimentalResults290
10.4TheProblemofAccurateCenterLocation295
10.4.1ObtainingaMethodforReducingComputationalLoad296
10.4.2ImprovementsontheBasicScheme299
10.4.3Discussion300
10.4.4PracticalDetails300
10.5OvercomingtheSpeedProblem302
10.5.1MoreDetailedEstimatesofSpeed303
10.5.2Robustness305
10.5.3ExperimentalResults306
10.5.4Summary307
10.6ConcludingRemarks310
10.7BibliographicalandHistoricalNotes311
10.8Problems312
CHAPTER11TheHoughTransformandItsNature
11.1Introduction315
11.2TheGeneralizedHoughTransform315
11.3SettingUptheGeneralizedHoughTransform-SomeRelevantQuestions317
11.4SpatialMatchedFilteringinImages318
11.5FromSpatialMatchedFilterstoGeneralizedHoughTransforms319
11.6GradientWeightingversusUniformWeighting320
11.6.1CalculationofSensitivityandComputationalLoad323
11.7Summary324
11.8ApplyingtheGeneralizedHoughTransformtoLineDetection325
11.9TheEffectsofOcclusionsforObjectswithStraightEdges327
11.10FastImplementationsoftheHoughTransform329
11.11TheApproachofGerigandKlein332
11.12ConcludingRemarks333
11.13BibliographicalandHistoricalNotes334
11.14Problem337
CHAPTER12EllipseDetection
12.1Introduction339
12.2TheDiameterBisectionMethod339
12.3TheChord-TangentMethod341
12.4FindingtheRemainingEllipseParameters343
12.5ReducingComputationalLoadfortheGeneralizedHoughTransformMethod345
12.5.1PracticalDetails349
12.6ComparingtheVariousMethods353
12.7ConcludingRemarks355
12.8BibliographicalandHistoricalNotes357
12.9Problems358
CHAPTER13HoleDetection
13.1Introduction361
13.2TheTemplateMatchingApproach361
13.3TheLateralHistogramTechnique363
13.4TheRemovalofAmbiguitiesintheLateralHistogramTechnique363
13.4.1ComputationalImplicationsoftheNeedtoCheckforAmbiguities364
13.4.2FurtherDetailoftheSubimageMethod366
13.5ApplicationoftheLateralHistogramTechniqueforObjectLocation368
13.5.1LimitationsoftheApproach370
13.6AppraisaloftheHoleDetectionProblem372
13.7ConcludingRemarks374
13.8BibliographicalandHistoricalNotes375
13.9Problems376
CHAPTER14PolygonandCornerDetection
14.1Introduction379
14.2TheGeneralizedHoughTransform380
14.2.1StraightEdgeDetection380
14.3ApplicationtoPolygonDetection381
14.3.1TheCaseofanArbitraryTriangle382
14.3.2TheCaseofanArbitraryRectangle383
14.3.3LowerBoundsontheNumbersofParameterPlanes385
14.4DeterminingPolygonOrientation387
14.5WhyCornerDetection?389
14.6TemplateMatching390
14.7Second-orderDerivativeSchemes391
14.8AMedian-Filter-BasedCornerDetector393
14.8.1AnalyzingtheOperationoftheMedianDetector394
14.8.2PracticalResults396
14.9TheHoughTransformApproachtoCornerDetection399
14.10ThePlesseyCornerDetector402
14.11CornerOrientation404
14.12ConcludingRemarks406
14.13BibliographicalandHistoricalNotes407
14.14Problems410
CHAPTER15AbstractPatternMatchingTechniques
15.1Introduction413
15.2AGraph-theoreticApproachtoObjectLocation414
15.2.1APracticalExample-LocatingCreamBiscuits419
15.3PossibilitiesforSavingComputation422
15.4UsingtheGeneralizedHoughTransformforFeatureCollation424
15.4.1ComputationalLoad426
15.5GeneralizingtheMaximalCliqueandOtherApproaches427
15.6RelationalDescriptors428
15.7Search432
15.8ConcludingRemarks433
15.9BibliographicalandHistoricalNotes434
15.10Problems437
PART33-DVISIONANDMOTION443
CHAPTER16TheThree-dimensionalWorld
16.1Introduction445
16.2Three-DimensionalVision-TheVarietyofMethods446
16.3ProjectionSchemesforThree-dimensionalVision448
16.3.1BinocularImages450
16.3.2TheCorrespondenceProblem452
16.4ShapefromShading454
16.5PhotometricStereo459
16.6TheAssumptionofSurfaceSmoothness462
16.7ShapefromTexture464
16.8UseofStructuredLighting464
16.9Three-DimensionalObjectRecognitionSchemes466
16.10TheMethodofBallardandSabbah468
16.11TheMethodofSilberbergetal.470
16.12Horaud’sJunctionOrientationTechnique472
16.13AnImportantParadigm-LocationofIndustrialParts476
16.14ConcludingRemarks478
16.15BibliographicalandHistoricalNotes480
16.16Problems482
CHAPTER17TacklingthePerspectiven-PointProblem
17.1Introduction487
17.2ThePhenomenonofPerspectiveInversion487
17.3AmbiguityofPoseunderWeakPerspectiveProjection489
17.4ObtainingUniqueSolutionstothePoseProblem493
17.4.1Solutionofthe3-PointProblem497
17.4.2UsingSymmetricalTrapeziaforEstimatingPose498
17.5ConcludingRemarks498
17.6BibliographicalandHistoricalNotes501
17.7Problems502
CHAPTER18Motion
18.1Introduction505
18.2OpticalFlow505
18.3InterpretationofOpticalFlowFields509
18.4UsingFocusofExpansiontoAvoidCollision511
18.5Time-to-AdjacencyAnalysis513
18.6BasicDifficultieswiththeOpticalFlowModel515
18.7StereofromMotion516
18.8ApplicationstotheMonitoringofTrafficFlow518
18.8.1TheSystemofBascleetal.518
18.8.2TheSystemofKolleretal.520
18.9PeopleTracking524
18.9.1SomeBasicTechniques526
18.9.2Within-vehiclePedestrianTracking528
18.10HumanGaitAnalysis530
18.11Model-basedTrackingofAnimals-ACaseStudy533
18.12Snakes536
18.13TheKalmanFilter538
18.14ConcludingRemarks540
18.15BibliographicalandHistoricalNotes542
18.16Problem543
CHAPTER19InvariantsandTheirApplications
19.1Introduction545
19.2CrossRatios:The“RatioofRatios”Concept547
19.3InvariantsforNoncollinearPoints552
19.3.1FurtherRemarksaboutthe5-PointConfiguration554
19.4InvariantsforPointsonConics556
19.5DifferentialandSemidifferentialInvariants560
19.6SymmetricalCrossRatioFunctions562
19.7ConcludingRemarks564
19.8BibliographicalandHistoricalNotes566
19.9Problems567
CHAPTER20EgomotionandRelatedTasks
20.1Introduction571
20.2AutonomousMobileRobots572
20.3ActiveVision573
20.4VanishingPointDetection574
20.5NavigationforAutonomousMobileRobots576
20.6ConstructingthePlanViewofGroundPlane579
20.7FurtherFactorsInvolvedinMobileRobotNavigation581
20.8MoreonVanishingPoints583
20.9CentersofCirclesandEllipses585
20.10VehicleGuidanceinAgriculture-ACaseStudy588
20.10.13-DAspectsoftheTask590
20.10.2Real-timeImplementation591
20.11ConcludingRemarks592
20.12BibliographicalandHistoricalNotes592
20.13Problems593
CHAPTER21ImageTransformationsandCameraCalibration
21.1Introduction595
21.2ImageTransformations596
21.3CameraCalibration601
21.4IntrinsicandExtrinsicParameters604
21.5CorrectingforRadialDistortions607
21.6Multiple-viewVision609
21.7GeneralizedEpipolarGeometry610
21.8TheEssentialMatrix611
21.9TheFundamentalMatrix613
21.10PropertiesoftheEssentialandFundamentalMatrices614
21.11EstimatingtheFundamentalMatrix615
21.12ImageRectification616
21.133-DReconstruction617
21.14AnUpdateonthe8-PointAlgorithm619
21.15ConcludingRemarks621
21.16BibliographicalandHistoricalNotes622
21.17Problems623
PART4TOWARDREAL-TIMEPATTERNRECOGNITIONSYSTEMS625
CHAPTER22AutomatedVisualInspection
22.1Introduction627
22.2TheProcessofInspection628
22.3ReviewoftheTypesofObjectstoBeInspected629
22.3.1FoodProducts629
22.3.2PrecisionComponents630
22.3.3DifferingRequirementsforSizeMeasurement630
22.3.4Three-dimensionalObjects631
22.3.5OtherProductsandMaterialsforInspection632
22.4Summary-TheMainCategoriesofInspection632
22.5ShapeDeviationsRelativetoaStandardTemplate634
22.6InspectionofCircularProducts635
22.6.1ComputationoftheRadialHistogram:StatisticalProblems636
22.6.2ApplicationofRadialHistograms641
22.7InspectionofPrintedCircuits642
22.8SteelStripandWoodInspection643
22.9InspectionofProductswithHighLevelsofVariability644
22.10X-rayInspection648
22.11TheImportanceofColorinInspection651
22.12BringingInspectiontotheFactory653
22.13ConcludingRemarks654
22.14BibliographicalandHistoricalNotes656
CHAPTER23InspectionofCerealGrains
23.1Introduction659
23.2CaseStudy1:LocationofDarkContaminantsinCereals660
23.2.1ApplicationofMorphologicalandNonlinearFilterstoLocateRodentDroppings663
23.2.2AppraisaloftheVariousSchemas664
23.2.3ProblemswithClosing665
23.3CaseStudy2:LocationofInsects665
23.3.1TheVectorialStrategyforLinearFeatureDetection666
23.3.2DesigningLinearFeatureDetectionMasksforLargerWindows669
23.3.3ApplicationtoCerealInspection670
23.3.4ExperimentalResults671
23.4CaseStudy3:High-speedGrainLocation673
23.4.1ExtendinganEarlierSamplingApproach673
23.4.2ApplicationtoGrainInspection675
23.4.3Summary679
23.5OptimizingtheOutputforSetsofDirectionalTemplateMasks680
23.5.1ApplicationoftheFormulas682
23.5.2Discussion683
23.6ConcludingRemarks683
23.7BibliographicalandHistoricalNotes684
CHAPTER24StatisticalPatternRecognition
24.1Introduction687
24.2TheNearestNeighborAlgorithm688
24.3Bayes’DecisionTheory691
24.4RelationoftheNearestNeighborandBayes’Approaches693
24.4.1MathematicalStatementoftheProblem693
24.4.2TheImportanceoftheNearestNeighborClassifier696
24.5TheOptimumNumberofFeatures696
24.6CostFunctionsandError-RejectTradeoff697
24.7TheReceiver-OperatorCharacteristic699
24.8MultipleClassifiers702
24.9ClusterAnalysis705
24.9.1SupervisedandUnsupervisedLearning705
24.9.2ClusteringProcedures706
24.10PrincipalComponentsAnalysis710
24.11TheRelevanceofProbabilityinImageAnalysis713
24.12TheRoutetoFaceRecognition715
24.12.1TheFaceasPartofa3-DObject716
24.13AnotherLookatStatisticalPatternRecognition:TheSupportVectorMachine719
24.14ConcludingRemarks720
24.15BibliographicalandHistoricalNotes722
24.16Problems723
CHAPTER25BiologicallyInspiredRecognitionSchemes
25.1Introduction725
25.2ArtificialNeuralNetworks726
25.3TheBackpropagationAlgorithm731
25.4MLPArchitectures735
25.5OverfittingtotheTrainingData736
25.6OptimizingtheNetworkArchitecture739
25.7HebbianLearning740
25.8CaseStudy:NoiseSuppressionUsingANNs745
25.9GeneticAlgorithms750
25.10ConcludingRemarks752
25.11BibliographicalandHistoricalNotes753
CHAPTER26Texture
26.1Introduction757
26.2SomeBasicApproachestoTextureAnalysis763
26.3Gray-levelCo-occurrenceMatrices764
26.4Laws’TextureEnergyApproach768
26.5Ade’sEigenfilterApproach771
26.6AppraisaloftheLawsandAdeApproaches772
26.7Fractal-basedMeasuresofTexture774
26.8ShapefromTexture775
26.9MarkovRandomFieldModelsofTexture776
26.10StructuralApproachestoTextureAnalysis777
26.11ConcludingRemarks777
26.12BibliographicalandHistoricalNotes778
CHAPTER27ImageAcquisition
27.1Introduction781
27.2IlluminationSchemes782
27.2.1EliminatingShadows784
27.2.2PrinciplesforProducingRegionsofUniformIllumination787
27.2.3CaseofTwoInfiniteParallelStripLights790
27.2.4OverviewoftheUniformIlluminationScenario793
27.2.5UseofLine-scanCameras794
27.3CamerasandDigitization796
27.3.1Digitization798
27.4TheSamplingTheorem798
27.5ConcludingRemarks802
27.6BibliographicalandHistoricalNotes803
CHAPTER28Real-timeHardwareandSystemsDesignConsiderations
28.1Introduction805
28.2ParallelProcessing806
28.3SIMDSystems807
28.4TheGaininSpeedAttainablewithNProcessors809
28.5Flynn’sClassification810
28.6OptimalImplementationofanImageAnalysisAlgorithm813
28.6.1HardwareSpecificationandDesign813
28.6.2BasicIdeasonOptimalHardwareImplementation814
28.7SomeUsefulReal-timeHardwareOptions816
28.8SystemsDesignConsiderations818
28.9DesignofInspectionSystems-TheStatusQuo818
28.10SystemOptimization822
28.11TheValueofCaseStudies824
28.12ConcludingRemarks825
28.13BibliographicalandHistoricalNotes827
28.13.1GeneralBackground827
28.13.2RecentHighlyRelevantWork829
PART5PERSPECTIVESONVISION831
CHAPTER29MachineVision:ArtorScience?
29.1Introduction833
29.2ParametersofImportanceinMachineVision834
29.3Tradeoffs836
29.3.1SomeImportantTradeoffs837
29.3.2TradeoffsforTwo-stageTemplateMatching838
29.4FutureDirections839
29.5Hardware,Algorithms,andProcesses840
29.6ARetrospectiveView841
29.7JustaGlimpseofVision?842
29.8BibliographicalandHistoricalNotes843
APPENDIXRobustStatistics
A.1Introduction845
A.2PreliminaryDefinitionsandAnalysis848
A.3TheM-estimator(InfluenceFunction)Approach850
A.4TheLeastMedianofSquaresApproachtoRegression856
A.5OverviewoftheRobustnessProblem860
A.6TheRANSACApproach861
A.7ConcludingRemarks863
A.8BibliographicalandHistoricalNotes864
A.9Problem865
ListofAcronymsandAbbreviations867
References869
AuthorIndex917
SubjectIndex925
……
