1、AgendaAgendaFoundationoCh14 Imaging System,Digitalization,Display,SoftwareoCh58 Histogram,Point Operations,Algebraic Operations,Geometric OperationsTheoryoCh912 Linear System,Fourier Frequency Transform,Filter Design,Discrete SamplingoCh1315 Orthogonal Radicle Transform,Wavelet Time-frequency Transf
2、orm,Optical Function TransformApplicationoCh1620 Image Restoration,Compression,Pattern RecognitionoCh2122 Color and Multi-Spectral Image Processing,Three Dimension Image Processing),(),(2yxgyxfED)(DR1DD)()(1DRDR)2log(21),(),(eDyxgyxfHDDless than D+xXX1TTW：noise coming from bit distributionA：filter0D
3、R(D)drentropy codingrate distortion functionThe distortion between The distortion between original image f(x,y)and original image f(x,y)and reconstruction imagereconstruction image g(x,y)is quantified by g(x,y)is quantified by the mean square error:the mean square error:2(,)(,)DEf x yg x ydoes it ex
4、ist such transform does it exist such transform T,and at the same time D is T,and at the same time D is minimumminimumminimum distortion minimum distortion transform coding is the transform coding is the most efficient at given most efficient at given distortion rate by rate distortion rate by rate
5、distortion functiondistortion function第8页（共17页）AgendaAgendaFoundationoCh14 Imaging System,Digitalization,Display,SoftwareoCh58 Histogram,Point Operations,Algebraic Operations,Geometric OperationsTheoryoCh912 Linear System,Fourier Frequency Transform,Filter Design,Discrete SamplingoCh1315 Orthogonal
6、Radicle Transform,Wavelet Time-frequency Transform,Optical Function TransformApplicationoCh1620 Image Restoration,Compression,Pattern RecognitionoCh2122 Color and Multi-Spectral Image Processing,Three Dimension Image ProcessingChapter 18 pattern recognitionimage segmentationinput imageobject typefea
7、ture vectorObject imageimagesegmentationfeatureextractionclassification“Bar”the three phases of pattern recognitionnxxx.21stepfunction1.object locator design(Chapt.18)Select the scene segmentation algorithm that will isolatethe individual objects in the image2.feature selection(Chapt.19)decide which
8、 properties of the objects are best to distinguish the object types and how to measure them3.classifier design(Chapt.20)establish the mathematical basis of the classification algorithm,and select the structure type of classifiers to be used 4.classifier training(Chapt.20)fix the various adjustable p
9、arameters in the classifier to suit the classified objects5.p e r f o r m a n c e evaluation(Chapt.20)estimate the expected error rates of the various possible misclassification0,0Dr02)(22rrrrrArrA2)(2lim/lim/lim)(000rBdrdrrDrArDrADHpDDDB1 21()()BTR THD dD1122111()()()()2BTR TP TA THD dDthe histogra
10、m and the profileDBr()Ap drrr()()()BDp DBp DAHD average boundary gradient22222(,)(,)(,)xf x yf x yf x yy21 22(,)(,)(1,1)(1,)(,1)g x yf x yf xyf xyf x y101202101121000121111000111101101101structuring element S：111111111111111111 101101111011011101structuring element Sxy imageDBS()B SBSS()B SBSS00001111100000000123450000000054321000000001232100003214 0567The boundary direction code1-41-31-52-21-22-11-11-61-81-71-9Line.NO100101102103104105106107sample210Object line segments
