您好,欢迎访问三七文档
当前位置:首页 > 商业/管理/HR > 经营企划 > 含噪图像中类线像素的检测淘汰竞争和紧支撑平滑滤波恢复算法
©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.:025322239(2009)1223367207(,100081)(PR),,,,,,,,,,,;;;;;TN911.73Adoi:10.3788/AOS20092912.3367RestorationAlgorithmtoLineSimilarPixelsinNoisyImagethroughDetection,Washout,CompetitionandTightFrameSmoothingFilteringHaoBingShenTingzhiTianHui(SchoolofInformationScienceandTechnology,BeijingInstituteofTechnology,Beijing100081,China)AbstractAnewdirectionaloperatorstructureinuniformstandardformatispresented.FromreferencetomanmadeGame,throughdetection,washout,competitionandtightframesmoothingfiltering,anonorthogonal,nonperfectreconstructionandnonrecursiveanalysisandsynthesisfilteringmacroplanandmicroalgorithmareproposed.Withsimplifiedandoptimizedstrategyofcoarselydetectinglinesthendelicatelysearchingzigzags,andundergamerulesofminorityobeymajorityandweakerfollowstronger,thetaskofrestoringlinesimilarpixelsfromnoisyimageiscompletedsuccessfully.Varityofexperimentsaremadetochecktheeffectiveness,consistencyandreliabilityofproposedalgorithm,andsomewellresultsareobtained.Atfinal,somefaithfulobjectiveevaluateonexperimentalresultsofnewmethodismade,andthelimitationandshortcomingofitispointedout.Keywordsimageprocessing;directionaloperator;detection;washout,competition;tightframesmoothingfiltering:2009201207;:2009202225:(60772066):(1962),,,,E2mail:haobing@bit.edu.cn:(1943),,,,E2mail:shentz@bit.edu.cn1,,,(,),PR[112],©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(),(),,[1316],PR,,[1720],,,()Radon,,[2124][25,26],,,[2729],2(),,,,[31],,PR,,,,,,,()()33.13.1.1,33,CCD,,,(0,0),113(a);(b);(c)1;(d);(e)-1Fig.1Directionaloperatorswithsizeof3(a)Observationwindow;(b)horizontaloperator;(c)operatorwithtangent1;(d)verticaloperator;(e)operatorwithtangent-1,Dl,l=14,x1=[0,0,0],x2=[-1,0,1],x3=[-1,0,1],x4=[1,0,-1];(1)y1=[-1,0,1],y2=[-1,0,1],y3=[0,0,0],y4=[-1,0,1];(2)xl,yl,l=14lDl,1,,Zl=xl,yl,l=14(3),wl=13[1,1,1],l=14(4)8633©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(1)(4),,;3.2,,MM4NM(M),;N,Zn,n4N;wn,n4N,nDn=Zn,wn,n4N(5)Zn=s,tZns,t,wn=wn,(s,t)s,tZn,n4N.22Fig.2Sketchofalldirectionaloperators21,2,3,4k=0,k=1,k=,k=-1.M,M;0k1,,1,1k2,-k-13,-1k04;,wl=1M1,,1M,l=14N.,,,21k=1;43121N11,N+14,,,(2),,,,,,;,(),(),;(),,,;,1,,-1,04.14.1.1(D)D,IJ(i,j),X(i,j),n=14N4NMM(i,j;n),i,jI,J,n=14N(6)M(i,j;n)=s,tZnwn(s,t)X(i+s,j+t),i,jI,J,n=14N(7)4NMen,n=14N,Men=i,jI,JM(i,j;n),n=14N(8)9633©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(W)Men(i,j),,,Fak=Fak(i,j;n,k),k=1,0,n=14N,k=1,2,i,jI,J(9)k=1,2,=1Fa(i,j)n,=-1Fa(i,j)n,=0Fa(i,j)T,Fa1T(n),orFa2T(n),(10)(i,j)n,Fa=Fa1n(),Tn4.2(C),,,FaIJ(i,j),(i,j)nIJFan,;,FaFa=Fani,j;G(n,n)n1:4N,i,jI,J,n=1,0,(11)n(9)Fa,nn;G,G(n,n)=n,n,n=1,(n,n),n1,(12)nFa,Fa1n1,Fa,,:,(),,f=n(n,n)nFan.(13)f0,,;f0,,,,(i,j)(n1,n2),(n1,n2)=argmaxnl,npnFani,j;(nl,np),iff0,i,jI,J,(14)(n1,n2)=argminnl,npnFani,j;(nl,np),iff0,i,jI,J.(15)5(F),:,,();,;,,Fa,Fa=Fai,j;(ni1,ni2),ni1,ni214N,i,jI,J.(16)IJ,,hs(),0733©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(i,j)(ni1,ni2),hs(i,j;r,t)=h(i+r,j+t)(r,t)Zhs.(17)Fs,Fs(i,j)=(r,t)Zhsh(i+r,j+t)Xi+r,j+t,i,jI,J.(18)Fs66.1,M=7,N=4,7,16,15Z1=(0,-3),(0,-2),(0,-1),(0,0),(0,1),(0,2),(0,3),Z2=(-1,-3),(-1,-2),(0,-1),(0,0),(0,1),(1,2),(1,3),Z3=(-2,-3),(-1,-2),(-1,-1),(0,0),(0,1),(1,2),(1,3),Z4=(-2,-3),(-2,-2),(-1,-1),(0,0),(0,1),(1,2),(2,3),Z5=(-3,-3),(-2,-2),(-1,-1),(0,0),(1,1),(2,2),(3,3),0,1/3,1/2,2/3,1w=171,1,1,1,1,1,1.T=1.05n,hs=131,1,1,56.2,rice,circuit,tireliftingbody,0.1,0.150.2,,,363rice(a);,(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)(b),(c),(d)Fig.3Noisingandlinesimilarpixelsrestoringresultofrice.originalimage(a);noisingimage,noisedeviationis(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)restoredimageof(b),(c),(d)4rcircuit(a);,(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)(b),(c),(d)Fig.4Noisingandlinesimilarpixelsrestoringresultofcircuit.originalimage(a);noisingimage,noisedeviationis(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)restoredimageof(b),(c),(d)1733©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.(a);,(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)(d),(b),(c)Fig.5Noisingandlinesimilarpixelsrestoringresultoftire.originalimage(a);noisingimage,noisedeviationis(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)restoredimageof(b),(c),(d)6liftingbody(a);,(b)0.1;(c),(d)0.2;(e),(f),(g)(b),(c),(d)Fig.6Noisingandlinesimilarpixelsrestoringresultofliftingbody.originalimage(a);noisingimage,noisedeviationis(b)0.1;(c)0.15;(d)0.2;(e),(f),(g)restoredimageof(b),(c),(d)lena,0.1,77lena(a);(b);(c)Fig.7Noisingandlinesimilarpixelsrestoringresultoflena(a)originalimage;(b)noisingimage;(c)restoredimage,3777.1,,,;;(,),10,10;101018,2326,277.2,rice,0.10.15,0.2;circuit,;tirelifitingbody,;lenacameraman(0.1),:,()7.3,2;;(2)8,PR,,,,,,2733©1994-2010ChinaAcademicJournalElectronicPublishingHouse.Allrightsreserved.
本文标题:含噪图像中类线像素的检测淘汰竞争和紧支撑平滑滤波恢复算法
链接地址:https://www.777doc.com/doc-616656 .html