基于Matlab的同态滤波器图像增强

:1002-2082(2010)04-0584-05Matlab马　龙天,张成义(,210044)摘　要:介绍同态滤波器的基本原理,基于Matlab软件,设计了适用的同态滤波器模型,并给出了其表达式参数,在此基础上,将同态滤波方法应用到彩色图像的光照补偿和对比度增强中。结果表明:经处理后的图像,局部对比度增强效果明显,较好地保持了图像的原始面貌,取得了预期的理想滤波效果。基于Matlab的同态滤波,能更好地适合人眼的视觉特性,提供了图像分析与判读的一种新方法。关键词:同态滤波;光照补偿;图像对比度:TN911.7;TP391.41:AOptimaldesignofhomorphicfilterbasedonMatlabMAYan,ZHANGCheng-yi(DepartmentofPhysics,NanjingUniversityofInformationScience&Technology,Nanjing210044,China)Abstract:Thebasicprincipleofthehomomorphicfilterisintroduced.AhomomorphicfiltermodelandparameterexpressionwasestablishedbyusingMatlab.Onthisbasis,thehomomorphicfilteringmethodwasappliedtothebrightnessofcolorimageandimagecontrastenhancement.Itisshownthatthemethodperformswellinenhancingthelocalcontrastofanimagewhilemaintainingitsglobalappearance,andtheexpectedfiltereffectisachieved.Experimentsprovethatthemethodissuitableforhumanvisualobservationtoconductimageanalysisandjudge.Keywords:homomorphicfilter;illuminationcompensation;imagecontrast,,,,,,,,,Matlab,1,,,,,:2009-09-13;:2010-05-04:(BJG008021):(1988-),,,,:,E-mail:[e-mail]31420107JournalofAppliedOpticsVol.31No.4Jul.2010,,,f(x,y)f(x,y)2:2,i(x,y)r(x,y)f(x,y)f(x,y)=i(x,y)r(x,y)(1)(1),2F(f(x,y))F(i(x,y))F(r(x,y))r(x,y),,i(x,y),,,22,z(x,y)=lnf(x,y)=lni(x,y)+lnr(x,y)(2),F(z(x,y))=F(lni(x,y))+F(lnr(x,y))(3)Z(u,v)=Fi(u,v)+Fr(u,v)(4)(4),,,,2:,2,,,H(u,v)Z(u,v),(4):S(u,v)=H(u,v)Z(u,v)=H(u,v)Fi(u,v)+H(u,v)Fr(u,v)(5),H(u,v),,:s(x,y)=F-1(S(u,v))=F-1(H(u,v)Fi(u,v)+H(u,v)Fr(u,v))(6),i(x,y)=F-1(H(u,v)Fi(u,v))r(x,y)=F-1(H(u,v)Fr(u,v))(7)s(x,y)=i(x,y)+r(x,y),,:g(x,y)=es(x,y)=ei(x,y)er(x,y)=i0(x,y)r0(x,y)(8)(8),H(u,v)1(4),1Fig.1Processingofhomomorphicfilter,,,,,,2Rh,Rl,D(u,v)(u,v)(u0,v0)2Rh1,Rl1,,,,H(u,v)=1-exp[-D(u,v)/2D02n](9)(9)5852010,31(4),:Matlab,:H(u,v)=(Rh-Rl)[1-e-c(D(u,v)/D0)2n]+Rl(10)D(u,v)=[(u-M/2)2+(v-N/2)2]12(11):D0u0v00D(u,v),MN,(M/2,N/2),,(10)2Fig.2Classicalhomomorphicfilterfunction3,,,H(u,v)=1/(1+D0/D(u,v))2n(12),:H(u,v)=(Rh-Rl)/(1+cD0/D(u,v))2n+Rl(13)3Fig.3AcommonhomomorphicfilterfunctionRh1,Rl1,,,D0,c,,RhRl3D0,,,,,MNf(x,y):F(u,v)=1MNM-1x=0N-1y=0f(x,y)e-j2P(uxM+vyN)(14):u=0,1,2,,M-1;v=0,1,,N-1:ûF(u,v)û=[Re2(u,v)+Im2(u,v)]12(15),Re(u,v)Im(u,v)F(u,v)D0:D0=D-=1MNM-1u=0N-1v=0D(u,v)(16),44Fig.4ImprovedhomomorphicfilterfunctionD(u,v)=1/1+D(u,v)D2+1/c1+DD(u,v)2(17)H(u,v)=(Rh-Rl)1+D(u,v)D-2+c1+DD(u,v)2-1+Rl(18),,4,Matlab,(13)(18)5862010,31(4),:Matlab,51011Table1HomomorphicfiltervalueRhRlc1.20.50.82.40.52.25Fig.5Rawimage6Fig.6Originalhistogram7Fig.7ButterworthfilterMatlab,,579,,,,,(),,(6810),,,8Fig.8Butterworthhistogram9Fig.9Improvedhomomorphicfilter10Fig.10Improvedhistogram5872010,31(4),:Matlab5,,,Matlab,,,:[1],.[J].,2007,23(6):264-266.CHENChun-ning,WANGYan-jie.Imagecontrastenhancementbyhomomorphicfilteringinfrequencyfield[J].MicrocomputerInformation,2007,23(6):264-266.(inChinesewithanEnglishabstract)[2].MATLAB[J].,2000(6):104-107.HUANGJian-ling.AnalysesandprocessingoffigureimageusingMATLAB[J].ComputerandModerni-zation,2000(6):104-107.(inChinesewithanEnglishabstract)[3],.[J].,2005,31(5):726-728.WANGYu,LONGXing-wu.Reducenon-uniformilluminationwithwavelet[J].OpticalTechnique,2005,31(5):726-728.(inChinesewithanEnglishabstract)[4],,,.[J].,2005,21(26):147-148.YILi-min,LIUYan-ying,GURei,etal.Acontrollablehistogramequalizationarithmetic[J].MicrocomputerInformation,2005,21(26):147-148.(inChinesewithanEnglishabstract)[5],.[J].,2001,29(4):531-533.ZHANGXin-ming,SHENLan-sun.Imagecontrastenhancementbywaveletbasedhomomorphicfiltering[J].ActaElectronicaSinica,2001,29(4):531-533.(inChinesewithanEnglishabstract)[6],,,.[J].,2008,35(3):97-101.ZHAOYao-hong,SHIZe-lin,LUOHai-bo,etal.Self-adaptivehistogramequalizationenhancementalgorithmforinfraredimage[J].Opto-ElectronicEngineering,2008,35(3):97-101.(inChinesewithanEnglishabstract)[7],,,.[J].,2010,31(2):221-224.KUANGHai-peng,WANGDe-jiang,ZHANGJing-guo,etal.Aerialimagewavelettransformationdenoisingbasedonmediumpre-filtering[J].JournalofAppliedOptics.2010,31(2):221-224.(inChinesewithanEnglishabstract)5882010,31(4),:Matlab