欢迎来到天天文库
浏览记录
ID:51813814
大小:816.56 KB
页数:9页
时间:2020-03-16
《可变阶次分数阶微分实现图像自适应增强.pdf》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库。
1、第34卷第4期沈阳工业大学学报Vol.34No.42012年7月JournalofShenyangUniversityofTechnologyJul.2012文章编号:1000-1646(2012)04-0446-09*可变阶次分数阶微分实现图像自适应增强1,2a2a2b11黄果,陈庆利,许黎,蒲亦非,周激流(1.四川大学计算机学院,成都610064;2.乐山师范学院a.智能信息处理及应用实验室,b.物电学院,四川乐山614000)摘要:为了改善图像增强的效果,提出了一种可变阶次的分数阶微分图像自适应增强算子.该算子的数学原理是根据图像的局部统计信息和结构特征来动态调整分数阶微分的
2、阶次.在图像增强的过程中,分数阶微分的阶次满足在图像“强边缘”处具有较大的微分阶次,在图像“弱边缘”或纹理处具有较小的微分阶次,在图像平滑处具有微小的微分阶次,在判断为噪声处具有负阶次.建立了图像增强效果和分数阶微分阶次之间的非线性量化关系,构造了相应的自适应函数,实现了图像的自适应增强.实验结果表明,在无噪声的情况下,该方法较传统的分数阶微分算子在图像增强方面效果更好;在有噪声的情况下,该方法具有一定的噪声免疫能力.关键词:自适应;分数阶微分;分数阶积分;图像增强;图像去噪;纹理增强;分数阶Fourier变换;信噪比中图分类号:TN391文献标志码:ARealizationofa
3、daptiveimageenhancementwithvariablefractionalorderdifferential1,2a2a2b11HUANGGuo,CHENQing-li,XULi,PUYi-fei,ZHOUJi-liu(1.SchoolofComputerScience,SichuanUniversity,Chengdu610064,China;2a.LaboratoryofIntelligentInformationProcessingandApplication,2b.SchoolofPhysicsandElectronics,LeshanNormalUnive
4、rsity,Leshan614000,China)Abstract:Inordertoimprovetheeffectofimageenhancement,anadaptiveimageenhancementoperatorwithvariablefractionalorderdifferentialwasproposed.Themathematicaltheoryoftheoperatorwasthattheorderoffractionalorderdifferentialwasdynamicallyadjustedaccordingtothelocalstatisticali
5、nformationandstructuralfeaturesofimages.Intheprocessofimageenhancement,theorderoffractionalorderdifferentialshouldbelargerinthestrongedgeregionoftheimage,smallerintheweakedgeortextureregionsoftheimage,andjerkwaterinthesmoothregionoftheimage.Inparticular,theorderoffractionalorderdifferentialsho
6、uldbenegativeinthejudgednoiseregion.Thenonlinearquantitativerelationshipbetweenimageenhancementeffectandorderoffractionalorderdifferentialwasestablished,andthecorrespondingadaptivefunctionwasconstructedtorealizetheadaptiveenhancementofimage.Theexperimentalresultsshowthattheproposedmethodexhibi
7、tsbetterimageenhancementeffectcomparedwiththetraditionalfractionalorderdifferentialoperatorinthenoise-freecase,andhascertainnoiseimmunityinthenoisecase.Keywords:adaptation;fractionalorderdifferential;fractionalorderintegral;imageenhance
此文档下载收益归作者所有