欢迎来到天天文库
浏览记录
ID:31988738
大小:5.37 MB
页数:89页
时间:2019-01-30
《进化多目标优化的稀疏重构方法-研究》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、万方数据AbstractCompressedsensing(CS)isanemergingresearchtopic.CSgenerallyincludesthreeaspects:thesparserepresentationofthesignal,theobservationmatrixandreconstructionalgorithm.ForCS,thesparserepresentationofthesignalisthebasis,designingtheobservationmatrixisthekeypoint,andreconstructionalgorithmi
2、susedforreconstructingtheimage.Themainobjectiveofthispaperistodesignagoodcompressedsensingimagereconstructionalgorithm.Atpresent,therearemanyreconstructionalgorithmssuchasgreedyalgorithm,convexoptimizationmethod.Forthefisttime,weputthethinkingofmulti-objectiveoptimizationevolutionaryalgorithmi
3、ntothecompressedsensingimagereconstruction,proposeevolutionarymulti-objectiveoptimizationbasedsparsereconstructionframework,andprovideadetailedalgorithmframeworkanddescriptioncombiningwithimage’swaveletstructuralinformationatthesametime.Themaininnovationofthispaperisasfollows:(1)Weproposeanevo
4、lutionarymulti-objectiveoptimizationbasedsparsereconstructionmethodandprovideadetailedalgorithmframework.Inordertosolveactualproblems,wedesignmulti-objectiveevolutionarycoding,consistentmutationoperatorandlocation-informationbasediterativehardthresholdingalgorithm.Wealsoverifythefeasibilityand
5、superiorityofthealgorithminthewaveletdomainbyaseriesofcomparativeexperiments.(2)Weproposeanewblockingandscatteringobservingway.Inthewaveletdomain,someblocksmaynotbesparseusingthetraditionalblockdividingapproach.Thenewblockingandscatteringobservingwayweproposesolvesthisproblembyallocatingthespa
6、rsitypressureoftheblockswhicharenotsparsetootherblocks.Weproposeorthogonalmatchingpursuitalgorithmanditerativehardthresholdingalgorithmbasedonthenewblockingandscatteringobservingway.Theeffectivenessandsuperiorityofthenewobservingwayandtheproposedalgorithmsareverifiedbyexperiments.(3)Weproposea
7、non-convexcompressedsensingimagereconstructionalgorithmbasedontheblockingandscatteringobservingwayandevolutionarymulti-objectiveoptimization,whichcombinestheadvantagesofboth.Wealsoprovideadetailedalgorithmframeworkanddescription.Because
此文档下载收益归作者所有