%0 Journal Article
%T 基于主动轮廓模型的肺纹理自动提取新方法
%A 刘军伟
%A 冯焕清
%A 周颖玥
%A 李传富
%J 自动化学报
%P 345-349
%D 2009
%R 10.3724/SP.J.1004.2009.00345
%X ？Computedtomography(CT)istheprimaryimagingmodalityforinvestigationoflungfunctionandlungdiseases.HighresolutionCTsliceimagesofchestcontainlotsoftextureinformation,whichprovidespowerfuldatasetstoresearchcomputeraid-diagnosis(CAD)system.Buttheextractionoflungtissuetexturesisachallengetask.Inthispaper,weintroduceanovelmethodbasedonlevelsettoextractlungtissuetexturetree,whichisautomaticandeffectual.Firstly,weproposeanimprovedimplicitactivecontourmodeldrivenbylocalbinaryfittingenergy,andtheparametersaredynamicandmodulatedbyimagegradientinformation.Secondly,anewtechniqueofpaintingbackgroundbasedonintensitynonlinearmappingisbroughtforwardtoremovetheinfluenceofbackgroundduringtheevolutionofsinglelevelsetfunction.Atlast,anumberofcontrastexperimentsareperformed,andtheresultsof3Dsurfacereconstructionshowourmethodisefficientandpowerfulforthesegmentationoffinelungtreetexturestructures.
%K Activecontourmodel
%K levelset
%K localbinaryfittingenergy
%K dynamicparametermodulation
%K 3Dsurfacereconstruction
%U http://www.aas.net.cn/CN/abstract/abstract15805.shtml