图像分割的商空间粒度原理_刘仁金.pdfVIP

28 10 V ol . 28 N o. 10 2005 10 CH IN ESE JO U RN A L O F CO M PU T ER S Oct . 2005 1) 2) 3) 刘仁金 黄贤武 1) ( 2370 12) 2) ( 230039) 3) ( 2 15006) . (X , f , #) Z ( [ X] , [ f ] , [ #] ) , . . ; ; ; ; T P 391 The Granular Theorem of Quotient Space in Image Segmentation L IU Ren-Jin 1) 2) H U A N G X ian- Wu3) 1) (Dep artment of Comp uter S cience T echnolog y , West A nhui Uni er sity , L ucan 2370 12) 2) (K ey L abora tory of I ntelligent Comp ut ing Sig nal P rocessing of M inistr y of Ed ucat ion, A nhui Uni er sity , H ef ei 230039) 3) ( S chool of Electronic Inf or mation E ngineer ing , S ooch ow Uni e rsity , Suz hou 2 15006) Abstract Based on t he quo t ient sp ace g ranular t heorem t he image segmentat ion concept is ana- ly zed and t he imag e seg ment at ion m et hods are st u died and t hen t he quot ient sp ace granular t heo- r em of image segm ent ation is dem on st rat ed . T he image segm ent atio n problem s are described w it h tr iple elements funct ion o f t he quot ient sp ace model ( X ,f , #) Z ([ X ] , [f ] , [ #] ) according to the quot ient sp ace g ranularit y com puting t he im ag e segmentat ion t heorem is present ed. T he met h- o ds of imag es segm ent ation b ased on hierarchical and synt hesis and combinat ional t echnique