基于Grassmann流形的仿射不变形状识别.pdfVIP

2012 2  ACTA AUTOMATICA SINICA February, 2012 Ä u Grassmann 6/ •  ØC/G£ O 4 + 1, 2, 3, 4 o2• 5 ¤L 1, 2, 3 Á ‡ D Ú  Kendall /G mn Ø · ^ u q C † , , ¤ § ¥ I u ) A C /3  õ / žA T^ • C †5Ǒ x . Ä u Grassmann 6/nØ ,  © • ØC/G mš5A , J Ñ Ä u Grassmann 6/ • Ø C /G £ OŽ { . Ž { Ä é Ô ö8 ¥ za /G © O OŽ /G Š   , 3 /G Š N  m ï õC þ   ©Ù ;  , ŠÿÁ /G ÿ  /G .ÿÁ /G Œq,a , é /G1 “d ©a . MPEG 7 /G ê ¢ L ², †D Ú Kendall /G © ¥Ä u Procrustean Ý þ£OŽ { ,  £OŽ {k ²` ³ ; ý ¢| ¥  I £ O ˜ Ú L ²,  Ž {é • C /k  · A å , 3 E,| ± p  V ÇE £ Ñ Ia O . … /G £O , Grassmann 6/ , • ØC , /G m, /G Š DOI 10.3724/SP.J.1004.2012.00248 Affine-invariant Shape Recognition Using Grassmann Manifold LIU Yun-Peng1, 2, 3, 4 LI Guang-Wei5 SHI Ze-Lin1, 2, 3 Abstract Traditional Kendall shape space theory is only applied to similar transform. However, geometric transforms of the object in the imaging process should be represented by affine transform at most situations. We analyze the nonlinear geometry structure of the affine invariant shape space and propose an affine-invariant shape recognition algorithm based on Grassmann manifold geometry. Firstly, we compute the mean shape and covariance for every shape class in the train sets. Then, we construct their norm probability models on the tangent space at each mean shape. Finally, we compute the maximum likelihood class according to the measured object and prior learned shape models. We use the proposed algorithm to recognize shapes in standard shape dataset and real images. Experiment r