Title :
3D Affine Moment Invariants for Surfaces
Author :
Xu, Dong ; Li, Hua
Author_Institution :
Nat. Res. Center for Intell. Comput. Syst., Chinese Acad. of Sci., Beijing
Abstract :
A new kind of weighted surface moments is defined which can achieve affine invariance for unclosed surfaces. We can also select a corresponding point and build a pseudo-solid for surface and compute the affine volume moment invariants. We test the invariance of the two affine moment invariants on a 3D terrain map and a 3D face scan. The experimental result reveals that the weighted surface moment invariants are highly robust to shape deformations and numerical computation.
Keywords :
image recognition; 3D affine moment invariants; 3D face scan; 3D terrain map; numerical computation; pseudo-solid; shape deformations; shape recognition; unclosed surfaces; weighted surface moment invariants; Biometrics; Computer vision; Intelligent systems; Object recognition; Remote sensing; Robustness; Shape; Solids; Surface fitting; Testing;
Conference_Titel :
Pattern Recognition, 2008. CCPR '08. Chinese Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-2316-3
DOI :
10.1109/CCPR.2008.26
