Title :
A Point Sets Smoothing Algorithm Using Curvature Flow
Author :
Cao, Xiao-ye ; Wang, Zhi-Yan ; Zhang, Yan-qing
Author_Institution :
Coll. of Comput. Sci. & Technol., South China Univ. of Technol., Guangzhou
Abstract :
We present a method for anisotropic fairing of point sets, which based on the covariance analysis and the efficiently constructed curvature. We use principal curvatures to detect features such as edges. Anisotropic curvature flow allows prominent surface features and other details of the objects to be treated in different ways. It is able to retain and even enhance important features such as edges and corners on the surface. We define such a diffusion flow which allows the point moving along the normal with a speed equal to the curvature. Since the point is moved along the normal direction, the proposed curvature flow will not introduce undesirable points drifting over the surface.
Keywords :
covariance analysis; edge detection; feature extraction; image denoising; image enhancement; smoothing methods; stereo image processing; anisotropic curvature flow; anisotropic point set fairing; covariance analysis; diffusion flow; edge detection; feature detection; feature enhancement; point set smoothing algorithm; principal curvature; surface features; Anisotropic magnetoresistance; Geometry; Least squares approximation; Multilevel systems; Noise reduction; Noise shaping; Signal processing algorithms; Smoothing methods; Surface fitting; Surface treatment;
Conference_Titel :
Wireless Communications, Networking and Mobile Computing, 2007. WiCom 2007. International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-1311-9
DOI :
10.1109/WICOM.2007.734