DocumentCode
2973263
Title
Study of the Point Element and Line Element Conic Curve Composite Algorithm
Author
Zheng-wu, Zhang
Author_Institution
Dept. of Mech. Eng., Shaanxi Univ. of Technol., Hanzhong, China
fYear
2010
fDate
25-27 June 2010
Firstpage
5621
Lastpage
5625
Abstract
The projective theory of conic curve is one of the most important part in projective geometry. The object recognition, scene reconstruction and motion analysis through corresponding conic curves are hot research areas in the computer vision. The conventional algorithm of point element and line element conic curve is to use several points or lines on conic curve. But in practice both of the points and lines on conic curve may be detected in computational processes, so the composite algorithm of conic curve must be established concerning with points and lines. The different bases of conic curves pencil through the pair of conic curves and conic curves pencil are proposed from the perspective of computational processes in this paper if all points and lines on the image plane are represented in terms of N-vectors, and the N-vectors of point element and line element conic curve composite algorithm is presented in the different conditions on the basis of study. The results of examples show that the algorithm is more utility and reliable than traditional algorithm.
Keywords
image motion analysis; object recognition; computer vision; line element conic curve composite algorithm; motion analysis; object recognition; projective theory; scene reconstruction; Algorithm design and analysis; Computer vision; Fitting; Image motion analysis; Image reconstruction; Object recognition; Optical variables measurement; base of conic curves pencil; composite algorithm; computer vision; point element and line element conic curve;
fLanguage
English
Publisher
ieee
Conference_Titel
Electrical and Control Engineering (ICECE), 2010 International Conference on
Conference_Location
Wuhan
Print_ISBN
978-1-4244-6880-5
Type
conf
DOI
10.1109/iCECE.2010.1365
Filename
5629533
Link To Document