DocumentCode
477909
Title
Methods for Acquiring Geometrical Properties in an Image Employed for Determining Image Stereo-Correspondence
Author
Koprowski, R. ; Wrobel, Zofia
Author_Institution
Dept. of Biomed. Comput. Syst., Silesia Univ., Sosnowiec
Volume
4
fYear
2008
fDate
18-20 Oct. 2008
Firstpage
68
Lastpage
71
Abstract
In the paper the evaluation of algorithms for determining graph vertices on the basis of binary images is presented. The covered algorithms employ the classic contour detection method as well as the Hough transform and the MPPM (minimum perimeter polygon method). The presented algorithms are the basis for image matching and determining the stereo correspondence of images. Stereo-pairs were acquired from a system for diagnosing faulty posture using the photogrammetric method. The algorithm operation is analyzed for the image matching technique that uses the optimal clique finding method.
Keywords
Hough transforms; edge detection; geometry; graph theory; image matching; pose estimation; stereo image processing; Hough transform; binary images; contour detection method; faulty posture diagnosing; geometrical properties; graph vertices; image matching; image stereo-correspondence; minimum perimeter polygon method; optimal clique finding method; photogrammetric method; Algorithm design and analysis; Computer science; Fuzzy systems; Image analysis; Image matching; Materials science and technology; Pattern analysis; Pattern matching; Shape; Zirconium;
fLanguage
English
Publisher
ieee
Conference_Titel
Fuzzy Systems and Knowledge Discovery, 2008. FSKD '08. Fifth International Conference on
Conference_Location
Jinan Shandong
Print_ISBN
978-0-7695-3305-6
Type
conf
DOI
10.1109/FSKD.2008.553
Filename
4666358
Link To Document