Title :
The Image Stabilization Based on the Line Characters
Author :
Xin, Chen ; Shuhong, Jiao
Author_Institution :
Inf. & Commun. Eng. Coll., Harbin Eng. Univ., Harbin, China
fDate :
March 31 2009-April 2 2009
Abstract :
A video stabilization algorithm based on the line characteristic is proposed. The rotation and excursion parameters between two images will be measured with the Hough transform technology as well as phase correlation technology. In order to eliminate the influence to the global situation which was born by the partial movement, the method with the whole line characters is adopted to get the movement parameter. The Bayesian estimation uses minimum mean square error rule, and it is adopted to get the relative motion parameters between two images based on the proposition of the movement parameter space model. And then the compensation of the movement can be gotten. The conclusion illuminates that this method can remove the high frequency vibration and retains camera´s active movement ideally.
Keywords :
Bayes methods; Hough transforms; image sequences; matrix algebra; mean square error methods; motion estimation; object detection; Bayesian estimation; Hough transform technology; camera active movement; high frequency vibration; line characters; minimum mean square error rule; movement parameter space model; phase correlation technology; video stabilization algorithm; Bayesian methods; Computer science; Educational institutions; Image edge detection; Laplace equations; Motion estimation; Optical sensors; Parameter estimation; Phase measurement; Rotation measurement; Bayesian estimation; Hough transform; Line characteristic; Phase correlation; Video stabilization;
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
DOI :
10.1109/CSIE.2009.639