Detection of Deformable Structures in Video by Polynomial Fitting Using an Efficient Hough Transform

Athwal, Cham (2012) Detection of Deformable Structures in Video by Polynomial Fitting Using an Efficient Hough Transform. In: Convergence and Hybrid Information Technology. Lecture Notes in Computer Science, 7425 . Springer Berlin Heidelberg, pp. 396-403. ISBN 978-3-642-32644-8

Full text not available from this repository.

Abstract

This paper describes a novel formulation of the Hough Transform technique to detect lines that can be approximated by polynomial curves between given end-points in bitmap images. The application domain is the analysis of the deformation of exible linear structures throughout a sequence of video frames. In many applications of this type it is possible to manually or semi-automatically identify two end points to the exible structure which then remains static or can be separately tracked throughout the video sequence. The example discussed in this study is the motion of belts between pulleys. The Hough Transform is used for the parameters of polynomial expressions based on combinations of simple parabolic and cubic curves. We demonstrate that these curves are rich enough to represent those that are typically found in the application domains considered. We present mathematical and algorithmic representations that enable intuitive and efficient computation of the Hough Transforms.

Item Type: Book Section
Additional Information: Submitted to REF 2014, UoA 11, Cham Athwal
Subjects: G400 Computer Science
Divisions: Faculty of Computing, Engineering and the Built Environment
Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology
Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology > Digital Media Technology
UoA Collections > UoA11: Computer Science and Informatics
Depositing User: Miss Jessica Baylis
Date Deposited: 06 Jun 2016 09:26
Last Modified: 06 Jun 2016 09:42
URI: http://www.open-access.bcu.ac.uk/id/eprint/238

Actions (login required)

View Item View Item

Research

In this section...