Paper
17 May 2006 Global motion estimation and target detection with region search
Author Affiliations +
Abstract
This paper proposes a global motion model estimation and target detection algorithm for surveillance and tracking applications. The proposed algorithm analyzes the foreground-background structure of a video frame, and detects objects with independent motions. Each video frame is first segmented into regions where image intensity and motion fields are homogeneous. Then global motion model fitting is accomplished using linear regression of motion vectors through iterations of region search. With the use of non-parametric estimation of motion field, the proposed methods is more efficient than direct estimation of motion parameter; and it is able to detect outliers where independent moving targets are located. The proposed algorithm is more computationally efficient than parametric motion estimation, and also more robust than a variety of background compensation based detection.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lei Ma, Jennie Si, and Glen P. Abousleman "Global motion estimation and target detection with region search", Proc. SPIE 6235, Signal Processing, Sensor Fusion, and Target Recognition XV, 623506 (17 May 2006); https://doi.org/10.1117/12.665801
Lens.org Logo
CITATIONS
Cited by 1 scholarly publication.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Motion estimation

Target detection

Motion models

Detection and tracking algorithms

Cameras

Image segmentation

Video

RELATED CONTENT

Collaborative tracking of objects in EPTZ cameras
Proceedings of SPIE (January 29 2007)
Robust automatic camera pointing for airborne surveillance
Proceedings of SPIE (August 14 2002)
Object tracking for video annotation
Proceedings of SPIE (November 02 2004)
A new method for target detection based on analysis of...
Proceedings of SPIE (October 30 2009)

Back to Top