Moving object segmentation for video surveillance and conferencing applications

Video surveillance and conferencing systems have an impressive spread both for their practical application and interest as research issue. The common approach used in such systems consists of a good segmentation of moving objects. This paper presents an algorithm for segmenting and extracting moving...

Full description

Saved in:
Bibliographic Details
Main Authors: Alsaqre, F.E., Yuan Baozong
Format: Conference Proceeding
Language:English
Subjects:
Change detection algorithms
Computerized monitoring
Image edge detection
Image segmentation
Object detection
Object segmentation
Shape
Video sequences
Video surveillance
Videoconference
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Video surveillance and conferencing systems have an impressive spread both for their practical application and interest as research issue. The common approach used in such systems consists of a good segmentation of moving objects. This paper presents an algorithm for segmenting and extracting moving objects suitable for surveillance and video conferencing applications, where a still background frame can be captured beforehand. Since edge detection is often used to extract accurate boundaries of the objects in the scene, the first step in our algorithm is accomplished by combining two kinds of edge points which are detected from the frame difference and the background subtraction. After removing edge points that belong to the background frame, the resulting moving edge map is fed to the object extracting step. A fundamental task in this step is to declare the candidates of the moving object, followed by applying morphological closing and opening operations. The algorithm is implemented on real video sequences and good segmentation performance is achieved.
DOI:10.1109/ICCT.2003.1209889