Real Time Foreground Segmentation for Video Sequences with Dynamic Background

Modeling background and segmentation of foreground objects are techniques required for various video surveillance and related applications. We propose an intensity and colour based background modeling technique for pixel level foreground segmentation of video sequences with dynamic background. It mo...

Full description

Saved in:
Bibliographic Details
Main Authors: Baser, Manit, Mittal, Miloni, Samaiya, Devesh
Format: Conference Proceeding
Language:English
Subjects:
Adaptation models
Dynamic Background
Foreground Segmentation
Image Processing
Real-time
Real-time systems
Streaming media
Surveillance
Training
Tuning
Video Processing
Video sequences
Video surveillance
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Modeling background and segmentation of foreground objects are techniques required for various video surveillance and related applications. We propose an intensity and colour based background modeling technique for pixel level foreground segmentation of video sequences with dynamic background. It models the background using initial few frames called training frames and calculates a threshold using statistical analysis on these frames. As the new frames appear in real time, the value of threshold is updated and the background model adapts to the changing background with each incoming frame. Quantitative results suggest that the proposed method can handle the sequences with highly dynamic background.
ISSN:2325-9418
DOI:10.1109/INDICON49873.2020.9342267