High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

A Novel Approach to Detect Foreground in Video Sequences Based on Mixture of Gaussians


Navneet S. Ghedia , Sanjaybhai Rajguru College of Engineering,Rajkot; Prof. Dr. C.H. Vithalani, Government Engineering College,Rajkot; Prof. Dr. Kiran Parmar, Retired Professor- GEC gandhinagar


video surveillance and monitoring, adaptive Gaussian mixture model, background model, Gaussian mixture density, foreground detection


Our aim is to develop a robust visual monitoring system that passively detects moving objects in a specified space and identify the activities of those objects. In the paper, we focus on foreground detection. Mixture of Gaussian is a very well know and friendly approach for background modelling to detect moving objects. Our method improves Gaussian mixture model by continuously updating the mixture parameters. It gives faster update and a smoother object mask. The Gaussian mixture model approach consists of different Gaussian distributions, mean, standard deviation, weight. So, focus is to develop a robust visual monitoring system which can work successfully against illumination variations, clutter background, slow moving objects, sawing trees and objects being introduced or removed from the frame.

Other Details

Paper ID: IJSRDV3I60636
Published in: Volume : 3, Issue : 6
Publication Date: 01/09/2015
Page(s): 1220-1224

Article Preview

Download Article