Abstract

Background subtraction is among the difficult activities in computer vision. Since the
nature of the environment may involve some changes due to light effect, dynamic background, shadow, camouflage effect, etc. Many moving object detection methods are proposed, but the majority of them fails to handle the multi-modality of scenes. Based on
the Artificial Immune Recognition System, efficient background subtraction methods are
proposed. In the first method, the Single Gaussian model is combined with the Artificial Immune Recognition System to better represent pixel variations in the scenes that
contain dynamic background. Artificial Immune Recognition System is used as a classification tool that separates antigens represented by the foreground pixels from antibodies
that modelled the background pixels. Each pixel in this proposition is modelled with
a feature vector contains Gaussian attributes. As a second contribution, we have used
the Artificial Immune Recognition System for managing the number of Gaussians dynamically in Gaussian Mixture Model instead to fix them a priori by the user. In this
contribution, a set of new Gaussians is generated using two different strategies: the first
one (Random generation) uses the Artificial Immune Recognition System for improving
the system decision, while in the second strategy (Directed generation), the Artificial
Immune Recognition System is used to improve the production of background models.
To reduce the effect of brightness, each frame in the video sequence is transformed from
the RGB to HSV color space. Artificial Immune Recognition System has also benefited
from some modifications to reduce research cost and to avoid the explosion of data in the
memory cells set. For reducing the research cost on the most representative memory cell
to the current antigen, the structure of the memory cells set is redefined as a binary tree
(kd-tree). Furthermore, we have added two mechanisms to the basic Artificial Immune
Recognition System to avoid the explosion of data in the memory cells set. The proposed
methods are implemented and tested on public datasets. The obtained results are largely
satisfactory compared to other state-of-the-art methods.

Information

Item Type:	Thesis
Divisions:
ePrint ID:	4429
Date Deposited:	2023-09-17
Further Information:	Google Scholar
URI:	https://www.univ-soukahras.dz/en/publication/article/4429

BibTex