Towards the design of smart video-surveillance system
Author | Beghdadi A. |
Author | Asim M. |
Author | Almaadeed N. |
Author | Qureshi M.A. |
Available date | 2019-11-03T11:47:38Z |
Publication Date | 2018 |
Publication Name | 2018 NASA/ESA Conference on Adaptive Hardware and Systems, AHS 2018 |
Resource | Scopus |
Abstract | Security and monitoring systems are increasingly demanding in terms of quality, reliability and flexibility especially those dedicated to video surveillance. The aim of this study is to identify some limiting factors in the existing video-surveillance systems and to propose a set of best practices for developing a smart platform for a security monitoring system incorporating advanced techniques for video processing and analysis. In this work, we focus on the effect of the video quality on the biometric part of the video-surveillance systems for public security. In such systems, face detection and recognition from video sequences acquired from surveillance cameras, are challenging tasks, due to presence of strong illumination variations, noise, and changes in facial expressions. In this paper, we mainly focus on the illumination issue occurred in video surveillance. The low light video data is processed using a perceptual based approach, namely multi-scale Retinex method, to improve the video quality, followed by face detection. The experimental results demonstrate significant performance improvement in face detection and recognition, by improving the illumination of video sequences over the unprocessed video data. � 2018 IEEE. |
Sponsor | This publication was made possible by NPRP grant # NPRP8-140-2-065 from the Qatar National Research Fund (a member of the Qatar Foundation). The statements made herein are solely the responsibility of the authors. |
Language | en |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Subject | enhancement face detection face recognition Retinex video surveillance |
Type | Conference Paper |
Pagination | 162 - 167 |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Computer Science & Engineering [2281 items ]