TRAP@NCI

Improving real-time face detection and recognition techniques with the help of cloud computing in real time

Bora, Amit (2015) Improving real-time face detection and recognition techniques with the help of cloud computing in real time. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science) - Submitted Version
Download (1MB) | Preview

Abstract

Cloud computing is transforming the business of information technology by deploying, managing and running all software on cloud which provide reliable, scalable and cost-effective technology. Cloud computing is gaining much popularity nowadays. Many organizations are switching to cloud to fulfill user requests as fast as possible. Cloud computing gives advantages such as mobility, low cost to user. Devices like laptop have low processing power and storage. Shifting processing power and storage of device to cloud, we can overcome drawbacks of devices. Face recognition and detection is a wide area of research but it is hard to achieve in real time. Many police departments are using face recognition for criminal person detection. As many people are using internet service, face detection with cloud computing is one of the important application. Devices like low configuration desktops face many challenges in their resources as low computing power, and storage.

The thesis proposes a software methodology that helps to overcome the challenges of low computing hardware. The thesis investigates the use of cloud computing to improve face detection and face recognition technology. The openstack platform was used to evaluate the performance of three different algorithms (haar cascade classifier, LBP cascade classifier, and eigenface) on four different databases. The face detection and recognition software was able to detect and recognize faces successfully. To conclude, we can use the software to any database which contains different number of images and hence we promote the usage of cloud to get the result of face detection and face recognition in real-time.

Item Type: Thesis (Masters)
Subjects: T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions: School of Computing > Master of Science in Cloud Computing
Depositing User: CAOIMHE NI MHAICIN
Date Deposited: 11 Feb 2016 12:10
Last Modified: 11 Feb 2016 12:10
URI: http://trap.ncirl.ie/id/eprint/2121

Actions (login required)

View Item View Item