DESIGN FRAMEWORK FOR INTERNET OF THINGS BASED NEXT GENERATION VIDEO SURVEILLANCE
Date
2018-01-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
Type
Thesis
Degree Level
Masters
Abstract
Modern artificial intelligence and machine learning opens up new era towards video
surveillance system. Next generation video surveillance in Internet of Things (IoT) environment is
an emerging research area because of high bandwidth, big-data generation, resource constraint
video surveillance node, high energy consumption for real time applications. In this thesis, various
opportunities and functional requirements that next generation video surveillance system should
achieve with the power of video analytics, artificial intelligence and machine learning are
discussed. This thesis also proposes a new video surveillance system architecture introducing fog
computing towards IoT based system and contributes the facilities and benefits of proposed system
which can meet the forthcoming requirements of surveillance. Different challenges and issues
faced for video surveillance in IoT environment and evaluate fog-cloud integrated architecture to
penetrate and eliminate those issues.
The focus of this thesis is to evaluate the IoT based video surveillance system. To this end,
two case studies were performed to penetrate values towards energy and bandwidth efficient video
surveillance system. In one case study, an IoT-based power efficient color frame transmission and
generation algorithm for video surveillance application is presented. The conventional way is to
transmit all R, G and B components of all frames. Using proposed technique, instead of sending
all components, first one color frame is sent followed by a series of gray-scale frames. After a
certain number of gray-scale frames, another color frame is sent followed by the same number of
gray-scale frames. This process is repeated for video surveillance system. In the decoder, color
information is formulated from the color frame and then used to colorize the gray-scale frames. In
another case study, a bandwidth efficient and low complexity frame reproduction technique that is
also applicable in IoT based video surveillance application is presented. Using the second
technique, only the pixel intensity that differs heavily comparing to previous frame’s
corresponding pixel is sent. If the pixel intensity is similar or near similar comparing to the
previous frame, the information is not transferred. With this objective, the bit stream is created for
every frame with a predefined protocol. In cloud side, the frame information can be reproduced by
implementing the reverse protocol from the bit stream.
Experimental results of the two case studies show that the IoT-based proposed approach
gives better results than traditional techniques in terms of both energy efficiency and quality of the video, and therefore, can enable sensor nodes in IoT to perform more operations with energy
constraints.
Description
Keywords
Video Surveillance, Internet of Things, Fog Computing
Citation
Degree
Master of Science (M.Sc.)
Department
Electrical and Computer Engineering
Program
Electrical Engineering