Paper Info
Title
ODROID XU4 based implementation of decision level fusion approach for matching computer generated sketches.
Abstract
•A novel feature extraction method was developed to extract features from computer generated sketches and face photos using supervised auto-encoder which is optimized to build deep architecture.•A novel classification approach was developed to match computer generated sketches and face photos using Parallel Convolutional Neural Network.•A novel decision level fusion was developed get the optimal match between computer generated sketches and face photos.•Implementing the proposed novel approach was carried out on ODROID-XU4 which is one of the latest single board computing device having extremely high energy efficiency and low power consumption.•The developed system can be used for real-time operation because of the very small form factor of ODROID-XU4 when compared to any other ARM based embedded devices.
Year
DOI
Venue
2016
10.1016/j.jocs.2016.07.013
Journal of Computational Science
Keywords
Field
DocType
Single board computer,Supervised auto-encoder,Deep architecture,Parallel convolutional neural network
Viola–Jones object detection framework,Android (operating system),Optimal matching,Computer graphics (images),Convolutional neural network,Computer science,Theoretical computer science,Feature extraction,Gigabit Ethernet,Single-board computer,Computer hardware,USB
Journal
Volume
ISSN
Citations 
16
1877-7503
8
PageRank 
References 
Authors
0.82
26
2
Name
Order
Citations
PageRank
Steven L. Fernandes131423.39
G. Josemin Bala2518.42