Title :
Implementation of object recognition and tracking algorithm on real-time basis
Author :
Jae Chang Kwak ; Tae Ryong Park ; Yong Seo Koo ; Kwang Yeob Lee
Author_Institution :
Dept. of Comput. Sci., Seo Kyeong Univ., Seoul, South Korea
Abstract :
This paper is intended to recognize and track objects on real time basis under the embedded environment. Speed and rate of recognition have been improved by strong directional SURF and optical flow, but its speed is not fast enough for the embedded system. In this paper, recognition and tracking objects by integration of ORB and optical flow with a parallelization using OpenMP are implemented. The result shows an improved processing speed.
Keywords :
feature extraction; object recognition; object tracking; ORB; OpenMP; Speeded Up Robust Feature; directional SURF; embedded system; object recognition; object tracking algorithm; optical flow; real time basis; Computer vision; Image motion analysis; Integrated optics; Object recognition; Optical imaging; Parallel processing; Real-time systems; BRIEF; ORB; OpenMP; SURF; optical flow;
Conference_Titel :
EUROCON, 2013 IEEE
Conference_Location :
Zagreb
Print_ISBN :
978-1-4673-2230-0
DOI :
10.1109/EUROCON.2013.6625254
