Title :
Fusion of three techniques for person detection in real time
Author :
Boukamcha, Hamdi ; Mohamed, Amr
Author_Institution :
EME Lab., Univ. of Monastir, Monastir, Tunisia
Abstract :
The present paper is a new technique for the detection of persons, with combination of the three approaches (Adaptive Gaussian Mixture, Objects Classification, and Stereo Vision) with the integration of the multithreaded architecture. The study and validation at high-level of these descriptors was used to determinate the advantages and drawbacks of each approach. Several tests were made to evaluate the precision, execution time, complexity and sensitivity to different situations.
Keywords :
Gaussian processes; image classification; mixture models; multi-threading; object detection; stereo image processing; adaptive Gaussian mixture; multithreaded architecture; objects classification; person detection; real time; stereo vision; technique fusion; Cameras; Classification algorithms; Instruction sets; Mobile communication; Stereo vision; Streaming media; Three-dimensional displays; Multi-threading; Person Detection; Stereo Vision; background-difference; classification;
Conference_Titel :
Computer Applications and Information Systems (WCCAIS), 2014 World Congress on
Conference_Location :
Hammamet
Print_ISBN :
978-1-4799-3350-1
DOI :
10.1109/WCCAIS.2014.6916573
