Title :
Human walker´s detection and localization
Author :
Ammar Cherif, Boudour ; Rokbani, Nizar ; Alimi, Adel M. ; Ben Khelifa, Mohamed Moncef
Author_Institution :
Dept. of Comput. Sci., Univ. of Sfax, Sfax, Tunisia
Abstract :
Vision based human robot interaction (HRI) is one of main interest in our research. The aim of this paper is the detection and localization of a human walker. Our system detects the human walking by tracking the skin, locating the head and the lower part of the body and determination of joints motion. We combine two approaches: skin colour detection and leg motion by the gradient method for gesture recognition. Experimental results have been extracted for a set of sequences in different environments.
Keywords :
gesture recognition; gradient methods; human-robot interaction; image colour analysis; image motion analysis; object detection; optical tracking; robot vision; body location; gesture recognition; gradient method; head location; human walker detection; human walker localization; human walking; joints motion; leg motion; skin colour detection; skin tracking; vision based human robot interaction; Detectors; Human robot interaction; Intelligent robots; Machine vision; Motion analysis; Motion detection; Object detection; Robot vision systems; Skin; Tracking;
Conference_Titel :
Signals, Circuits and Systems (SCS), 2009 3rd International Conference on
Conference_Location :
Medenine
Print_ISBN :
978-1-4244-4397-0
Electronic_ISBN :
978-1-4244-4398-7
DOI :
10.1109/ICSCS.2009.5412217
