Real-time human activity recognition

Albukhary, N. and Mohd. Mustafah, Yasir (2017) Real-time human activity recognition. In: 6th International Conference on Mechatronics - ICOM'17, 8th–9th August 2017, Kuala Lumpur, Malaysia.

	PDF - Published Version Restricted to Repository staff only Download (843kB) | Request a copy
	PDF (SCOPUS) - Supplemental Material Restricted to Repository staff only Download (475kB) | Request a copy

Abstract

The traditional Closed-circuit Television (CCTV) system requires human to monitor the CCTV for 24/7 which is inefficient and costly. Therefore, there’s a need for a system which can recognize human activity effectively in real-time. This paper concentrates on recognizing simple activity such as walking, running, sitting, standing and landing by using image processing techniques. Firstly, object detection is done by using background subtraction to detect moving object. Then, object tracking and object classification are constructed so that different person can be differentiated by using feature detection. Geometrical attributes of tracked object, which are centroid and aspect ratio of identified tracked are manipulated so that simple activity can be detected.

Item Type:	Conference or Workshop Item (Plenary Papers)
Additional Information:	5107/62903
Uncontrolled Keywords:	Background subtraction, Closed circuit television, Feature detection, Human activities, Human activity recognition, Image processing technique, Object classification, Object Tracking
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Mechatronics Engineering
Depositing User:	Dr Yasir Mohd Mustafah
Date Deposited:	21 Mar 2018 15:41
Last Modified:	26 Jun 2018 16:09
URI:	http://irep.iium.edu.my/id/eprint/62903

Actions (login required)

View Item