An improved algorithm for moving object tracking based on frame and edge differences.

M. F.  Sarwar; A.  Basit; H. Z.  Ahmad

No Detail Found!

An improved algorithm for moving object tracking based on frame and edge differences.

Author(s)

M. F. Sarwar Department of Electronic Engineering, International Islamic University, Islamabad, Pakistan

A. Basit TPD, Pakistan Institute of Nuclear Science and Technology (P1NSTECH), Nilore, Islamabad, Pakistan

H. Z. Ahmad Department of Computer Sciences, Bahria University, Islamabad, Pakistan

Abstract

Object tracking is gaining interest of researchers in the field of image processing and computer vision. Many methods have been proposed by the researchers in this field. In this paper, an improved algorithm for moving object tracking is proposed based on the frame difference and edge difference methods. Canny edge detector is applied to detect edges of current and previous frames and to get the difference of both edge images. Afterwards, the simple frame difference method is applied on both frames then the result is combined with the resulting image obtained after edge difference. Improved Otsu method is used to threshold the image and morphological filtering is applied to remove noise. Subsequently connectivity analysis is carried out to obtain the moving objects. This algorithm takes advantage of both frame difference and edge difference methods to improve the accuracy in detecting the moving objects. Experiments are performed on various videos which show efficient results in very short time.

Publication Details

Page(s) 9-13

DOI DOI not available

Published Journal: Nucleus, Volume: 52, Issue: 1, Year: 2015

Request the full text from PASTIC