A Novel pixel 8-neighborhood frame differencing algorithm for segmentation of moving object in video

Author

Sheena, C.V. ; Narayanan, N.K.

Author_Institution

Dept. of Inf. Technol., Kannur Univ., Kannur, India

fYear

2015

fDate

5-7 March 2015

Firstpage

1

Lastpage

4

Abstract

Segmentation of moving object from a video sequence is a fundamental and critical task in many computer vision applications. This paper present a simple method for the segmentation of the moving object in video sequences based on frame-differencing algorithm. The method consider neighborhood of each pixel in each frame for segmenting moving object. The experiment performed using KTH Action database. By visual inspection of segmented frame it is found that the proposed method is capable of removing impulse noise compared to conventional frame-differencing algorithm.

Keywords

image motion analysis; image segmentation; impulse noise; object tracking; KTH Action database; computer vision applications; frame-differencing algorithm; impulse noise; moving object segmentation; segmented frame; video sequence; visual inspection; Image color analysis; Image recognition; Image resolution; Image segmentation; Pixel 8-Neighbourhood; Segmentation; frame-differencing; thresholding;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical, Computer and Communication Technologies (ICECCT), 2015 IEEE International Conference on

Conference_Location

Coimbatore

Print_ISBN

978-1-4799-6084-2

Type

conf

DOI

10.1109/ICECCT.2015.7226021

Filename

7226021