An adaptive morphological filter for moving object segmentation | [Journal of Theoretical and Applied Information Technology • 2020]

Author(s):

1. MUHAMMAD HAMEED SIDDIQI: College of Computer and Information Sciences, Jouf University,Sakaka, Aljouf, Kingdom of Saudi, Arabia

2. MADALLAH ALRUWAILI: College of Computer and Information Sciences, Jouf University,Sakaka, Aljouf, Kingdom of Saudi, Arabia

3. M. M. KAMRUZZAMAN: College of Computer and Information Sciences, Jouf University,Sakaka, Aljouf, Kingdom of Saudi, Arabia

4. SAAD ALANAZI: College of Computer and Information Sciences, Jouf University,Sakaka, Aljouf, Kingdom of Saudi, Arabia

5. SAID ELAIWAT: College of Computer and Information Sciences, Jouf University,Sakaka, Aljouf, Kingdom of Saudi, Arabia

6. FAYADH ALENEZI: Department of Electrical Engineering, College of Engineering, Jouf University, Sakaka, Aljouf, Kingdom of Saudi Arabia

Abstract:

This work presents a system that combined well-known morphological filters in order to find true moving regions from a sequence of images. For localizing the changed region, a block-based change segmentation process is proposed. This change region is naturally a coarse region and also contains some holes. To compensate this, we used an edge-based dilation to get an anisotropic expansion of the coarse image. Then the segmentation is generated using watershed algorithm. To prevent over segmentation, we used a specially weighted gradient image to achieve segmentation. Also, we removed some local minima from that gradient image. Finally, a fusion is applied on morphological filters. Furthermore, we calculated the coverage ratio of edge pixels of each segmented region. Comparing with a converge threshold, we determined whether the segmented region is truly belongs to a moving region or not. In the end, the experimental result demonstrated the validity of our proposed method.

Page(s): 2755-2762

DOI: DOI not available

Published: Journal: Journal of Theoretical and Applied Information Technology, Volume: 98, Issue: 14, Year: 2020

Keywords:

Morphological Filter

References:

References are not available for this document.

Citations

Citations are not available for this document.

Citations

Downloads

Views