MRF and multiagent system based approach for image segmentation

Kamal E. Melkemi, Mohamed Batouche, Sebti Foufou

Research output: Contribution to conferencePaper

Abstract

Simulated Annealing (SA) and Iterated Conditional Modes (ICM) are two of the Markov Random Fields (MRF) model based appi'oaches for image segmentation. In practice, the ICM provides reasonable segmentations compared to the SA and was the most robust in most cases. However, the ICM strongly depends on the initialization phase. In this work, we develop a new approach for image segmentation based on Multiagent System (MAS) in order to produce good segmentations. We consider a set of segmentation agents and a coordinator agent. Each segmentation agent is able to segment the image by ICM starting from its own initialization. However, the coordinator agent diversifies the initial configurations using crossover and mutation operators known in the Genetic Algorithms (GAs). We can consider this model as a hybridization of ICM and GAs. The role of this hybridization is to help in the task of segmentation intensification in order to accede to good configurations.

Original languageEnglish (US)
Pages1499-1504
Number of pages6
StatePublished - Dec 1 2004
Event2004 IEEE International Conference on Industrial Technology, ICIT - Hammamet, Tunisia
Duration: Dec 8 2004Dec 10 2004

Other

Other2004 IEEE International Conference on Industrial Technology, ICIT
CountryTunisia
CityHammamet
Period12/8/0412/10/04

Fingerprint

Multi agent systems
Image segmentation
Simulated annealing
Genetic algorithms
Mathematical operators

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Melkemi, K. E., Batouche, M., & Foufou, S. (2004). MRF and multiagent system based approach for image segmentation. 1499-1504. Paper presented at 2004 IEEE International Conference on Industrial Technology, ICIT, Hammamet, Tunisia.

MRF and multiagent system based approach for image segmentation. / Melkemi, Kamal E.; Batouche, Mohamed; Foufou, Sebti.

2004. 1499-1504 Paper presented at 2004 IEEE International Conference on Industrial Technology, ICIT, Hammamet, Tunisia.

Research output: Contribution to conferencePaper

Melkemi, KE, Batouche, M & Foufou, S 2004, 'MRF and multiagent system based approach for image segmentation' Paper presented at 2004 IEEE International Conference on Industrial Technology, ICIT, Hammamet, Tunisia, 12/8/04 - 12/10/04, pp. 1499-1504.
Melkemi KE, Batouche M, Foufou S. MRF and multiagent system based approach for image segmentation. 2004. Paper presented at 2004 IEEE International Conference on Industrial Technology, ICIT, Hammamet, Tunisia.
Melkemi, Kamal E. ; Batouche, Mohamed ; Foufou, Sebti. / MRF and multiagent system based approach for image segmentation. Paper presented at 2004 IEEE International Conference on Industrial Technology, ICIT, Hammamet, Tunisia.6 p.
@conference{9f2c589279e04dd080550ae10a1fe51d,
title = "MRF and multiagent system based approach for image segmentation",
abstract = "Simulated Annealing (SA) and Iterated Conditional Modes (ICM) are two of the Markov Random Fields (MRF) model based appi'oaches for image segmentation. In practice, the ICM provides reasonable segmentations compared to the SA and was the most robust in most cases. However, the ICM strongly depends on the initialization phase. In this work, we develop a new approach for image segmentation based on Multiagent System (MAS) in order to produce good segmentations. We consider a set of segmentation agents and a coordinator agent. Each segmentation agent is able to segment the image by ICM starting from its own initialization. However, the coordinator agent diversifies the initial configurations using crossover and mutation operators known in the Genetic Algorithms (GAs). We can consider this model as a hybridization of ICM and GAs. The role of this hybridization is to help in the task of segmentation intensification in order to accede to good configurations.",
author = "Melkemi, {Kamal E.} and Mohamed Batouche and Sebti Foufou",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
pages = "1499--1504",
note = "2004 IEEE International Conference on Industrial Technology, ICIT ; Conference date: 08-12-2004 Through 10-12-2004",

}

TY - CONF

T1 - MRF and multiagent system based approach for image segmentation

AU - Melkemi, Kamal E.

AU - Batouche, Mohamed

AU - Foufou, Sebti

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Simulated Annealing (SA) and Iterated Conditional Modes (ICM) are two of the Markov Random Fields (MRF) model based appi'oaches for image segmentation. In practice, the ICM provides reasonable segmentations compared to the SA and was the most robust in most cases. However, the ICM strongly depends on the initialization phase. In this work, we develop a new approach for image segmentation based on Multiagent System (MAS) in order to produce good segmentations. We consider a set of segmentation agents and a coordinator agent. Each segmentation agent is able to segment the image by ICM starting from its own initialization. However, the coordinator agent diversifies the initial configurations using crossover and mutation operators known in the Genetic Algorithms (GAs). We can consider this model as a hybridization of ICM and GAs. The role of this hybridization is to help in the task of segmentation intensification in order to accede to good configurations.

AB - Simulated Annealing (SA) and Iterated Conditional Modes (ICM) are two of the Markov Random Fields (MRF) model based appi'oaches for image segmentation. In practice, the ICM provides reasonable segmentations compared to the SA and was the most robust in most cases. However, the ICM strongly depends on the initialization phase. In this work, we develop a new approach for image segmentation based on Multiagent System (MAS) in order to produce good segmentations. We consider a set of segmentation agents and a coordinator agent. Each segmentation agent is able to segment the image by ICM starting from its own initialization. However, the coordinator agent diversifies the initial configurations using crossover and mutation operators known in the Genetic Algorithms (GAs). We can consider this model as a hybridization of ICM and GAs. The role of this hybridization is to help in the task of segmentation intensification in order to accede to good configurations.

UR - http://www.scopus.com/inward/record.url?scp=27944466090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=27944466090&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:27944466090

SP - 1499

EP - 1504

ER -