Your browser version is outdated. We recommend that you update your browser to the latest version.

Manuscript Title: Multiobjective Optimization Approaches in Image Segmentation – The Directions and Challenges

Author : Bong Chin-Wei and Mandava Rajeswari

Email : bongwendy@gmail.commandava@cs.usm.my

Abstract: A new trend of problem formulation for image segmentation is to use multiobjective optimization approach in its decision making process. Multiobjective formulations are realistic models for many complex optimization problems. In many real-life problems, objectives under consideration conflict with each other, and optimizing a particular solution with respect to a single objective can result in unacceptable results with respect to the other objectives. The aim of this paper is to provide a comprehensive review of multiobjective optimization in image segmentation problem. The models are classified according to relevant features, such as the various aspects of the optimization approaches used, various possible problem formulations, type of datasets modelled and the scope of the applications. Through our analysis of the current state of the research, we diagnose some of the directions and challenges for modelling the image segmentation problem with multiple objectives criteria. In this review, we consolidate the selected material in the literature, including more than 80 studies dated 1990 or later.

Keywords: Multiobjective Optimization, soft-computing techniques, image segmentation, clustering, classification

Vol 2 (1)