The goal of this project:

• Implement a multi-objective evolutionary elgorithm for color image segmentation
• Implement a simple weighted-sum genetic algorithm for color image segmentation
• Compare the results of the two implemented solutions on several benchmark problems
• Test and analyze the effects of MOEA(s) in optimizing multiple objectives simultaneously.

Image segmentation is a fundamental process in many image, video and computer vision applications. The main goal is to partition an image into separate regions of pixels, which ideally correspond to different real-world objects.

Original Image	Type 1 segmentation	Type 2 segmentation

NSGA-II is a modification of NSGA. NSGA-II computes the cost of an individual x by taking into account not only the individuals that dominate it, but also the individuals that it dominates. For each individual, we also compute a crowding distance by finding the distance to the nearest individuals along each objective function dimension.

• A Multi-Objective Evolutionary Algorithm for Color Image Segmentation. Ripon, Ali, Newaz, Ma

• Introduction to Evolutionary Computing. Smith, Eiben