Nakib Amir - Metaheuristics for Medicine and Biology

In this work, a new formulation of the image thresholding problem as a shortest path problem was proposed. Then, a hybrid ant colony system based algorithm to solve was used to solver it. Moreover, a new segmentation criterion was proposed, named the biased survival exponential entropy. This criterion considers the cumulative distribution of the gray level information and takes into account spatial quality of the segmentation result. It is well known that the ant colony optimization algorithms are slow; in order to solve this problem we enhanced the classical ant colony system by hybridizing it with a local search algorithm. The original Ant Colony System (ACS) [] was applied in different real world applications. Since the formulation of the problem as a graph optimization problem, ACS can be applied by associating two measures to each arc: the closeness , and the pheromone trail . ACS uses a mechanism based on three main operations: (1) the state transition rule provides a direct way to balance between exploration of new edges and exploitation of a priori and accumulated knowledge about the problem. (2) The global updating rule is applied only to edges that belong to the best ant tour. (3) While ants construct a solution, a local pheromone updating rule (local updating rule, for short) is applied.

In ACS the state transition rule is as follows: an ant positioned at the node r chooses the city v to move to by applying the rule given below.