Optimizing Stack Filters by Clone Selection Algorithm

The design of stack filters can be formulated as a course of optimizing positive Boolean functions (PBF). In this paper, we use clone selection algorithms (CSA) to optimize stack filters under the mean absolute error (MAE) criterion. The optimal stack filters are used to restore images corrupted with noise. The outcomes are compared with that those optimized by genetic algorithms (GA) and by particle swarm optimization (PSO) Algorithms. Simulation results show that the filters optimized by this method can preserve details well and suppress noise efficiently at the same time. Otherwise, the time spending in computing error cost is much less than that using other two algorithms