Using Fuzzy Logic and Particle Swarm Optimization to design an image filter
Hung-Hsu Tsai, Ji-Shiang Shih, Bae-Muu Chang · 2012
This paper presents an Image Filter with noise detector using Fuzzy Logic and Particle Swarm Optimization (PSO), which is called the IFFLPSO filter, for removal and restoration of impulse noises. In IFFLPSO filter, the fuzzy logic is employed to efficiently design the noise detector. The proposed filter effectively judges the input pixel vector whether it is corrupted or not. Meanwhile, the particle swarm optimization algorithm (PSO) is utilized so as to optimize the noise detectors to enhance the noise detection performance. Subsequently, in order to enhance the restoration performance of proposed filter, the color ratio of spot's region in the restored image is employed to determine the spot's color. Also, the pixel vectors with different color ratios in the spot region are detected. Finally, the vector median filter is utilized to restore the corrupted pixels. Experimental results demonstrate that the proposed image filter outperforms the existing other well-known filters in restoration performance. And the system can be widely applied in microarray image processing.