An Image Filter with a hybrid impulse detector based on decision tree and Particle Swarm Optimization
Hung-Hsu Tsai, Xuan-Ping Lin, Bae-Muu Chang · 2009
The paper proposes an Image Filter with a hybrid impulse detector based on Decision Tree (DT) and Particle Swarm Optimization (PSO), which is called the IFDTPSO filter, for the recovery of the corrupted image by impulse noises. The Several impulse noise detectors are combined in the design of the IFDTPSO filter to form an impulse noise detector (IND) which is designed by DT and PSO to effectively detect corrupted pixels of noisy images. The IND can correctly classify pixels as either noise-free or noise-corrupted, and then, restoring process utilizes the median filter to powerfully recover corrupted pixels from noisy images. Experimental results demonstrate that the IFDTPSO filter outperforms the existing well-known methods.