ACO Based Adaptive Filter for High Density Impulse Noise

In this paper, we proposed an ACO based adaptive non-casual linear prediction technique for vector median filter to remove high density impulse noise from color images. It is a new method for impulsive noise reduction and edge preservation in images. Generally, when an image is affected by high density of impulse noise, homogeneity among the pixels is distorted. Images of different characteristics corrupted with a wide range of impulsive noise densities using impulsive noise model is examined using the proposed method. This paper, based on the basic ant colony algorithm and integrating with the genetic algorithm, proposes an image restoration processing method based on hybrid ant colony algorithm. This method transforms the optimal population information of genetic algorithm into the original pheromone concentration matrix of ant colony algorithm and uses it to compute the parameters of degradation function to get a precise estimate of the original image. By analyzing and comparing the restoration results, the method of this paper cannot only overcome the influence of noises, but it can also make the image smoother with no fringe effects in the edges and excellent visual effects, verifying its practicability. The proposed filter improves the Peak Signal to Noise Ratio (PSNR).