A Cellular Automata Method for Filtering Digital Images

摘要

In most applications involving images or image processing one of the most common problems is the presence of noise. This fact often requires the application of some kind of smoothing procedure in order to attenuate noise and quantization error. This paper presents the application of two-dimensional cellular automata to the problem of noise removal in digital images. A simple dynamical rule for the cellular automata is proposed that maps an initial configuration representing the original noisy image, to other configuration corresponding to an image with diminished noise. The proposed method is compared, over several example images with a classical method (gaussian filter). The experiments show a reasonable good performance regardless of the nature of noise.