We do to curb noise and improve image quality, you must first pre-processing of images, that is, smooth images. The purpose of image smoothing is to improve the image of the visual effects and improve image clarity in order to become more conducive to the computer image processing and analysis of various characteristics. We use a classic cellular automata “Game of Life,” conducted a study about image smoothing based on game of life to image that a cell for each pixel, using a radius of 1 neighbors of Moore type. We transform the image of the different rules, the use of two-dimensional cellular automaton model area of the smoothed image. We experimentally demonstrate that the smoothing method with the median filter to the normalized mean square error and peak signal to noise ratio comparison method revealed that the more obvious advantages.