Genetic Algorithms-Based Simulation Of Diffusion Of Chinese Ink Drawing

With the development of computer graphic non-photorealistic rendering and Chinese traditional ink drawing art, studies of the simulation of Chinese ink wash drawing is one of the most challenging work. These Chinese ink wash modeling methods improve computer art and digital art.The most important effect of Chinese drawing is ink diffusion, which is also the hardest. By analyzing the Chinese ink wash drawing material and diffusion mechanism, a new method based on genetic algorithm is presented to model ink particle character and action.Our method includes tow parts: object and material of simulation. The former is about the characters of ink stroke by watching lots of physical Chinese wash ink drawing works. The simulation objects contain original ink stroke, fields of diffusion and irregular diffusion edge. The latter is more essential and the core of this paper. The simulation work involves xuan-paper model, water ink model and ink diffusion model. The task of paper model is to simulate the characters of Chinese xuan-paper. These characters embrace microcosmic fiber random and macroscopic construction texture. We simulate microcosmic fiber random character by Gaussian distribution and the component of this paper is called as"paper cell". We synthesize macroscopic construction texture by Ashikhmin algorithm and get the characters of texture using k-means.Water ink model simulates the original strokes by binary gene encoding, initializing population and gene mutation.Ink diffusion model builds a mechanism of diffusion action based on diffusion process points. It constructs a list of mapping that describes the correspondence of Minkowski distance to the rate gene mutation so that gene mutation can happen suitably by itself. As a result, modeling the diffusion of Chinese ink wash drawing can be finished automatically.Some physical laws are taken into account in building the simulation model , but this is not a strict and exact physical process simulation. We aimed at characterizing the specialties of the Chinese painting and regenerating the extrusive artistic effects that are concerned eagerly by the artists and building a software system.