Abstract: Gaussian diffusion mode of smokescreen is not suitable for complex atmospheric conditions and it is not easy to accurately obtain statistical properties of particles by utilizing random-walk mode, so the practicability of these two kinds of modes is limited. In view of the problems mentioned above, the revised random-walk smokescreen simulation model method was put forward, in the method, mathematical morphology algorithm was utilized to detect the data on smokescreen edge in a video in real-time and the model was rectified in real-time on basis of such data. The test result indicates that the revised model is suitable for actual smokescreen modeling and simulation under complex atmospheric conditions. The precision of the generated simulation smokescreen is controllable and the model is approximate to the real smokescreen. The method realizes precise simulation of smokescreen and solves the problems that the random-walk model and gauss smokescreen simulation model are not practical.

Key words: smokescreen, mathematical morphology, edge detection, random-walk mode, simulation