Traditional static flow field visualization methods suffer from many problems, such as a lack of continuity expression in the vector field, uneven distribution of seed points, messy visualization, and time-consuming calculations. In response to these problems, this paper proposes a multi-scale mapping method based on real-time feature streamlines. The method uses feature streamlines to solve the problem of continuity expression in flow fields and introduces a streamline tracking algorithm which combines adaptive step length with velocity matching to render feature streamlines in a real-time and multi-scale way. In order to improve the stability and uniformity of the seed point layout, this method uses two different point placement methods which utilize a global regular grid distribution algorithm and feature area random distribution algorithm. In addition, this method uses a collision detection algorithm to detect and deal with the unreasonable covering between streamlines, which alleviates visual confusion in the resulting drawing. This method also uses HTML5 Canvas to render streamlines, which greatly improves the drawing speed. Therefore, use of this method can not only improve the uniformity of the seed point layout and the speed of drawing but also solve the problems of continuity expression in the vector field and messy visualization.
Read full abstract