Details

Visually effective rendering is a common challenge for visualizing fractals, attractors, and other datasets. In many cases the data density varies very significantly. If we directly map the computational solution to dots or other graphics primitives on the 2D canvas, in some areas the data could be sparse, while in other areas the plots overlap, leading to a flat appearance and loss of detail. This Demonstration suggests one possible and simple approach to enhance the visual appearance to some extent, using density for tone mapping, gamma adjustment, Gaussian blur, and colorization.

Reference

[1] S. Draves and E. Reckase, "The Fractal Flame Algorithm," pp. 6–8. http://flam3.com/flame.pdf.