Topological flow volume extraction from time-surface maps

The input–output behavior or flow transfer function of typical mixing processes is highly relevant to the analysis of the dynamic system and its mixing quality. We aim to visualize this behavior by extracting topologically relevant flow volumes from statistics accumulated during particle traversal of the flow field. To guarantee a sufficiently dense sampling of the flow field, we use adaptive time-surfaces for the computation of these trajectory statistics. The proposed volume extraction technique operates in parameter space of the computed time-surfaces and facilitates fast extraction of boundary geometry at different levels of detail. Our results visualize flow transfer functions in the form of volumes for extrema of different time-surface statistics and demonstrate their benefit for flow analysis.

► We transfer the notion of impulse responses and transfer functions to flow fields.
► Details on the numerical computation of such transfer functions are given.
► We propose a fast and robust treatment of topological features of these mappings.
► A robust and fast volume extraction technique for these features is developed.