Using Maxima Score for the Extraction and Visualization of Fluid Flow Structures

The extraction of important flow features can facilitate easier cognition of fluid flows, enabling users to improve object/fluid interactions. This paper presents an approach to extract flow features, which can be applied to data of any domain. Our approach uses a normalized "maxima score" for predictable output across datasets. This score is calculated based on a given point\´s scalar intensity relative to its neighbors. This score can be used to reveal areas of interest within the flow field. We present two techniques for visualization of the maxima score, one based on isosurfaces and one based on edge tracing from image processing. We have successfully applied the maxima score approach to synthetic, computational, and empirical datasets.