How to define neighbors in the feature space is important to visualization. Neighbors based not only on physical attributes but also on the feature attributes will provides a criterion to reduce complexity. We compare several distance measure methods in vortex visualization. These methods include λ2 distance, Δ distance, Q distance, Γ2 distance and normalized Euclidean distance of physical data from dataset. We compare these distance measure methods on three kinds of data samples from some fluid dynamic data set. We aim to find a measure to determine feature neighbors in flow data.
展开▼