GLoG: Laplacian of Gaussian for Spatial Pattern Detection in Spatio-Temporal Data

Boundary detection has long been a fundamental tool for image processing and computer vision, supporting the analysis of static time-varying data. In this work, we built upon theory Graph Signal Processing to propose novel boundary filter in context graphs, having as main application scenario visual spatio-temporal More specifically, equivalent graphs so-called Laplacian Gaussian edge filter, which is widely used processing. The proposed able reveal interesting spatial patterns while still enabling definition entropy time slices. reveals degree randomness slice, helping users identify expected unexpected phenomena over time. effectiveness our approach appears applications involving synthetic real data sets, show that methodology uncover temporal phenomena. provided examples case studies make clear usefulness mechanism support analytic tasks