Adaptive 4D Change Detection Based on PSI

Inhalt:
Previously, we proposed a PSI-based 4D change detection to detect disappearing and emerging PS points (3D) along a time sequence (1D). The change points are detected according to their change indices, which are statistically modelled from their temporal coherences of divided SAR image sets. However, heuristic selection of change index thresholds brings vulnerability and causes loss of change points. In this study, we refine our method to be adaptive, in which the change points are extracted based on statistical characteristics of change indices. Our experiment validates this adaptive approach and shows increase of change points compared with the old version.