Saliency-Guided Change Detection for Aerial and Remote Sensing Imageries

Change detection from aerial and remote sensing imageries is an important research area with a wide range of applications in urban and environmental studies, emergency management, etc. Due to wide variability in real-world complex scenes, change detection from aerial and remote sensing imageries remains a challenging problem. In this paper, we explore using two-dimensional (2D) within-images and between-images co-occurrence histogram saliency to guide principal component pursuit (PCP). Our experimental results show that the saliency-guided PCP method can effectively extract salient physical changes while reducing non-salient environmental and acquisition changes.