Foreground model for background subtraction with blind updating
Haixia Wang, Li Shi · 2016
Background subtraction is a key pre-processing step for several video automatic operations. Various techniques have been proposed to perform background subtraction automatically in complex environments. Visual background extractor (ViBe) is a popular background subtraction technique that can initialize its model in a single frame, adapt to the environment changes and achieve satisfactory subtraction results. In this paper, we propose to use ViBe with blind updating which can more quickly adapt to dynamic environment changes. We propose foreground model with adaptive updating strategy to assist the ViBe with blind updating to detect slow moving object without introducing the ghost phenomenon. Experimental results have verified the performance of the proposed technique.