Background Modeling Method Based On Quad Tree Decomposition and Contrast Measure
Samah Elkah, Siham Aqel, My Abdelouahed Sabri, Abdellah Aarab · Procedia Computer Science · 2019
Motion detection is a complex task that builds upon a range of underlying components. In order to assure accurate object detection, it is quite requisite to implement these components meticulously, especially in the presence of complex objects in the scene. This paper proposes a new block-based method of background modeling in dynamic scenes using quad tree decomposition and contrast measure. Experimental results show that our approach can effectively generate the background even under challenging circumstances.