PROJECT TITLE :

Background Modeling by Stability of Adaptive Features in Complex Scenes - 2018

ABSTRACT:

The one-feature-primarily based background model typically fails in complicated scenes, since a pixel is better described by many options, that highlight different characteristics of it. So, the multi-feature-based mostly background model has drawn a lot of attention recently. In this Project, we propose a unique multi-feature-based mostly background model, named stability of adaptive feature (SoAF) model, that utilizes the stabilities of different features in a pixel to adaptively weigh the contributions of these features for foreground detection. We do that mainly due to the very fact that the options of pixels within the background are often more stable. In SoAF, a pixel is described by several options and each of those options is depicted by a unimodal model that offers an initial label of the target pixel. Then, we tend to live the stability of every feature by its histogram statistics over a time sequence and use them as weights to assemble the aforementioned unimodal models to yield the ultimate label. The experiments on some customary benchmarks, that contain the complex scenes, demonstrate that the proposed approach achieves promising performance compared with some state-of-the-art approaches.