Object’s Action Detection using GMM Algorithm for Smart Visual Surveillance System

• Published in: Procedia computer science Vol. 133; pp. 276 - 283
• Main Authors: V, Venkata Prasad, Rayi, Chandra Sekhar, Naveen, Cheggoju, satpute, Vishal R.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2018
• Subjects: Gaussian Mixture Model (GMM); object’s action detection; surveillance; object’s action detection; Gaussian Mixture Model (GMM); surveillance
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2018.07.034

In modern world Object’s action detection is significant for efficient video surveillance. There are several techniques used for object detection i.e., to extract foreground from background like variance method and GMM algorithm out of which GMM algorithm is the most efficient technique used. In this paper, we focused on detecting object’s action by using GMM algorithm. For this, we have applied GMM algorithm and then by using the foreground image from GMM, object’s action i.e. whether the object is static or moving is detected. This algorithm can be further used in robotic vision so that robot could detect the object’s action and then can provide the related status. And also, it has a wide range of applications such as Video surveillance for unmanned Railway-level crossing, Traffic signals controlling under video surveillance etc.