Abstract: Detection of moving objects in video streams is that the first relevant step of data and background subtraction may be a very talked-about approach for foreground segmentation. in this system, we are going to simulate different background subtraction strategies to beat the matter of illumination variation, background clutter and shadows. detecting and tracking of frame elements is very important in understanding human activities. Intelligent and automatic security surveillance systems became an energetic analysis space in recent time because of associate increasing demand for such systems publicly areas like airports, underground stations and mass events. during this context, pursuit of stationary foreground regions is one in every of the foremost vital needs for surveillance systems supported the pursuit of abandoned or purloined objects or pose vehicles. it's terribly difficult for somebody's operator to effectively observe events as they happen. Recently computer vision analysis has got to address ways in which to automatically a number of this knowledge, to help human operators. Video surveillance system is a process of monitoring and analyzing video sequences for the purpose of checking the behavior, activities and other certain information in a video sequence. In this system, the system will calculate the number of people present in house, room etc. and on that bases it increases or decreases the speed of fan.
Keywords: R-CNN, Automatic Detection, Tracking, Real-Time Video
| DOI: 10.17148/IJARCCE.2019.8517