• Vishwa Pratap Singh
• Vasudev Sharma
• Dhanshri Prihar

Video surveillance frameworks produce gigantic measures of information for display and storage. Long hours of monitoring by human of the surveillance video is unfeasible and incapable. Programmed abnormalities motion detection framework which can adequately pull in administrator consideration and trigger recording is consequently the fruitful way of video surveillance in dynamic scenarios. This paper proposes a solution for real-time detection of abnormal motion. The proposed technique is appropriate for present day video surveillance models, where restricted computing power or processing control is accessible close to the camera for communication and compression. The algorithm utilizes the macroblock motion vectors which are produced regardless as a major aspect of the compression process of videos. Motion vectors have been used to derive motion features. The factual distribution of such features amid natural movement is assessed via training. Test results exhibit real time operations which are reliable enough.

Activity Detection, abnormal activity

"Abnormal Activity Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.674-678, May-2019, Available :http://www.jetir.org/papers/JETIR1905G97.pdf

"Abnormal Activity Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp674-678, May-2019, Available at : http://www.jetir.org/papers/JETIR1905G97.pdf