A Machine Learning Model for Human Motion Detection and Event Feedback

Abstract

Fundamentally it is tough to answer, what is a peculiar event. Completely normal event in one situation would possibly turn out to be bizarre in the any other situation. For instance, reflect on consideration on the activities like, ‘a cattle is grazing in a giant open field’ and ‘a cattle is grazing in a house backyard’. In video processing the place the heritage get subtracted, each are the equal events; besides the situation the place the motion of grazing is being performed. An essential aspect that separates both activities aside is the normal frequency of their occurrences. The grazing is common action and occurs often in an open grass field unlike the residence backyard, which makes it normal tournament for that situation. When the grazing is repeatedly performed many instances in the house outside (probably in suburb) then it becomes regular pastime and have to be classified as the ordinary event. Hence anomalous activities happen enormously infrequently. After performing experimental results for clustering performance evaluation on artificial dataset show that the clustering outperforms the other clustering methods, for clustering the single dominant class from the dataset.