Motion Detection Using Heuristic AI based Machine Learning Approaches

Abstract

Due to the rise in shootings, knife assaults, terrorist attacks, etc. that occur in public spaces around the world, it has become crucial to identify suspicious activity in these areas. This study employs convolutional neural networks and deep learning to identify suspicious activity in videos and photos. We examine various CNN architectures and contrast their precision. We describe the design of our system, which can analyze live video feed from cameras and determine whether an activity is suspicious or not. We also make suggestions for potential future advancements in the field of detecting suspicious activity.

Topics & Concepts