Abnormal Activity Detection in Crowded Areas
Sohan Tummalapalli, Durga Jayasai Pillagolla, Aparna Vaishnavi Karpurapu, Narayana Swami Potla, Iwin Thanakumar Joseph · 2024
Detecting abnormal activity in crowded areas poses a significant challenge due to the intricate nature of crowd dynamics and the difficulty in defining what constitutes normal behaviour. It is an effort to maintain security and prevent crime. In this paper we examine the advancements in identifying abnormal activity within crowded areas. Our focus lies on techniques that utilize computer vision and machine learning to detect abnormal behaviour in videos and images of populated scenes. We also discuss the challenges in identifying abnormal activity in certain locations and discuss possible directions for further investigation.