Suspicious Activity Detection in Video Surveillance using Convolutional Neural Network

Video surveillance systems have become an integral part of ensuring public safety and security are a concern in multiple areas, such as transportation, retail, and critical infrastructure.Detecting suspicious activities within vast amounts of video volume is difficult, necessitating development of sophisticated methods.A new strategy is proposed in this study to detect suspicious activity in video surveillance using Convolutional Neural Networks (CNNs).The proposed system leverages the deep learning capabilities of CNNs to automatically extract relevant features from video frames and identify unusual or suspicious behaviors.Key steps in the methodology include data preprocessing, frame extraction, and CNN model training.By utilizing a large dataset of labelled video clips, the CNN model learns to recognize patterns associated with suspicious activities, such as loitering, violence, or trespassing.In our proposed system we are developing a video surveillance system using CNN which can generate alerts for the suspicious activities.