Detection of Unusual Activity in Surveillance Video Scenes Based on Deep Learning Strategies

Abstract

In today's world, abnormal activity indicates threats and risks to others. An anomaly can be defined as something that deviates from what is expected, common, or normal. Because it is difficult to continuously monitor public spaces, intelligent video surveillance is necessary. When artificial intelligence, machine learning, and deep learning were introduced into the system, the technology had progressed much too far. Different methods are in place using the above combinations to help distinguish various suspicious activities from the live tracking of footage. Human behavior is the most unpredictable, and determining whether it is suspicious or normal is quite tough. In an academic setting, a deep learning Technique is utilized to detect normal or abnormal behavior and sends an alarm message to the appropriate authorities if suspicious activity is predicted. Monitoring is frequently carried out by extracting successive frames from a video. The framework is split into two sections. The features are calculated from video frames in the first phase, and the classifier predicts whether the class is suspicious or normal in the second part based on the obtained features. This paper proposes an effective method to design a system that automatically detects any unexpected or abnormal circumstance and alerts the appropriate authority and it can be used in both indoor and outdoor settings in an academic area. The proposed system was able to achieve an accuracy rate of 95.3 percent.