Predicting Anomalous and Consigning Apprise During Heists

Nowadays, there are many problems due to robberies, and there are many solutions for this but this can't be stopped at that moment. The crime charge has increased because there aren't many security guards outside the cell. Consequently, there is a critical need for a sophisticated device that can detect an anomaly i.e., face occlusion and weapon detection are considered anomalous behavior. With the current system, the ongoing illegal behavior cannot be reported to the nearby protective authorities before the criminal may get away. With the aid of Convolutional neural network (CNN), YOLO algorithms, and SMTP for sending emails.this machine has provided a technique to identify and categories whether or not it's unusual behavior or not, and to send alerts to the security government. In which the MobileNetV2 model of CNN algorithm is used to detect the face mask of the burglar and YOLOv4 for detecting objects like guns, rifles, and knives. When compared to the ResNet50 model, MobileNetV2 gives high accuracy with less time consumption. With the number of crimes increasing as well as the lack of safety, there is a growing necessity to produce this kind of security gadget. This method helps in reporting the ongoing unlawful activity to the local government's concerned protection agency in order to prevent another robbery from occurring.