Real Time Violence Detection Using Autonomous Intelligent Surveillance Robot

Intelligent surveillance robots dominate the entire world with its efficiency and accuracy helping in performing dangerous task.The surveillance robot helps to survey the area. The violence keeps on growing day by day.This paper presents an intelligent surveillance robot capable of detecting and preventing violence in two modes - motion-controlled and autonomous. The motion-controlled mode allows the user to manually control the robot's movement and camera orientation, while the autonomous mode allows the robot to navigate and monitor the environment independently. In both modes, the robot employs a deep learning algorithm taken from CNN and LSTM for accurately detecting violence and sending a notification through email with the image of the violent activity.On a authenticated dataset, the proposed concept has been examined, the outcomes looks promising in regards to accuracy and speed.The integration of motion control and autonomous operation modes makes the surveillance robot versatile and effective in a range of surveillance scenarios. Overall, this project demonstrates the potential of machine learning and robotics for enhancing security and surveillance measures.