ABSTRACT
The drone surveillance system becomes predominantly important in times of wars for safety and security. Drone surveillance provides visual imagery or video of the site. UAVs flying in military and police operations would aid in various surveillance operations by providing high quality footage to identify various targets. Most drones just provide live video feed through cameras attached to them. They are unable to identify the objects present at the site. This paper proposes the development of a machine learning- based approach for the detection of animate objects in real-time video streams captured from the drone. It also involves the development of a ground controller which will be used to operate the drone in manual mode. The controller has many options for taking off the drone, landing, manual control of the drone, destination setting, etc.
