Detection and Location Estimation of Object in Unmanned Aerial Vehicle using Single Camera and GPS

Abstract

It is important for an Unmanned Aerial Vehicle (UAV) to detect any object in its view. This enables the UAV to locate the object with respect to itself and is required for locking and tracking the object. Object detection and location estimation makes the UAV capable of manipulating the environment as well as to follow the target. In the present work, object detection has been carried out using You Only Look Once (YOLO) to detect the object in the image stream of the Robot Operating System (ROS) bag file. The GPS information of the UAV is used to further calculate the GPS coordinates of the object. The images are acquired using a single monocular camera.