Pengidentifikasian dan Pencarian Manusia Berbasis Citra Menggunakan Unmanned Aerial Vehicle

Abstract

Indonesia often suffers from disasters triggered by natural damage caused by human actions such as floods and landslides. The difficulty of finding victims of natural disasters is the cause of the slow process of evacuation of victims. Because the time required to perform a search is relatively long, it causes a large number of unsaved victims. The availability of camera and UAV equipment and image processing and control technologies can be used to facilitate the search for victims of natural disasters. Image processing is an image processing technique that converts the input image into another better quality image. An image processing device mounted on a UAV that can fly at a certain height can search with missions that have been entered on the software mission planner. This tool is also more accurate because it comes with a GPS that reports the coordinates of the UAV. From the tests conducted with three high variations of 5 meters, 10 meters and 15 meters, obtained an average photo of 28, 51 and 85 photos for each height respectively. This identification system works well. It is known from that detected objects can be captured images.  There are differences of number images being captured in each height, this happens because the camera specification is not supporting, as well as the influence of wifi router signal is disconnected at a certain distance. The distance between the UAV and the wifi router will affect the process of detecting objects as well as image capture.