A drone is a type of Unmanned Aerial Vehicles (UAV) that most of the times can have four propellers. They can be used in many applications, one of those is to move through places of difficult access. Besides drones practicity over other aerial vehicles, its price is way lower compared to large vehicles, which turns them attractive to many activities. Also it offers safety in dangerous situations, like fires or accidents, as it doesnt need an on-board pilot. In a system with autonomous flight the concern with its landing and recharging of the batteries, which does not last more than a few minutes, arises. Using on-board devices, like its cameras and GPS modules, it is possible to implement functions to optimize its capabilities. With the goal to present a solution to such problem, this essay proposes a model which utilizes image recognition to allow a drone to land in an autonomous system. This landing routine based on its image turns flight and landing into autonomous processes, without human intervention.