Autonomous unmanned aerial vehicle (UAV) operations heavily depend on the global positioning system (GPS) for localization and mission planning. However, there are environments, like dense urban areas, areas below dense canopy cover, and cloudy weather conditions, where GPS positioning is restricted or unavailable. Besides, the extensive dependence on GPS technology creates vulnerabilities to GPS spoofing, where signals are manipulated to interfere with navigation systems illicitly. This paper proposes a method for estimating desired object geo-location as well as UAV geo-position for uninterrupted autonomous navigation in the satellite-out-of-coverage area. Identifying the unknown object followed by obtaining its Cartesian coordinate in the aerial image through multiple frame rotations, the projection of the geo-location from the Cartesian coordinates to the satellite map, and estimating the self geoposition for navigation are the key contributions of this work. The performance of the proposed model is assessed through field experiments. The result shows the obtained geo-location of the object is very close to the actual geo-location with minimal error.