JAUE2018-051 A Study on the Graveyards Recognition in Forest Area using Deep-Learning based UAV Image

Abstract

Urban development projects are accelerating and the demand for SOC infrastructure is increasing, so wide land and forest area development is underway. In this way, graveyards are revealed in the process of changing and maintaining the existing forest area condition, and relocating graves is essential in development. So far, the survey of forest areas has become diverse and complicated, such as the shape and characteristics of the target area, and the environmental conditions of forests. Therefore, it is time for systematic countermeasures. At present, field survey methods using field visits and tape measures are the main approach. In recent years, UAV has been attracting interest and is being used in various ways as a way to construct rapid and accurate 3D spatial information at low cost. In this study, the spatial information was constructed by using UAV and the input image and the matching parts in the constructed digital map were explored. It is expected that the method proposed in this study will contribute to the exploration of graveyards in the forest development area.