Supportive Information to Find Victims from Aerial Video in Search and Rescue Operation

An unmanned aerial vehicle (UAV) is expected to be one of the powerful IoT tools in a search and rescue (SAR) operation. Notably, an aerial video would be most promising. However, in a SAR operation, a searching task via an aerial video compels laborious tasks such as high concentration to practitioners. In this paper, to reduce such a laborious task, we propose a method to provide supportive information, i.e., the number of persons, the time and thumbnails, for finding victims via an aerial video with an object detection algorithm. Also, to use it in a SAR operation, we need to meet the requirements for a dataset and lightweight process. Through the evaluation, we showed that we could get supportive information reasonably and displayed them on a GUI.