This paper describes an attempt to develop an automatic unmanned aerial vehicle (UAV) control system for emergency rescue and investigation. A cascade classifier system using Haar-like features for object recognition from a free viewpoint is also presented. Conventional emergency rescue systems have a shortcoming in terms of their performance. Thus, we introduce an automatic UAV control system in order to overcome the shortcoming of conventional emergency rescue systems in an emergency situation. The proposed system involves two processing stages. One involves a capturing and recognition of target objects, and the other involves controling the UAV using digital intelligence. Object detection in many types of images as well as real-time simulation methods continue to be one of the most common techniques for information processing and applications. The fundamental concepts of the proposed system and some experimental results obtained by using related processing techniques are also described.
J-STAGEがリニューアルされました！https://www.jstage.jst.go.jp/browse/-char/ja/
