“…Vision analysis techniques can provide high-accuracy obstacle representations. Although the high computation overhead has hindered the adoption of this approach in the auto-navigation of indoor mini quadrotors [5], the recent technological advances in the on-boardscale micro-computers, such as Raspberry Pi, Intel Edison, 1 Mr. Shin is a graduate student of the computer science and information technology (CSIT) department at the University of the District of Columbia (UDC), Washington, DC 20008, donghyeok.shin@udc.edu 2 Dr Kim is with Faculty of the CSIT department at UDC, Washington, DC 20008,junwhan.kim@udc.edu 3 Drs Yu, and Jeong are with Clearton, LLC (www.clearton.com),{byu, djeong}@clearton.com and Arduino, have been significant [6], and the distribution of advanced vision analysis techniques in open source libraries, such as OpenCV [7], has been accelerated. In the light of these recent developments, we focus on designing a visionbased method for detecting and avoiding obstacles for realtime autonomous navigation of mini quadrotors in indoor environments with moving obstacles.…”