Point Based Motion Detection on UAV Cameras

Motion detection from video is a hot topic in computer vision. Also it is an important part of smart video surveillance system. In today's world surveillance system is playing an important role in the field of security. Moving object detection has been widely used in video surveillance systems. Besides, correct detection of moving objects is important part of video processing and surveillance systems. Thanks to development of hardare features of cameras, number of proposed methods are increasing. In this paper we develop a novel method for optical flow based motion detection in videos. In this approach by applying feature extraction method, corner points are obtained. After, optical How and transformation matrix using this points are calculated. If the difference of results is bigger than the threshold, this points mark as motion points. Results showed that our algoritm produced promissing results for moving cameras.