Pose Estimation of Flying Object Based on Monocular Camera and Feature Detection

Abstract

Position measurement plays an important role in the modern control, national defense, aerospace and medical fields. Pose measurement system should have characteristics of high precision, strong real-time performance and easy installation. Compared with other measuring systems, visual measuring system has many advantages, such as simple structure, low cost and high resolution. Compared with the double (multiple) camera, monocular camera detection do not have to match features of difficulty, occupy large space, and adjust inconveniently. Moreover, the field of pose measurement based on monocular camera still has a lot of blank, so this project is researching in pose measurement using monocular camera. This project was tested with the help of Frisbees. Ultimately, we got the method of how to measure the rotating objects, found how to measure the triaxial pitch and speed of the flying objects, and presented a monocular vision theory model of the aircraft model according to the characteristics of the movement.