Camera calibration for unmanned aerial vehicle mapping

Abstract

Unmanned Aerial Vehicle (UAV) can be used to acquire highly accurate data in deformation survey and low-cost digital cameras are commonly used in UAV mapping. Thus, camera calibration is considered important in high-accuracy UAV mapping using low-cost digital camera. The main focus of this study is to calibrate UAV camera at different camera distances and to assess the accuracy of the image mapping. The scope of this study includes camera calibration for short and long interval range and UAV image mapping accuracy assessment using calibration parameters of different camera distances. The camera distances for the image calibration acquisition and mapping accuracy assessment were 2, 3, 4, 5, and 6 metres for the short interval range using the Sony F828 camera, and 1.4, 15, 25, and 55 metres for the long interval range using Sony NEX6. The study was conducted on a flat football field of about 2,500 square metres and in the 3D Measurement Laboratory, both located in Universiti Teknologi Malaysia. The large calibration field and a portable calibration frame were used as the tools for the camera calibration and for checking the mapping accuracy at different camera distances. Australis software was used to perform the camera calibration and image mapping processes. The results show that camera distance changes the camera calibration parameters, i.e. principal point (xp, yp), lens distortion (k1, k2, k3, p1, p2), and affinity (b1). From different camera distances in calibration process, only specific camera distance resulting the best highly accurate UAV mapping, which can achieve millimetre and sub-millimetre levels. In conclusion, camera calibration using UAV can be done at several distance intervals to choose the best camera parameters for highly accurate UAV mapping.