3D tracking between satellites using monocular computer vision

[摘要] Visually estimating three-dimensional position, orientation and motion, between an observerand a target, is an important problem in computer vision. Solutions which compute threedimensionalmovement from two-dimensional intensity images, usually rely on stereoscopicvision. Some research has also been done in systems utilising a single (monocular) camera.This thesis investigates methods for estimating position and pose from monocular image sequences.The intended future application is of visual tracking between satellites flying in closeformation. The ideas explored in this thesis build on methods developed for use in camera calibration,and structure from motion (SfM). All these methods rely heavily on the use of differentvariations of the Kalman Filter.After describing the problem from a mathematical perspective we develop different approachesto solving the estimation problem. The different approaches are successfully tested on simulatedas well as real-world image sequences, and their performance analysed.

[发布机构] Stellenbosch University