The article raises the problem of navigation and positioning of unmanned aerial vehicles without using satellite navigation systems, such as GPS and GLONASS. The idea of creating a navigation system using computer vision technology, which is a subsection of the discipline of artificial intelligence, is described herein. The article gives a brief description of the core concepts and problems of one of the main subsections of computer vision – identification of images and finding of specific points on the image. Also algorithms for determination of local features on the image, that are clearly differing in their operation principle and purposes, have been reviewed and analysed from the point of view of identifying specific points on the image. Based on the operation principles of the reviewed algorithms, their material advantages and disadvantages, which directly affect the efficiency of using these algorithms in non-satellite navigation system, have been identified. In the conclusion, several key aspects that should be taken into account in selection of appropriate algorithm for development of a navigation system are highlighted.
REVIEW OF IMAGE FEATURE POINTS DETECTION ALGORITHMS
2018
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
Metadata by DOAJ is licensed under CC BY-SA 1.0
Modifications to Pose Estimation Algorithms that Allow for Near-Planar Feature Points
British Library Conference Proceedings | 2008
|