| Autor: |
Ulf Bestmann, Peter Hecker, Maik Angermann, Andreas Dekiert, S. Wolkow |
| Rok vydání: |
2019 |
| Předmět: |
|
| Zdroj: |
The International Technical Meeting of the The Institute of Navigation. |
| ISSN: |
2330-3646 |
| DOI: |
10.33012/2019.16733 |
| Popis: |
This paper discusses the unique advantages of and challenges facing infrared cameras in the domain of optical relative positioning during landing by runway detection. It introduces a novel histogram-based algorithm for temperature mapping of the infrared data which aims to counteract the atmospheric absorption of infrared radiation as well as influences like an obstructed field-of-view or extreme temperatures. These effects often have negative ramifications when using conventional temperature mapping algorithms and result in an infrared image which is unsuitable for runway detection. Furthermore, this paper proposes an adaptive Canny threshold parameter estimation. It is optimized to utilize the high contrast in the infrared image provided by the presented temperature mapping algorithm and aims at increasing the ratio of extracted runway features to total number of extracted features from the image. The presented algorithms were evaluated against an infrared based system using legacy algorithms as well as against a system using visual light cameras. Flight test data was gathered with two research aircrafts, Dornier Do 128-6 and Diamond DA42, at two different airports in Braunschweig and Wiener Neustadt. Results indicate, that the proposed algorithms either increase the distance of first runway detection by several times or even make the detection possible at all, thus also significantly increasing the availability of the runway detection. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
