Real-time infinite horizon tracking with data fusion for augmented reality in a maritime operations context.

Olivier Hugues 1 Jean-Marc Cieutat 1 Pascal Guitton 2
1 ESTIA Recherche
2 Inria Siège
Abstract : In this paper, we propose a method for real-time horizon tracking (i.e., separation line between the sky and the sea) in a maritime operations context. We present the fusion of an image processing algorithm with the data obtained from the inertial measurement unit (IMU). The initial aim is to filter out environmental conditions using inertial information in order to combine a video stream with onboard electronic charts. This is achieved by the detection of the horizon with an image processing algorithm in an area defined by the IMU. We then present an evaluation of the algorithm with regard to the rate of detection of the horizon and the impact of the image resolution on the computational time. The purpose of developing this method is to create an augmented reality maritime operations application. We combine the video stream with electronic charts in a single display. We use the position of the horizon in the image to split the display into different areas. Then, we use transparency to display the video, the electronic charts or both.
Keywords : Image processing inertial augmented reality
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [5 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-00787072
Contributor : Olivier Hugues <>
Submitted on : Monday, February 11, 2013 - 11:58:45 AM
Last modification on : Wednesday, November 29, 2017 - 3:06:11 PM
Document(s) archivé(s) le : Saturday, April 1, 2017 - 9:59:12 PM

File

manuscrit.pdf
Files produced by the author(s)

Identifiers

Collections

INRIA

Citation

Olivier Hugues, Jean-Marc Cieutat, Pascal Guitton. Real-time infinite horizon tracking with data fusion for augmented reality in a maritime operations context.. Virtual Reality, Springer Verlag, 2013, 17 (3), pp.10. ⟨10.1007/s10055-013-0234-9⟩. ⟨hal-00787072⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

269

Files downloads

385

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD