Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones

Nicklas Haagh Christensen, Frederik Falk, Oliver Gyldenberg Hjermitslev, Rikke Gade

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

We design a feature-based model to estimate and predict the free height of a fixed-wing drone flying at altitudes up to 100 meters above terrain using the stereo vision principles and a one-dimensional Kalman filter. We design this using a single RGB camera to assess the viability of sequential images for height estimation, and to assess which issues and pitfalls are likely to affect such a system. This model is tested on both simulation data flying above flat and varying terrain, as well as data from a real test flight. Simulation RMSE ranges from 10.7% to 21.0% of maximum flying height. Real estimates vary significantly more, resulting in an RMSE of 27.55% of median flying height of one test flight. Best MAE was roughly 17%, indicating the error to expect from the system. We conclude that feature-based detection appears to be too heavily influenced by noise introduced by the drone and other uncontrollable parameters to be used in reliable height estimation.
Original languageEnglish
Title of host publication Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Volume5
PublisherSCITEPRESS Digital Library
Publication dateFeb 2019
Pages751-759
ISBN (Electronic)978-989-758-354-4
DOIs
Publication statusPublished - Feb 2019
Event14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Visigrapp 2019) - Prague, Czech Republic
Duration: 25 Feb 201927 Feb 2019

Conference

Conference14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Visigrapp 2019)
CountryCzech Republic
CityPrague
Period25/02/201927/02/2019

Cite this

Christensen, N. H., Falk, F., Hjermitslev, O. G., & Gade, R. (2019). Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones. In Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Vol. 5, pp. 751-759). SCITEPRESS Digital Library. https://doi.org/10.5220/0007256107510759
Christensen, Nicklas Haagh ; Falk, Frederik ; Hjermitslev, Oliver Gyldenberg ; Gade, Rikke. / Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones. Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. Vol. 5 SCITEPRESS Digital Library, 2019. pp. 751-759
@inproceedings{2ddd18be26e14350b19fadf5f3fab3f8,
title = "Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones",
abstract = "We design a feature-based model to estimate and predict the free height of a fixed-wing drone flying at altitudes up to 100 meters above terrain using the stereo vision principles and a one-dimensional Kalman filter. We design this using a single RGB camera to assess the viability of sequential images for height estimation, and to assess which issues and pitfalls are likely to affect such a system. This model is tested on both simulation data flying above flat and varying terrain, as well as data from a real test flight. Simulation RMSE ranges from 10.7{\%} to 21.0{\%} of maximum flying height. Real estimates vary significantly more, resulting in an RMSE of 27.55{\%} of median flying height of one test flight. Best MAE was roughly 17{\%}, indicating the error to expect from the system. We conclude that feature-based detection appears to be too heavily influenced by noise introduced by the drone and other uncontrollable parameters to be used in reliable height estimation.",
author = "Christensen, {Nicklas Haagh} and Frederik Falk and Hjermitslev, {Oliver Gyldenberg} and Rikke Gade",
year = "2019",
month = "2",
doi = "10.5220/0007256107510759",
language = "English",
volume = "5",
pages = "751--759",
booktitle = "Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications",
publisher = "SCITEPRESS Digital Library",

}

Christensen, NH, Falk, F, Hjermitslev, OG & Gade, R 2019, Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones. in Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. vol. 5, SCITEPRESS Digital Library, pp. 751-759, 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Visigrapp 2019), Prague, Czech Republic, 25/02/2019. https://doi.org/10.5220/0007256107510759

Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones. / Christensen, Nicklas Haagh; Falk, Frederik; Hjermitslev, Oliver Gyldenberg; Gade, Rikke.

Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. Vol. 5 SCITEPRESS Digital Library, 2019. p. 751-759.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones

AU - Christensen, Nicklas Haagh

AU - Falk, Frederik

AU - Hjermitslev, Oliver Gyldenberg

AU - Gade, Rikke

PY - 2019/2

Y1 - 2019/2

N2 - We design a feature-based model to estimate and predict the free height of a fixed-wing drone flying at altitudes up to 100 meters above terrain using the stereo vision principles and a one-dimensional Kalman filter. We design this using a single RGB camera to assess the viability of sequential images for height estimation, and to assess which issues and pitfalls are likely to affect such a system. This model is tested on both simulation data flying above flat and varying terrain, as well as data from a real test flight. Simulation RMSE ranges from 10.7% to 21.0% of maximum flying height. Real estimates vary significantly more, resulting in an RMSE of 27.55% of median flying height of one test flight. Best MAE was roughly 17%, indicating the error to expect from the system. We conclude that feature-based detection appears to be too heavily influenced by noise introduced by the drone and other uncontrollable parameters to be used in reliable height estimation.

AB - We design a feature-based model to estimate and predict the free height of a fixed-wing drone flying at altitudes up to 100 meters above terrain using the stereo vision principles and a one-dimensional Kalman filter. We design this using a single RGB camera to assess the viability of sequential images for height estimation, and to assess which issues and pitfalls are likely to affect such a system. This model is tested on both simulation data flying above flat and varying terrain, as well as data from a real test flight. Simulation RMSE ranges from 10.7% to 21.0% of maximum flying height. Real estimates vary significantly more, resulting in an RMSE of 27.55% of median flying height of one test flight. Best MAE was roughly 17%, indicating the error to expect from the system. We conclude that feature-based detection appears to be too heavily influenced by noise introduced by the drone and other uncontrollable parameters to be used in reliable height estimation.

U2 - 10.5220/0007256107510759

DO - 10.5220/0007256107510759

M3 - Article in proceeding

VL - 5

SP - 751

EP - 759

BT - Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

PB - SCITEPRESS Digital Library

ER -

Christensen NH, Falk F, Hjermitslev OG, Gade R. Assessing Sequential Monoscopic Images for Height Estimation of Fixed-Wing Drones. In Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications. Vol. 5. SCITEPRESS Digital Library. 2019. p. 751-759 https://doi.org/10.5220/0007256107510759