Robust Vision-Based Runway Detection through Conformal Prediction and Conformal mAP

Alya Zouzou, Léo Andéol, Mélanie Ducoffe, Ryma Boumazouza
Proceedings of the Fourteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 266:515-534, 2025.

Abstract

We explore the use of conformal prediction to provide statistical uncertainty guarantees for runway detection in vision-based landing systems (VLS). Using fine-tuned YOLOv5 and YOLOv6 models on aerial imagery, we apply conformal prediction to quantify localization reliability under user-defined risk levels. We also introduce Conformal mean Average Precision (C-mAP), a novel metric aligning object detection performance with conformal guarantees. Our results show that conformal prediction can improve the reliability of runway detection by quantifying uncertainty in a statistically sound way, increasing safety on-board and paving the way for certification of ML system in the aerospace domain.

Cite this Paper

BibTeX
@InProceedings{pmlr-v266-zouzou25a, title = {Robust Vision-Based Runway Detection through Conformal Prediction and Conformal mAP}, author = {Zouzou, Alya and And\'{e}ol, L\'{e}o and Ducoffe, M\'{e}lanie and Boumazouza, Ryma}, booktitle = {Proceedings of the Fourteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {515--534}, year = {2025}, editor = {Nguyen, Khuong An and Luo, Zhiyuan and Papadopoulos, Harris and Löfström, Tuwe and Carlsson, Lars and Boström, Henrik}, volume = {266}, series = {Proceedings of Machine Learning Research}, month = {10--12 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v266/main/assets/zouzou25a/zouzou25a.pdf}, url = {https://proceedings.mlr.press/v266/zouzou25a.html}, abstract = {We explore the use of conformal prediction to provide statistical uncertainty guarantees for runway detection in vision-based landing systems (VLS). Using fine-tuned YOLOv5 and YOLOv6 models on aerial imagery, we apply conformal prediction to quantify localization reliability under user-defined risk levels. We also introduce Conformal mean Average Precision (C-mAP), a novel metric aligning object detection performance with conformal guarantees. Our results show that conformal prediction can improve the reliability of runway detection by quantifying uncertainty in a statistically sound way, increasing safety on-board and paving the way for certification of ML system in the aerospace domain.} }
Endnote
%0 Conference Paper %T Robust Vision-Based Runway Detection through Conformal Prediction and Conformal mAP %A Alya Zouzou %A Léo Andéol %A Mélanie Ducoffe %A Ryma Boumazouza %B Proceedings of the Fourteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2025 %E Khuong An Nguyen %E Zhiyuan Luo %E Harris Papadopoulos %E Tuwe Löfström %E Lars Carlsson %E Henrik Boström %F pmlr-v266-zouzou25a %I PMLR %P 515--534 %U https://proceedings.mlr.press/v266/zouzou25a.html %V 266 %X We explore the use of conformal prediction to provide statistical uncertainty guarantees for runway detection in vision-based landing systems (VLS). Using fine-tuned YOLOv5 and YOLOv6 models on aerial imagery, we apply conformal prediction to quantify localization reliability under user-defined risk levels. We also introduce Conformal mean Average Precision (C-mAP), a novel metric aligning object detection performance with conformal guarantees. Our results show that conformal prediction can improve the reliability of runway detection by quantifying uncertainty in a statistically sound way, increasing safety on-board and paving the way for certification of ML system in the aerospace domain.
APA
Zouzou, A., Andéol, L., Ducoffe, M. & Boumazouza, R.. (2025). Robust Vision-Based Runway Detection through Conformal Prediction and Conformal mAP. Proceedings of the Fourteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 266:515-534 Available from https://proceedings.mlr.press/v266/zouzou25a.html.

Related Material