Manhole Cover Localization in Aerial Images with a Deep Learning Approach

Benjamin Commandre 1, 2 Driss En-Nejjary 3, 1 Lionel Pibre 3 Marc Chaumont 3 Carole Delenne 2, 1 Nanée Chahinian 1
2 LEMON - Littoral, Environnement : Méthodes et Outils Numériques
CRISAM - Inria Sophia Antipolis - Méditerranée
3 ICAR - Image & Interaction
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Urban growth is an ongoing trend and one of its direct consequences is the development of buried utility networks. Locating these networks is becoming a challenging task. While the labeling of large objects in aerial images is extensively studied in Geosciences, the localization of small objects (smaller than a building) is in counter part less studied and very challenging due to the variance of object colors, cluttered neighborhood, non-uniform background, shadows and aspect ratios. In this paper, we put forward a method for the automatic detection and localization of manhole covers in Very High Resolution (VHR) aerial and remotely sensed images using a Convolutional Neural Network (CNN). Compared to other detection/localization methods for small objects, the proposed approach is more comprehensive as the entire image is processed without prior segmentation. The first experiments using the Prades-Le-Lez and Gigean datasets show that our method is indeed effective as more than 49% of the ground truth database is detected with a precision of 75%. New improvement possibilities are being explored such as using information on the shape of the detected objects and increasing the types of objects to be detected, thus enabling the extraction of more object specific features.
Type de document :
Communication dans un congrès
ISPRS Hannover Workshop: HRIGI 17 – CMRT 17 – ISA 17 – EuroCOW 17, Jun 2017, Hannover, Germany. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, XLII-1/W1, pp.333-338, 2017, <http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-1-W1/index.html>. <10.5194/isprs-archives-XLII-1-W1-333-2017>
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01556762
Contributeur : Carole Delenne <>
Soumis le : lundi 10 juillet 2017 - 14:52:41
Dernière modification le : mardi 11 juillet 2017 - 01:13:03

Fichier

isprs-archives-XLII-1-W1-333-2...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Collections

Citation

Benjamin Commandre, Driss En-Nejjary, Lionel Pibre, Marc Chaumont, Carole Delenne, et al.. Manhole Cover Localization in Aerial Images with a Deep Learning Approach. ISPRS Hannover Workshop: HRIGI 17 – CMRT 17 – ISA 17 – EuroCOW 17, Jun 2017, Hannover, Germany. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, XLII-1/W1, pp.333-338, 2017, <http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-1-W1/index.html>. <10.5194/isprs-archives-XLII-1-W1-333-2017>. <hal-01556762>

Partager

Métriques

Consultations de
la notice

110

Téléchargements du document

21