Share Email Print
cover

Proceedings Paper • new

Machine learning approaches for defect classification on aircraft fuselage images aquired by an UAV
Author(s): Julien Miranda; Jannic Veith; Stanislas Larnier; Ariane Herbulot; Michel Devy
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In order to ease visual inspections of exterior aircraft fuselage, new technical approaches have been recently deployed. Automated UAVs are now acquiring high quality images of the aircraft in order to perform offline analysis. At first, some acquisitions are annotated by human operators in order to provide a large dataset required to train machine learning methods, especially for critical defects detection. An intrinsic problem of this dataset is its extreme imbalance (i.e there is an unequal distribution between classes): the rarest and most valuable samples represent few elements among thousands of annotated objects. Deep Learning-only based approaches have proven to be very effective when a sufficient amount of data is available for each desired class, whereas less complex systems such as Support Vector Machine theoretically need less data, and few-shot learning dedicated methods (Matching Network, Prototypical Network, etc.) can learn from only few examples. Those approaches are compared on our applicative case. Preliminary results show the existence of empirical frontiers in term of training dataset volume that indicate which approach might be favored. Based on those results, we propose a method to combine different approaches in order to achieve best performances on categorical accuracy, with special attention paid to underrepresented category.

Paper Details

Date Published: 16 July 2019
PDF: 8 pages
Proc. SPIE 11172, Fourteenth International Conference on Quality Control by Artificial Vision, 1117208 (16 July 2019); doi: 10.1117/12.2520567
Show Author Affiliations
Julien Miranda, Lab. d'Analyse et d'Architecture des Systèmes du, CNRS (France)
Univ. de Toulouse (France)
Donecle (France)
Jannic Veith, Donecle (France)
Swiss Federal Institute of Technology-ETHZ (Switzerland)
Stanislas Larnier, Donecle (France)
Ariane Herbulot, Lab. d'Analyse et d'Architecture des Systèmes du, CNRS (France)
Univ. de Toulouse (France)
Michel Devy, Lab. d'Analyse et d'Architecture des Systèmes du, CNRS (France)
Univ. de Toulouse (France)


Published in SPIE Proceedings Vol. 11172:
Fourteenth International Conference on Quality Control by Artificial Vision
Christophe Cudel; Stéphane Bazeille; Nicolas Verrier, Editor(s)

© SPIE. Terms of Use
Back to Top