Hybridization of deep and prototypical neural network for rare defect classification on aircraft fuselage images acquired by an unmanned aerial vehicle.
Julien MirandaJannic VeithStanislas LarnierAriane HerbulotMichel DevyPublished in: J. Electronic Imaging (2020)
Keyphrases
- unmanned aerial vehicles
- neural network
- input image
- image database
- image data
- defect classification
- object recognition
- image classification
- rigid body
- acquired images
- ground truth
- three dimensional
- image analysis
- image features
- image retrieval
- image understanding
- image annotation
- image registration
- edge detection
- neural network is trained
- back propagation
- pattern recognition