Learning Iterative Processes with Recurrent Neural Networks to Correct Satellite Image Classification Maps

Emmanuel Maggiori 1 Guillaume Charpiat 2 Yuliya Tarabalka 1 Pierre Alliez 1
1 TITANE - Geometric Modeling of 3D Environments
CRISAM - Inria Sophia Antipolis - Méditerranée
2 TAO - Machine Learning and Optimisation
CNRS - Centre National de la Recherche Scientifique : UMR8623, Inria Saclay - Ile de France, UP11 - Université Paris-Sud - Paris 11, LRI - Laboratoire de Recherche en Informatique
Abstract : While initially devised for image categorization, convolutional neural networks (CNNs) are being increasingly used for the pixelwise semantic labeling of images. However, the proper nature of the most common CNN architectures makes them good at recognizing but poor at localizing objects precisely. This problem is magnified in the context of aerial and satellite image labeling, where a spatially fine object outlining is of paramount importance. Different iterative enhancement algorithms have been presented in the literature to progressively improve the coarse CNN outputs, seeking to sharpen object boundaries around real image edges. However, one must carefully design, choose and tune such algorithms. Instead, our goal is to directly learn the iterative process itself. For this, we formulate a generic iterative enhancement process inspired from partial differential equations, and observe that it can be expressed as a recurrent neural network (RNN). Consequently, we train such a network from manually labeled data for our enhancement task. In a series of experiments we show that our RNN effectively learns an iterative process that significantly improves the quality of satellite image classification maps.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/hal-01388551
Contributor : Emmanuel Maggiori <>
Submitted on : Thursday, October 27, 2016 - 11:12:29 AM
Last modification on : Tuesday, April 17, 2018 - 9:08:49 AM

File

Maggiori_RNN2016.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01388551, version 1
  • ARXIV : 1608.03440

Citation

Emmanuel Maggiori, Guillaume Charpiat, Yuliya Tarabalka, Pierre Alliez. Learning Iterative Processes with Recurrent Neural Networks to Correct Satellite Image Classification Maps. 2016. ⟨hal-01388551⟩

Share

Metrics

Record views

600

Files downloads

798