FN Archimer Export Format
PT J
TI Domain-Adversarial Neural Networks for Deforestation Detection in Tropical Forests
BT 
AF SOTO VEGA, Pedro Juan
   Costa, Gilson A.
   Feitosa, Raul Q.
   Ortega, Mabel X.
   Bermudez, Jose D.
   Turnes, Javier N.
AS 1:1;2:2;3:3;4:3;5:3;6:4;
FF 1:PDG-REM-BEEP-LEP;2:;3:;4:;5:;6:;
C1 French Research Institute for Exploitation of the Sea, Brest, France.
   State University of Rio de Janeiro, Rio de Janeiro, Brazil.
   Department of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, Rio de Janeiro, Brazil.
   Department of System Design Engineering, University of Waterloo, Waterloo, Canada
C2 IFREMER, FRANCE
   UNIV RIO DE JANEIRO, BRAZIL
   PONTIFICIA UNIV CATOLICA RIO DE JANEIRO, BRAZIL
   UNIV WATERLOO, CANADA
SI BREST
SE PDG-REM-BEEP-LEP
UM BEEP-LM2E
IN WOS Ifremer UMR
   copubli-int-hors-europe
   copubli-sud
IF 4.8
TC 10
UR https://archimer.ifremer.fr/doc/00762/87439/92949.pdf
LA English
DT Article
DE ;Feature extraction;Training;Neurons;Task analysis;Neural networks;Forestry;Noise measurement;Change detection;deep learning (DL);deforestation detection;domain adaptation (DA);remote sensing (RS)
AB Many deep-learning-based, domain adaptation methods for remote sensing applications rely on adversarial training strategies to align features extracted from images of different domains in a shared latent space. However, the performance of such representation matching techniques is negatively impacted when class occurrences in the target domain, for which no labelled data is available during training, are highly imbalanced. In this work, we propose a deep-learning-based representation matching approach for domain adaptation in the context of change detection tasks. We further evaluate the approach in a deforestation mapping application, characterized by a high-class imbalance between the deforestation and no-deforestation classes. The domains represent different sites in the Amazon and Brazilian Cerrado biomes. To mitigate the class imbalance problem, we devised an unsupervised pseudo-labeling scheme based on Change Vector Analysis that prevents the feature alignment to be biased towards the over-represented class. The experimental results indicate that the proposed approach can improve the accuracy of cross-domain deforestation detection. 
PY 2022
SO Ieee Geoscience And Remote Sensing Letters
SN 1545-598X
PU Institute of Electrical and Electronics Engineers (IEEE)
VL 19
IS 2504505
UT 000782795900001
DI 10.1109/LGRS.2022.3163575
ID 87439
ER

EF