The 2018 IEEE GRSS Data Fusion Contest, organized by the IADF TC in collaboration with the University of Houston, was opened on January 15, 2018. The test phase (and the evaluation server) was opened on March 13 and closed on March 25. We received over 1300 submissions!

We would like to thank all the participants for their submissions. Evaluation and ranking were conducted by the Organizers, who carefully checked all valid submissions. The winners are reported below along with a brief overview of their methods and their Overall Accuracy (OA) and Kappa statistic over all the undisclosed ground truth pixels.


1st Place

(1st Place in Data Fusion Classification Challenge & 1st Place in Multispectral LiDAR Classification Challenge)
Team Name: Gaussian
Authors: Yonghao Xu, Bo Du, and Liangpei Zhang (Wuhan University, China)
Approach: Fully convolutional networks and post-classification with topological relationships among different objects
OA = 80.78 kappa = 0.80 (Data Fusion Classification Challenge)
OA = 81.07 kappa = 0.80 (Multispectral LiDAR Classification Challenge)


2nd Place

(2nd Place in Data Fusion Classification Challenge)
Team Name: dlrpba
Authors: Daniele Cerra, Miguel Figueiredo Vaz Pato, Emiliano Carmona, Jiaojiao Tian, Seyed Majid Azimi, Rupert Müller, Ksenia Bittner, Corentin Henry, Eleonora Vig, Franz Kurz, Reza Bahmanyar, Pablo d‘Angelo, Kevin Alonso, Peter Fischer, and Peter Reinartz (German Aerospace Center, Germany)
Approach: Deep convolutional and shallow neural networks on a simplified set of classes, completed by a series of specific detectors and ad hoc classifiers
Metrics: OA = 80.74 kappa = 0.80 (Data Fusion Classification Challenge)


3rd Place

(1st Place in Hyperspectral Classification Challenge)
Team Name: challenger
Authors: Shuai Fang, Dou Quan, Lei Zhang, and Ligang Zhou (Xidian University, China)
Approach: Two-branch convolutional neural network
Metrics: OA = 77.39 kappa = 0.73 (Hyperspectral Classification Challenge)


3rd Place, ex aequo

(2nd Place in Multispectral LiDAR Classification Challenge & 3rd Place in Data Fusion Classification Challenge)
Team Name: AGTDA
Authors: Sergey Sukhanov, Dmitrii Budylskii, Ivan Tankoyeu, Roel Heremans, and Christian Debes (AGT International, Germany)
Approach: Ensemble learning based on several classifiers, including convolutional neural networks, gradient boosting machines, and random forests, followed by post-processing techniques
OA = 79.79 kappa = 0.79 (Data Fusion Classification Challenge)
OA = 78.05 kappa = 0.77 (Multispectral LiDAR Classification Challenge)

Each team will present their methods in a special session at the forthcoming IGARSS conference in Valencia that will take place on Wednesday, July 25 (session WE1.R5: IEEE GRSS Data Fusion Contest). Their papers will be included in the IGARSS proceedings.