Convolutional Neural Networks (CNN) have been being widely employed to solve the challenging remote sensing task of aerial scene classification. Nevertheless, it is not straightforward to find single CNN models that can solve all aerial scene classification tasks, allowing the development of a better alternative, which is to fuse CNN-based classifiers into an ensemble. However, an appropriate choice of the classifiers that will belong to the ensemble is a critical factor, as it is unfeasible to employ all the possible classifiers in the literature. Therefore, this work proposes a novel framework based on meta-heuristic optimization for creating optimized ensembles in the context of aerial scene classification. The experimental results were performed across nine meta-heuristic algorithms and three aerial scene literature datasets, being  compared in terms of effectiveness (accuracy), efficiency (execution time), and behavioral performance in different scenarios.  Our results suggest that the Univariate Marginal Distribution Algorithm shows more effective and efficient results than other commonly used meta-heuristic algorithms, such as Genetic Programming and Particle Swarm Optimization.

@inproceedings{Ferreira2020:ICPR,
 author    = {Álvaro R. Ferreira Jr and  Gustavo H. de Rosa and  João P. Papa and Gustavo Carneiro and  Fabio A. Faria},
 title     = {Creating Classifier Ensembles through Meta-heuristic Algorithms for Aerial Scene Classification},
booktitle = {Intl. Conf. on Pattern Recognition (ICPR)},
 year      = {2020},
 pages     ={to appear}
 }

Code: https://github.com/gugarosa/evolutionary_ensembles/edit/master/README.md