The proposed method is based on neural networks by modeling the data marginal distribution with the graph Laplacian built with both labeled and unlabeled samples, at the same time, optimizing neural networks layers in a single process, back-propagating the gradient of a Maximum Margin based objective function. Therefore, the proposed approach gives rise to an operational classifier, as opposed to previously presented semi-supervised scenarios. Results demonstrate the improved classification accuracy and scalability of this approach on SAR image classification problems.
CITATION STYLE
Sun, L., Li, X., Zhang, Q., & Ma, M. (2015). Semisupervised classification of SAR images by maximum margin neural networks method. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9242, pp. 513–521). Springer Verlag. https://doi.org/10.1007/978-3-319-23989-7_52
Mendeley helps you to discover research relevant for your work.