期刊
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
卷 17, 期 6, 页码 1087-1091出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LGRS.2019.2937872
关键词
Feature extraction; Deep learning; Computer architecture; Data mining; Remote sensing; Convolutional neural networks; Multilayer perceptrons; Aerial images; convolutional neural networks (CNNs); cross-entropy (CE); multilayer perceptron (MLP); neural architecture search network mobile (NASNet Mobile); transfer learning
Recently, deep convolutional neural networks (DCNNs) have gained great success in classifying aerial images, but in this area, the existence of the hard images, due to their innate characteristics, and weak focus of the network on them, due to the use of the cross-entropy (CE) loss, lead to reducing the accuracy of classification of aerial images. Moreover, since the last convolutional layer in a CNN has highly class-specific information, giving equal importance to all the channels causes to extract less discriminative features in comparison to weighting each of the channels adaptively. The fact that data labeling as well as creating ground truth on large data set is expensive is another point of concern in this regard. To address these problems, we have proposed a novel method for classification of aerial images. Our method includes proposing a new loss function, which enhances the focus of the network on hard examples by adding a new term to CE as a penalty term, bringing about the state-of-the-art results; designing a new multilayer perceptron (MLP) as a classifier, in which the used attention mechanism extracts more discriminative features by weighting each of the channels adaptively; and applying transfer learning strategy by adopting neural architecture search network mobile (NASNet Mobile) as a feature descriptor for the first time in the field of aerial images, which can mitigate the aforementioned costs. As indicated in the results, our proposed method outperforms the existing baseline methods and achieves state-of-the-art results on all three data sets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据