Towards better exploiting convolutional neural networks for remote sensing scene classification

• Published in: Pattern recognition Vol. 61; pp. 539 - 556
• Main Authors: Nogueira, Keiller, Penatti, Otávio A.B., dos Santos, Jefersson A.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2017
• Subjects: Aerial scenes; Convolutional neural networks; Deep learning; Feature extraction; Fine-tune; Hyperspectral images; Remote sensing; Deep learning; Fine-tune; Hyperspectral images; Aerial scenes; Feature extraction; Convolutional neural networks; Remote sensing
• ISSN: 0031-3203; 1873-5142
• DOI: 10.1016/j.patcog.2016.07.001

We present an analysis of three possible strategies for exploiting the power of existing convolutional neural networks (ConvNets or CNNs) in different scenarios from the ones they were trained: full training, fine tuning, and using ConvNets as feature extractors. In many applications, especially including remote sensing, it is not feasible to fully design and train a new ConvNet, as this usually requires a considerable amount of labeled data and demands high computational costs. Therefore, it is important to understand how to better use existing ConvNets. We perform experiments with six popular ConvNets using three remote sensing datasets. We also compare ConvNets in each strategy with existing descriptors and with state-of-the-art baselines. Results point that fine tuning tends to be the best performing strategy. In fact, using the features from the fine-tuned ConvNet with linear SVM obtains the best results. We also achieved state-of-the-art results for the three datasets used. •Analysis of the generalization power of ConvNets for remote sensing datasets.•Comparative analysis of ConvNets and low-level and mid-level feature descriptors.•Evaluation and analysis of three strategies to exploit existing ConvNets in different scenarios.•Evaluation of ConvNets with state-of-the-art baselines.