Towards better exploiting convolutional neural networks for remote sensing scene classification

作者：

Highlights：

• 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.

摘要

Highlights•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.

论文关键词：Deep learning,Convolutional neural networks,Fine-tune,Feature extraction,Aerial scenes,Hyperspectral images,Remote sensing

论文评审过程：Received 1 February 2016, Revised 1 June 2016, Accepted 1 July 2016, Available online 2 July 2016, Version of Record 13 October 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.07.001