What is it about?
We combine a convolutional neural network with a recurrent neural network in an encoder-decoder fashion to tackle the remote sensing image classification problem. We test the proposed solution on three benchmark remote sensing datasets (i.e., RS-19, UC-Merced and Brazilian Coffee Scenes datasets).
Featured Image
Why is it important?
Our solution achieves the state-of-the-art results on three benchmark remote sensing datasets which demonstrates the effectiveness of using automatically-learned features with deep neural networks.
Perspectives
I believe this article points out that there is need for large-scale remote sensing image classification datasets to generate more powerful and generalizable deep learning models.
Dr Ferda Ofli
Qatar Computing Research Institute
Read the Original
This page is a summary of: Recurrent Neural Networks for Remote Sensing Image Classification , IET Computer Vision, May 2018, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/iet-cvi.2017.0420.
You can read the full text:
Resources
Contributors
The following have contributed to this page
