Tools and resources for remote sensing image classification using convolutional neural networks (convnets).
Collection SAT CNN Models contains code and pretrained convnet models for classification of satellite images. The convnets are trained on publicly available SAT-4 and SAT-6 datasets of satellite images (http://csc.lsu.edu/~saikat/deepsat/). The collection contains the following datasets:
SAT-4 Models - CNN models trained on SAT-4 dataset
SAT-6 Models - CNN models trained on SAT-6 dataset
Tools & Results - code for convnet training and image classification, conference paper and presentation.
Collection HRRS Image Classification contains code and pretrained convnet models for classification of high resolution remote sensing (HRRS) images. The convnets are trained/fine-tuned on publicly available AID dataset of HRRS images (http://captain.whu.edu.cn/project/AID/). The collection contains the following datasets:
Trained convnet models - models trained from scratch on AID dataset
Fine-tuned convnet models - models pretrained on ImageNet and fine-tuned on AID dataset
Tools & Results - code of an experimental framework for training and testing different convnet topologies for HRRS image classification, conference paper and presentation.
The Space team has made the following datasets and collections publicly available. You must be a logged-in member of the Space to access all the datasets and collections.
There are no datasets associated with this Space
Viewing most recent collections View All Collections
There are no published datasets associated with this Space.