- DataSet Quality Analysis
- Change Detection highlighter
- Features extraction and completion
- Provides several command line tools, you can combine together to build your own workflow
- Follows geospatial standards to ease interoperability and data preparation
- Build-in cutting edge Computer Vision model and loss implementations (and allows to replace by your owns)
- Support either RGB or multibands imagery (as multispectral or hyperspectral)
- Allows Data Fusion (from imagery or rasterized vectors)
- Web-UI tools to easily display, hilight or select results
- High performances, PyTorch based, and native multi GPUs support
rsp compare
Compute composite images and/or metrics to compare several XYZ dirs.rsp cover
Generate a tiles covering, in csv format: X,Y,Zrsp download
Downloads tiles from a remote server (XYZ, WMS, or TMS)rsp extract
Extracts GeoJSON features from OpenStreetMap .pbfrsp predict
Predict masks, from given inputs and an already trained modelrsp rasterize
Rasterize vector geospatial features (GeoJSON or PostGIS), to raster tilesrsp subset
Filter images in a slippy map dir using a csv tiles coverrsp tile
Tile a rasterrsp train
Trains a model on a datasetrsp vectorize
Extract simplified GeoJSON features from segmentation masks
pip3 install RoboSat.pink
conda create -n robosat_pink python=3.6
conda activate robosat_pink
pip install robosat.pink
git clone https://github.com/datapink/robosat.pink
pip3 -e ./robosat.pink
Ubuntu 18.04:
sudo sh -c "apt update && apt install -y build-essential python3-pip"
pip3 install RoboSat.pink && export PATH=$PATH:~/.local/bin
CentOS 7:
sudo sh -c "yum -y update && yum install -y python36 wget && python3.6 -m ensurepip"
pip3 install --user RoboSat.pink
Add Nvidia GPU drivers:
wget http://us.download.nvidia.com/XFree86/Linux-x86_64/418.43/NVIDIA-Linux-x86_64-418.43.run
sudo sh NVIDIA-Linux-x86_64-418.43.run -a -q --ui=none
- To test RoboSat.pink install, launch in a terminal:
rsp -h
- Upon your
pip
PATH setting, you could have to update it:export PATH=$PATH:.local/bin
- To use the development version:
pip3 install git+https://github.com/datapink/robosat.pink
- Requires: Python 3.6 or later, and Nvidia GPU with at least 6Go RAM.
- Training and validation datasets have to be tiled, using XYZ tiles format.
- A Dataset directory, so containing XYZ tiles, can be split as:
dataset
├── training
│ ├── images
│ └── labels
└── validation
├── images
└── labels
- Tiles images formats could be either: JPEG, WEBP, GeoTIFF, PNG.
- Tiles labels are expected to be PNG with single band and indexed palette.
- Tools producing XYZ tiles directory as output, also allows to easily generate a web map client, for visual inspection.
- Following schema, show several paths to create your own training dataset from several kinds of input data.
NOTA: several inputs connected to a single arrow point means a logical OR (e.g. WMS or XYZ or TMS).
RoboSat.pink use cherry-picked Open Source libs among Deep Learning, Computer Vision and GIS stacks.- Historical MapBox RoboSat github directory (not active anymore)
- Christoph Rieke's Awesome Satellite Imagery Datasets
- Mr Gloom's Awesome Semantic Segmentation
- The Lovász-Softmax loss: A tractable surrogate for the optimization of the IoU measure in neural networks
- U-Net: Convolutional Networks for Biomedical Image Segmentation
- Deep Residual Learning for Image Recognition
- Angiodysplasia Detection and Localization Using Deep Convolutional Neural Networks
- TernausNetV2: Fully Convolutional Network for Instance Segmentation
- Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps
-
Pull Requests are welcome ! Feel free to send code... Don't hesitate either to initiate a prior discussion throught tickets on any implementation question.
-
If you want to collaborate through code production and maintenance on a long term basis, please get in touch, co-edition with an ad hoc governance can be considered.
-
If you want a new feature, but don't want to implement it, DataPink provide core-dev services.
-
Expertise and training on RoboSat.pink are also provided by DataPink.
-
And if you want to support the whole project, because it means for your own business, funding is also welcome.
- Daniel J. Hofmann https://github.com/daniel-j-h
- Olivier Courtin https://github.com/ocourtin