RoboSat.pink

Computer Vision framework for GeoSpatial imagery

Purposes:

• DataSet Quality Analysis
• Change Detection highlighter
• Features extraction and completion

Main Features:

• 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, Data Augmentation and Loss implementations (and allows to replace by your owns)
• Support either RGB and multibands imagery, and allows Data Fusion
• Web-UI tools to easily display, hilight or select results (and allow to use your own templates)
• High performances
• Eeasily extensible by design

Documentation:

Tutorials:

• RoboSat.pink 101
• How to use plain OpenData to create a decent training OpenDataSet
• How to extend RoboSat.pink features, and use it as a Framework

Config file:

• RoboSat.pink configuration file

Tools:

• rsp cover Generate a tiles covering, in csv format: X,Y,Z
• rsp download Downloads tiles from a remote server (XYZ, WMS, or TMS)
• rsp extract Extracts GeoJSON features from OpenStreetMap .pbf
• rsp rasterize Rasterize vector features (GeoJSON or PostGIS), to raster tiles
• rsp subset Filter images in a slippy map dir using a csv tiles cover
• rsp tile Tile raster coverage
• rsp train Trains a model on a dataset
• rsp export Export a model to ONNX or Torch JIT
• rsp predict Predict masks, from given inputs and an already trained model
• rsp compare Compute composite images and/or metrics to compare several XYZ dirs
• rsp vectorize Extract simplified GeoJSON features from segmentation masks
• rsp info Print RoboSat.pink version informations

Presentations slides:

• @SOTM 2019
• @FOSS4G 2019

Installs:

With PIP:

pip3 install RoboSat.pink                                     # For latest stable version

or

pip3 install git+https://github.com/datapink/robosat.pink     # For current dev version

With Ubuntu 19.04, from scratch:

sudo sh -c "apt update && apt install -y build-essential python3-pip"
pip3 install RoboSat.pink && export PATH=$PATH:~/.local/bin
wget http://us.download.nvidia.com/XFree86/Linux-x86_64/430.40/NVIDIA-Linux-x86_64-430.40.run
sudo sh NVIDIA-Linux-x86_64-430.40.run -a -q --ui=none

With CentOS 7, from scratch:

sudo sh -c "yum -y update && yum install -y python36 wget && python3.6 -m ensurepip"
pip3 install --user RoboSat.pink
sudo sh -c "yum groupinstall -y 'Development Tools' && yum install -y kernel-devel epel-release"
wget http://us.download.nvidia.com/XFree86/Linux-x86_64/430.40/NVIDIA-Linux-x86_64-430.40.run
sudo sh NVIDIA-Linux-x86_64-430.40.run -a -q --ui=none

NOTAS:

• Requires: Python 3.6 or 3.7
• GPU is not strictly mandatory, but rsp train and rsp predict would be -that- slower without.
• To test RoboSat.pink install, launch in a terminal: rsp -h or rsp info
• Upon your pip PATH setting, you may have to update it: export PATH=$PATH:.local/bin
• If needed, to remove pre-existing Nouveau driver: sudo sh -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf && update-initramfs -u && reboot"

Architecture:

RoboSat.pink use cherry-picked Open Source libs among Deep Learning, Computer Vision and GIS stacks.

GeoSpatial OpenDataSets:

• Christoph Rieke's Awesome Satellite Imagery Datasets
• Zhang Bin, Earth Observation OpenDataset blog
• Wehwu's Awesome Remote Sensing Change Detection

Bibliography:

• U-Net: Convolutional Networks for Biomedical Image Segmentation
• Deep Residual Learning for Image Recognition
• Aggregated Residual Transformations for Deep Neural Networks
• Wide Residual Networks
• TernausNetV2: Fully Convolutional Network for Instance Segmentation
• The Lovász-Softmax loss: A tractable surrogate for the optimization of the IoU measure in neural networks
• Albumentations: fast and flexible image augmentations

Contributions and Services:

• Pull Requests are welcome ! Feel free to send code... Don't hesitate either to initiate a prior discussion via gitter or ticket on any implementation question. And give also a look at Makefile rules.

• 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.

Requests for funding:

We've already identified several new features and research papers able to improve again RoboSat.pink, your funding would make a difference to implement them on a coming release:

• Increase (again) prediction accuracy :

  • on low resolution imagery
  • even with few labels
  • feature extraction when they are (really) close
  • with multibands and Data Fusion

• Add support for :

  • MultiClass classification
  • Linear features extraction
  • PointCloud data support
  • Time Series Imagery
  • StreetView Imagery

• Improve (again) performances

Authors:

• Olivier Courtin https://github.com/ocourtin
• Daniel J. Hofmann https://github.com/daniel-j-h

Citing:

  @Manual{,
    title = {{RoboSat.pink} Computer Vision framework for GeoSpatial Imagery},
    author = {Olivier Courtin, Daniel J. Hofmann},
    organization = {DataPink},
    year = {2019},
    url = {http://RoboSat.pink},
  }