SALaD
SALaD (Semi-Automatic Landslide Detection) is a landslide mapping system. SALaD utilizes Object-based Image Analysis and Random Forest to map landslides. It requires optical imagery, a DEM, corner coordinates of a training area, and manually mapped landslides within the training area. The code is built to run primarily on a Linux.
Installation
- Install Singularity
User can find thorough instruction for installation of Singularity >=3.0.0 on different operating systems (Linux, Windows or Mac) here:
https://sylabs.io/guides/3.0/user-guide/installation.html
- Build a Singularity container from Singularity definition file
Singularity definition files can be used as the target when building a container. Assuming user has the definition file called SALaD.def (see below), the container (named ilab-salad.sif) can be built with the command:
$ sudo singularity build ilab-salad.sif SALaD.def
For more details, check:
https://sylabs.io/guides/3.0/user-guide/build_a_container.html
If you don’t have root access on a Linux machine or want to host your container on the cloud, you can build the container on the Remote Builder: https://cloud.sylabs.io/builder
Executing SALaD from a container
singularity run -B <local_dir> < path_to_singularity_container>/ilab-salad.sif python <path_to_scripts>/driver.py -i "image.tif" -d "srtm.tif" -l "manual_landslide.shp" -lx 308335 -ly 3114295 -rx 312440 -ry 3109225 -rmi 2 -rma 32 -s 2 -p "<path_to_input_data_folder>" -op "<path_to_output_folder>" -r "landslide_SALaD.shp"
Example definition file (SALaD.def)
Bootstrap: docker
FROM: nvidia/cuda:10.1-cudnn7-devel-ubuntu18.04
%labels
Author Remi Cresson <remi.cresson[at]irstea[dot]fr>
Version v1.0.0
%help
========================================================================
- Orfeo Toolbox (without Tensor Flow)
========================================================================
%environment
# ------------------------------------------------------------------
# Add important environment variables
# ------------------------------------------------------------------
export PATH="$PATH:/work/otb/superbuild_install/bin/"
export PYTHONPATH="/work/otb/superbuild_install/lib/otb/python:$PYTHONPATH"
export OTB_APPLICATION_PATH="/work/otb/superbuild_install/lib/otb/applications"
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/work/otb/superbuild_install/lib/:/work/tf/installdir/lib/"
# set PYTHONPATH for access to OTB application
export PYTHONPATH="/usr/local/otb/src/innovation-lab:$PYTHONPATH"
%post
apt-get update -y \
&& apt-get upgrade -y \
&& apt-get install -y --no-install-recommends \
git
# retrieve OTB source from git repository and open permissions
mkdir -p /usr/local/otb
git clone --single-branch --branch otb-container https://github.com/nasa-nccs-hpda/innovation-lab.git /usr/local/otb
chmod a+rwx -R /usr/local/otb
apt-get update -y \
&& apt-get upgrade -y \
&& apt-get install -y --no-install-recommends \
sudo \
ca-certificates \
curl \
make \
cmake \
g++ \
gcc \
git \
libtool \
swig \
xvfb \
wget \
autoconf \
automake \
pkg-config \
zip \
zlib1g-dev \
unzip \
&& rm -rf /var/lib/apt/lists/*
# ---------------------------------------------------------------------------
# OTB and TensorFlow dependencies
# ---------------------------------------------------------------------------
apt-get update -y \
&& apt-get upgrade -y \
&& apt-get install -y --no-install-recommends \
freeglut3-dev \
libboost-date-time-dev \
libboost-filesystem-dev \
libboost-graph-dev \
libboost-program-options-dev \
libboost-system-dev \
libboost-thread-dev \
libcurl4-gnutls-dev \
libexpat1-dev \
libfftw3-dev \
libgdal-dev \
libgeotiff-dev \
libglew-dev \
libglfw3-dev \
libgsl-dev \
libinsighttoolkit4-dev \
libkml-dev \
libmuparser-dev \
libmuparserx-dev \
libopencv-core-dev \
libopencv-ml-dev \
libopenthreads-dev \
libossim-dev \
libpng-dev \
libqt5opengl5-dev \
libqwt-qt5-dev \
libsvm-dev \
libtinyxml-dev \
qtbase5-dev \
qttools5-dev \
default-jdk \
python3-pip \
python3.6-dev \
python3.6-gdal \
python3-setuptools \
libxmu-dev \
libxi-dev \
qttools5-dev-tools \
bison \
software-properties-common \
dirmngr \
apt-transport-https \
lsb-release \
gdal-bin \
&& rm -rf /var/lib/apt/lists/*
# ---------------------------------------------------------------------------
# Python packages
# ---------------------------------------------------------------------------
ln -s /usr/bin/python3 /usr/bin/python \
&& python3 -m pip install --upgrade pip \
&& python3 -m pip install pip six numpy wheel mock keras future
# ---------------------------------------------------------------------------
# Build OTB: Stage 1 (clone)
# ---------------------------------------------------------------------------
mkdir -p /work/otb \
&& cd /work/otb \
&& git clone https://gitlab.orfeo-toolbox.org/orfeotoolbox/otb.git otb \
&& cd otb \
&& git checkout release-7.0
# ---------------------------------------------------------------------------
# Build OTB: Stage 2 (superbuild)
# ---------------------------------------------------------------------------
mkdir -p /work/otb/build \
&& cd /work/otb/build \
&& cmake /work/otb/otb/SuperBuild \
-DUSE_SYSTEM_BOOST=ON \
-DUSE_SYSTEM_CURL=ON \
-DUSE_SYSTEM_EXPAT=ON \
-DUSE_SYSTEM_FFTW=ON \
-DUSE_SYSTEM_FREETYPE=ON \
-DUSE_SYSTEM_GDAL=ON \
-DUSE_SYSTEM_GEOS=ON \
-DUSE_SYSTEM_GEOTIFF=ON \
-DUSE_SYSTEM_GLEW=ON \
-DUSE_SYSTEM_GLFW=ON \
-DUSE_SYSTEM_GLUT=ON \
-DUSE_SYSTEM_GSL=ON \
-DUSE_SYSTEM_ITK=ON \
-DUSE_SYSTEM_LIBKML=ON \
-DUSE_SYSTEM_LIBSVM=ON \
-DUSE_SYSTEM_MUPARSER=ON \
-DUSE_SYSTEM_MUPARSERX=ON \
-DUSE_SYSTEM_OPENCV=ON \
-DUSE_SYSTEM_OPENTHREADS=ON \
-DUSE_SYSTEM_OSSIM=ON \
-DUSE_SYSTEM_PNG=ON \
-DUSE_SYSTEM_QT5=ON \
-DUSE_SYSTEM_QWT=ON \
-DUSE_SYSTEM_TINYXML=ON \
-DUSE_SYSTEM_ZLIB=ON \
-DUSE_SYSTEM_SWIG=OFF \
-DOTB_WRAP_PYTHON=OFF \
&& make -j $(grep -c ^processor /proc/cpuinfo)
# ---------------------------------------------------------------------------
# Build OTB: Stage 3 (bindings)
# ---------------------------------------------------------------------------
cd /work/otb/otb/Modules/Remote \
&& git clone https://github.com/remicres/otbtf.git \
&& cd /work/otb/build/OTB/build \
&& cmake /work/otb/otb \
-DOTB_WRAP_PYTHON=ON \
-DPYTHON_EXECUTABLE=/usr/bin/python3.6 \
-Dopencv_INCLUDE_DIR=/usr/include \
-DModule_OTBTensorflow=ON \
-DOTB_USE_TENSORFLOW=OFF \
-DTENSORFLOW_CC_LIB=/work/tf/installdir/lib/libtensorflow_cc.so \
-DTENSORFLOW_FRAMEWORK_LIB=/work/tf/installdir/lib/libtensorflow_framework.so \
-Dtensorflow_include_dir=/work/tf/installdir/include/ \
&& cd /work/otb/build/ \
&& make -j $(grep -c ^processor /proc/cpuinfo)
# ---------------------------------------------------------------------------
# Install Packages and Retrieve Source Code
# ---------------------------------------------------------------------------
pip3 install --upgrade richdem==0.3.4
pip3 install --upgrade fiona==1.8.13
pip3 install --upgrade geopandas==0.7.0
pip3 install --upgrade numba==0.49.1
pip3 install --upgrade pandas==1.0.3
pip3 install --upgrade peakutils==1.3.3
pip3 install --upgrade rasterstats==0.14.0
pip3 install --upgrade scikit-learn==0.21.3
# Below added 7/15/21
sudo apt-get update
sudo apt-get install -y libspatialindex-dev
pip3 install Pysal==1.14.4
pip3 install rtree==0.8.3
GitHub - nasa/SALaD at pythonawesome.com
SALaD (Semi-Automatic Landslide Detection) is a landslide mapping system. SALaD utilizes Object-based Image Analysis and Random Forest to map landslides. It requires optical imagery, a DEM, corner ...
nasa
