DDPSC PhenotyperCV

PhenotyperCV is a header-only C++11 library that has a large range of functionality for image-based plant phenotyping.

Included is a single source file that, when built, has workflows for many pratical applications. The created executable has multiple features that are selected for using -m flag indicating the "mode" you'd like execute. As development continues, new modes will become available to expand the use of this program to problems outside the framework of Bellwether platform. Futher information can be found using the -h flag.

Custom workflows, with the aid of functions in this library, are also made possible and are intended to ease the difficulty in processing images from any source. A common bottleneck of image processing is not knowing how to perform specific tasks and this library is designed with that in mind.

Usage

Visit the wiki page for more info on how to use included source file and how to create your own custom workflows.

Building the program on a local computer

PhenotyperCV is dependent on two packages: OpenCV and Eigen3. Additionally, the OpenCV installation must have been with the extra modules enabled, namely: aruco, ml, and ximgproc. This program must be compiled from source and is made easier with cmake

(Mac Only) Installing Homebew

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

Installing CMake

For Ubuntu users:

sudo apt-get install cmake

For MacOSX users:

brew install cmake

Making a directory for all builds

mkdir big_build && cd big_build
export my_path=$(pwd)

Getting Eigen3 and installing

wget "https://gitlab.com/libeigen/eigen/-/archive/3.3.7/eigen-3.3.7.tar.bz2" -O"eigen-3.3.7.tar"
tar -xf eigen-3.3.7.tar && cd eigen-3.3.7
mkdir build && cd build
cmake -DCMAKE_INSTALL_PREFIX=/usr/local ..
sudo make install
sudo cp -r /usr/local/include/eigen3/Eigen /usr/local/include/
cd $my_path

Getting OpenCV extra modules

wget https://github.com/opencv/opencv_contrib/archive/4.5.2.tar.gz -O"opencv_contrib-4.5.2.tar.gz"
tar -xzf opencv_contrib-4.5.2.tar.gz
cd opencv_contrib-4.5.2/modules
rm -rv !("aruco"|"ximgproc") 
cd $my_path

Getting OpenCV and installing

wget https://github.com/opencv/opencv/archive/4.5.2.tar.gz -O"opencv-4.5.2.tar.gz"
tar -xzf opencv-4.5.2.tar.gz
cd opencv-4.5.2
mkdir build && cd build
cmake .. -DOPENCV_EXTRA_MODULES_PATH=$my_path/opencv_contrib-4.5.2/modules
make -j8
sudo make install
cd $my_path

Getting ZBar and installing

For Ubuntu users:

sudo apt-get install libzbar-dev libzbar0

For MacOSX users:

brew install zbar

Getting PhenotyperCV and building

git clone https://github.com/jberry47/ddpsc_phenotypercv
cd ddpsc_phenotypercv && mkdir build && cd build
cmake ..
make
export PATH=$PATH:$my_path/ddpsc_phenotypercv/build

Building the program on DDPSC Infrastructure

Both OpenCV and Eigen3 depedencies are in unconventional locations that are not found with cmake. To build the program on the infrastructure, there is a bash script in the misc directory of this repo that links to everything needed, and can be run with:

cd ddpsc_phenotypercv/misc
./pull_compile_phenocv.sh
cd ../../

Alternatively, a pre-built executable exists already in /home/jberry/programs/PhenotyperCV.

After a successful build, you need to include the libraries to OpenCV and ZBar in your LD_LIBRARY_PATH

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/shares/bioinfo/installs/opencv-3.3.0/install/lib:/bioinfo/lib

It is recommended that you add this line to your bash_profile or bash_rc so you don't have to add the libraries every time you want to use PhenotyperCV.

Testing the build

After you create the executable, test to make sure it can be used by running PhenotyperCV -h which should display a help page.

DDPSC Datasci Cluster Usage

Example condor job file and accompanying executable for processing images on the DDPSC infrastructure can be found in the misc directory

• phenotypercv.submit - condor submit file that MUST BE EDITED to your file paths and image location
• run_phenocv.sh - the executable that the job file calls and MUST BE EDITED to your file paths and image location