An implimentation of the following article
- https://shop.cqpub.co.jp/hanbai/books/MIF/MIF201805.html
- "特別企画 あなたの知らない生体計測の世界"
you can detect blood vessels from a RGB JPEG image. No need to use IR camera
git clone https://github.com/ice-github/ImageInformationAnalyzer
cd ImageInformationAnalyzer
git submodule init
git submodule update
mkdir ThirdParty/install
INSTALL_DIR=`pwd`/ThirdParty/install
cd ThirdParty/Eigen
mkdir build && cd build
cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_DIR ..
make && make install
cd ../../
*you should consider removing "build" directory because CMake may refer the build directory as Eigen3 dir
cd ThirdParty/glog
mkdir working && cd working
cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_DIR -DWITH_GFLAGS=false ..
make && make install
cd ../../
cd ThirdParty/ceres-solver
mkdir working && cd working
cmake -DCMAKE_INSTALL_PREFIX=$INSTALL_DIR -DEIGEN_INCLUDE_DIR=$INSTALL_DIR/include/eigen3 -DMINI GLOG=true -DBUILD_EXAMPLES=false -DBUILD_TESTING=false -DGFLAGS=false ..
make && make install
cd ../../
*you may encount errors about map/set but the errors can be fixed by adding std::, e.g. std::map
cmake -DOpenCV_DIR=${opencv cmake directory} -DEigen3_DIR=$INSTALL_DIR/share/eigen3/cmake -DCeres_DIR=$INSTALL_DIR/CMake ..
*OpenCV_DIR indicates the directory of OpenCVConfig.cmake file. e.g. opencv/build
./sample ${Project directory}/img/Hand.jpg
*you can change the luminance by using the setting window and "Apply" button