This was written to process C. elegans chemotaxis behavior and follows a simple flow: if I know how many worms there are in an experiment (say, 8) then I should be able to start from a point where I can see all 8 worms and track outwards from that. This was written to see whether that is a better way of doing things.
YAWT uses CMake with >= Qt 6.10 and OpenCV. Basler Pylon support is not part of the current build.
Install Qt 6 and OpenCV built for the same compiler you use with CMake. For Visual Studio 2022, OpenCV's CMake package directory is commonly:
C:\opencv\build\x64\vc16\libConfigure with explicit Qt and OpenCV package paths:
cmake -S . -B build\windows-msvc -G Ninja `
-DCMAKE_PREFIX_PATH=C:\Qt\6.7.0\msvc2019_64 `
-DOpenCV_DIR=C:\opencv\build\x64\vc16\lib
cmake --build build\windows-msvcThe same values can be set in Qt Creator under the kit's CMake configuration:
CMAKE_PREFIX_PATH=C:/Qt/6.7.0/msvc2019_64
OpenCV_DIR=C:/opencv/build/x64/vc16/lib
To build and package from an x64 MSVC command prompt with Qt on PATH:
set OpenCV_DIR=C:\opencv\build
scripts\package_windows_msvc.cmd --cleanThe packaged app is written to dist\windows\yawt, with a zip archive beside it.
The packaging script copies OpenCV runtime DLLs and explicitly looks for opencv_videoio_ffmpeg*.dll and openh264*.dll. Those are needed for OpenCV MP4/H264 video I/O, including crop export. If the script prints a warning for either DLL, copy the missing DLL from your OpenCV install next to yawt.exe, or use an OpenCV build that includes the FFmpeg video I/O backend.