Manual image annotation creates rectangular images with selected areas of interest (ROI). User opens image and selects rectangular areas of interest.
After selecting rectangles and pressing menu button program cuts rectangle images from the bigger image.
All parameters are saved in configuration INI file config.ini,
which is in the temp directory.
The output of the application is the set of rectangular images
in the temp directory.
Also it saves rectangular areas for the image in TXT format. You could open saved areas, modify them and cut image again.
Tested on Windows and Ubuntu for Python 3.7.
External libraries:
- Pillow to open images of various formats.
To start Manual Image Annotation with rectangles:
# Install additional packages
pip install pillow
# Run it
python runme.pySoftware architecture:
