Splits photographs in a scanned image. If you have a pile of photographs to scan and archive, scan multiple photographs together and use photodetector to automatically generate separate photograph files.
Based on Python, OpenCV for image segmentation, tilt correction, trim, and recursive path traversal.
Prerequisites: git, python3.
Download repo:
git clone https://github.com/bebaek/photodetector.git
Install:
pip install photodetector
Run command line applicaton:
photodetector PATH
PATH is either filename or directory name. Resulting files will be saved in
out directory.
Use optional parameters as needed, for example:
photodetector --left-trim 40 --thresh 200 --outdir outgoing/ incoming/
Run photodetector -h to see the full list of options.
In order to use in your Python script or application, import the package and
use the main class ImageProcessor:
from photodetector import ImageProcessor
path = 'foo/*.jpg'
ip = ImageProcessor()
ip.load(path)
ip.extract_photos()
ip.save(outdir='foo-out')
Output images may have wrong orientations. Use the accompanying command to interactively fix that:
photorot SOURCE_FOLDER TARGET_FOLDER
Run the standard pip command:
pip uninstall photodetector
Prerequisites: git, bash, python3, python3-venv.
Download repo:
git clone https://github.com/bebaek/photodetector.git
Run setup script to create the virtual environment and install photodetector package in the development mode:
bash ci/setup-devenv.sh
Run unit tests as needed at the root of the repo:
python -m unittest
Conform to PEP 8 and check the code style:
flake8