Makes One Image Overview (OIO) from video scan of threaded insert. The whole video scan has to include axial movement only in one direction.

Before running, navigate to configuration file located at /src/config/config.py. Define paths to input and output folder and other configurable parameters, mostly to adapt it to your environment.

• --mode {single,multiple}: Specify the mode of operation (single video or multiple videos). If single you have to use argument --video_name. If multiple all files from INPUT_FOLDER (which is specified in configuration file config.py) are processed as videos.
• --video_name: (Required if using single mode) Path to the video file.
• --calc_rot_per_frame: (Default is False) Set to True to calculate its own rotation per frame, not using the precalculated one. Not recommended since it greatly increases computational time.

To build the application image:

docker build -t oio-pipeline .

To run the application using Docker, you need to specify the config file path and input and output directories. Use the -v flag to bind your local paths. Examples:

docker run -v /path/to/input:/input -v /path/to/output:/output -v /path/to/config/config.py:/app/src/config/config.py oio-pipeline --mode multiple

docker run -v /path/to/input:/input -v /path/to/output:/output -v /path/to/config/config.py:/app/src/config/config.py oio-pipeline --mode single --video_name "video_name.mp4"

1. Download and install FFmpeg

2. Install requirements, for example with pip:

pip install -r requirements.txt

3. Run the main script, for example:

python src/scripts/main.py --mode single --video_name "video_name.mp4"