charuco-calibration

Script to find the distortion parameters of a Raspberry Pi Camera. Tested with the Raspberry Pi Camera Module 3 WIDE.

Adapted from Using ChArUco boards in OpenCV.

You need a Charuco Board.

Grab a bunch of images at high resolution on the Raspberry Pi Camera covering all the field of view.

This command will result in about 60 images:

rpicam-still --timeout 120000 --timelapse 2000 -o image%04d.jpg

and move it to a fast computer (Python script tested on MacOS Sequoia):

python ./calibration-charuco.py

If the script fails at some image number, find the image file name for that number (marked with # prefix) and start over.

After a successful execution it will end up with a YAML file similar to this calibration_chessboard.yaml (obtained from a Raspberry Pi Camera Module 3 WIDE, Infrared) and at the end it will open two windows to show the comparison between original and undistorted images.