This project contains a suite of camera calibration proecdures for opencv-python

Current implementation contains:

• Single camera calibration using aruco marker configuration.

Installation

pip install git+https://github.com/hcilab-uofm/camera_calibration_suit

Usage

Single camera calibration

1. Generate the marker configuration image:

   run-calibration generate-chessboard

2. Capture a set of images from the camera of the chessboard pattern geneated and place them in a directory. You can use the following for that

   run-calibration capture-images captured_chessboard_images

   Where captured_chessboard_images is a directory.

3. Run the process-chessboard-images command:

   run-calibration process-chessboard-images captured_chessboard_images JPG
   

   This saves the parameters in a file named camera_parameters.npy Assuming the direcotry containing the images is named captured_chessboard_images, and the image extension is JPG. Use -d to display the images used to generated the parameters.

4. Use the generated parameters to undistort an image from the camera:

import cv2 import numpy as np image = cv2.imread("test.jpg") camera_matrix, distortion_coefficients = np.load("camera_parameters.npy", allow_pickle=True) cv2.undistort(image, camera_matrix, distortion_coefficients, None)

Capture images

To capture images you can use the following command:

poetry run run-calibration capture-images image-dir

Where images will be stored in image-dir. Also has the following options:

• -e, (--image-ext): Image extension to use (defult jpg).
• -c (--camera-or-file): camera index or file name to pass to opencv. Defaults to 0, which is the 0 index camera.
• -t, (--wait-time-ms): Wait time for each frame in milliseconds (default 30).
• -h, (--display-height): Display height (default 720).
• --help

This can be used as a common tool as well.