A simple OpenCV checkerboard camera calibration python package
Project description
OpenCV Camera Calibration
A simple tool for checkerboard camera calibration if you are tired of re-writing from the OpenCV docs on camera calibration.
Install
pip install opencv_calibrate
Prerequisites
This package relies on ffmpeg, please ensure it is installed. See python ffmpeg for more.
Usage
import opencv_calibrate
# image
image_directory_path = "example_images/"
# image params
checkerboard = (10, 7)
ret, camera_parameters = opencv_calibrate.image(image_directory_path, checkerboard)
# or with video
# video
video_path = "/path/to/video"
# video params
ret, camera_parameters = opencv_calibrate.video(video_path)
Then we can acess the parameters in the camera_parameters object:
# camera intrinsic matrix
camera_matrix = camera_parameters.camera_matrix
# distortion matrix
distortion_matrix = camera_parameters.distortion_matrix
# rotation vectors
rotation_vecs = camera_parameters.rotation_vecs
# translation vectors
translation_vecs = camera_parameters.translation_vecs
# projection error
projection_error = camera_parameters.projection_error
# save parameters
camera_parameters.save("/path/to/dir/")
# load
from opencv_calibrate import CameraParameters
# from npz
camera_parameters.load_from_npz("/path/to/params.npz")
# from yaml
camera_parameters.load_from_yaml("/path/to/params.yaml")
or from the command line (this is also callable by python -m opencv_calibrate.calibrate
):
calibrate [-h] [--video VIDEO] [--image_dir IMAGE_DIR] [--output_dir OUTPUT_DIR] [--checkerboard CHECKERBOARD]
Camera calibration of videos and images containing checkerboard run with DEBUG>1 for more outputs
options:
-h, --help show this help message and exit
--video VIDEO Path to the video file
--image IMAGE_DIR
Path to the directory containing images
--output_dir OUTPUT_DIR
Path to the output directory
--checkerboard CHECKERBOARD
Number of rows and columns in comma seperated format e.g. "9, 6"
This will output a YAML file with the parameters:
camera_matrix:
- [f_x, 0.0, c_x]
- [0.0, f_y, c_y]
- [0.0, 0.0, 1.0]
distortion_matrix:
- [k1, k2, p1, p2 k3]
projection_error: float
and a .npz numpy file with addition of the extrinsic matrix.
Run with DEBUG environment variable for more outputs, e.g.
DEBUG=1 calibrate --image /path/to/images
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
opencv_calibrate-0.1.0.tar.gz
(8.3 kB
view hashes)
Built Distribution
Close
Hashes for opencv_calibrate-0.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | bf7661727b84bb09142c78359ccada38d2ff6a55a28b690330951d85bc5f1113 |
|
MD5 | 7650a03da83c113fed88d278b313147b |
|
BLAKE2b-256 | 43a5678e8b850c41ccb1c62e861e90a781a289a94448e624927ca91f8d48c510 |