Camera Calibration

About Camera and Camera Calibration Ref: https://3dvision.princeton.edu/courses/COS429/2015fa/slides/02_camera/

Figure 1: Camera, generated with Leonardo.ai

prompt: On the left side, a person is staring at the sky and the sky is full of stars and also the Saturn is there. And the chair and trees are, too. And a 3D coordinate system axes are passing through the view. The three axes, x, y, and z should be all shown. Following one of the axes, there's a camera. Starting from the point where the camera is, the view is converted to a 2D world. There should be a coordinate system representation including 2 axes. On the right side, where axes for 2D are, the image frames are flying and piled up.

Cameras capture the 3D world and record the view into a 2D image, also called frame in video.
While converting the dimesion, distortion occurs and to correct it, we use 'calibration'. It's getting camera matrix and distortion coefficients, necessary for undistortion process.
This Repository contains the basic source codes for camera calibration and image undistortion using OpenCV.

Installation

1. Install the packages from the requirements.txt

while installing the packages, if the errors occur, please refer to ERROR_REFERENCE.md.

$ python -m venv .venv
$ source .venv/bin/activate (Mac)
  .venv\Scripts\Activate.ps1 (Windows)
$ pip install --upgrade pip
$ pip install -r requirements.txt

2. Install OpenCV

The file, requirements.txt doesn't contain opencv-python. It's because we can also install it with pip install but for better stability to use gstreamer, downloading the source file from the OpenCV website and building the codes manually is recommended.

The link for the source code and the guidance to use it is as follows:
https://docs.opencv.org/4.x/da/df6/tutorial_py_table_of_contents_setup.html

Usage

run the final cell of the ipynb file or execute the py file.

1. to Correct the distortion of an image:

• source code: calibration.py
• source code with comments to understand the process: calibration.ipynb

2. to Calibrate the camera with a video file:

• source code: calibration_with_mp4.py
• source code with comments to understand the process: calibration_with_mjepg.ipynb

3. to record a video with a camera

• source code: video_recording.py
• source code with comments to understand the process: video_recording.ipynb