Camera Calibration Tool

This repository contains a Python-based camera calibration tool that uses OpenCV to calibrate cameras and correct lens distortion. The tool performs camera calibration using a chessboard pattern and provides functions for undistorting images.

Features

• Camera calibration using chessboard patterns
• Extraction of camera matrix, distortion parameters, rotation vectors, and translation vectors
• Image undistortion with two methods: standard undistortion and remapping
• Reprojection error calculation to assess calibration quality
• Support for batch processing of calibration images

Requirements

• Python 3.x
• OpenCV (cv2)
• NumPy
• Glob (for file searching)

See requirements.txt for the exact package versions.

Installation

1. Clone this repository:

   git clone <repository-url>
   

2. Install the required dependencies:

   pip install -r requirements.txt
   

Usage

1. Prepare a set of calibration images containing a chessboard pattern (default: 9x6 inner corners)
2. Place the images in a directory (e.g., calib_images/)
3. Run the script:

   python camera_calib.py
   

The script will:

• Process all images matching the pattern calib_images/*.jpeg
• Detect chessboard corners in each image
• Calculate camera calibration parameters
• Save undistorted versions of the images
• Calculate and display the reprojection error

Configuration

The calibration parameters can be adjusted in the CameraCaliberation class:

• chessboard_size: Number of inner corners per chessboard row and column (default: (9, 6))
• frame_size: Size of the images used for calibration (default: (1280, 720))
• criteria: Termination criteria for corner sub-pixel accuracy

Classes and Methods

CameraCaliberation

Main class for performing camera calibration:

• caliberate(img_loc): Performs camera calibration using images from the specified location
• undistort(cameraMatrix, distParam, img_dist, remap=False): Corrects distortion in a single image
• reprojection_error(rvecs, tvecs, cameraMatrix, distParam): Calculates reprojection error

Output

The script produces:

• Camera matrix and distortion parameters
• Undistorted images saved as undistorted_image.jpg or undistorted_image_remap.jpg
• Reprojection error metrics

License

This project is licensed under the MIT License - see the LICENSE file for details.