https://github.com/qureshizawar/camera_optimisation
This project aims to use optimisation to build a real world model of a camera suitable for use in computer vision. The model is built considering a fixed focus, pin hole camera design.
https://github.com/qureshizawar/camera_optimisation
Last synced: about 2 months ago
JSON representation
This project aims to use optimisation to build a real world model of a camera suitable for use in computer vision. The model is built considering a fixed focus, pin hole camera design.
- Host: GitHub
- URL: https://github.com/qureshizawar/camera_optimisation
- Owner: qureshizawar
- Created: 2019-05-17T19:14:00.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2019-05-17T19:14:19.000Z (almost 6 years ago)
- Last Synced: 2025-01-13T09:11:59.563Z (4 months ago)
- Language: MATLAB
- Homepage:
- Size: 219 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# b1_camera_optimisation
This project aims to use optimisation to build a real world model of a camera suitable for use in computer vision.
The model is built considering a fixed focus, pin hole camera design. A grid of points was used as a calibration object
First the camera model and a calibration grid is positioned somewhere in space. A
noisy image of the grid is generated and some outliers added. A transformation matrix (which transforms
points from an object frame to the camera frame) called a homography is estimated with the help of
RANSAC to reject outliers. A number of images are used to estimate homographies from which a matrix
(k-matrix) entailing the intrinsic properties of the camera model is built. This estimated k-matrix is then
optimized to produce the model for the camera.
Load RunOptimiseKMatrix.m for full estimation and optimisation of a camera K- matrix
Please read Calibrating a digital camera for computer vision.pdf for more information