Abstract
A method for distortion correction of un-calibrated lens with unkown collinear feature points is proposed and the improved fast distortion correction calculating method based on pre-computing template is implemented. Without known camera parameters and any known world coordinates point in 3D space, the distortion parameters of camera lens can be estimated based on the perspective projection invariant of straight lines. The proposed distortion correction method is easy to be performed and characterized by simple algorithm, fewer known conditions and fast computing speed. Some experiments to evaluate the performance of this approach on real images are reported, and the results show that this method is feasible for the machine vision systems. The improved method with pre-computing template drastically reduces the computation charge. And real-time applications of the image distortion correction are improved.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call