A novel camera calibration method based on genetic algorithm

A camera calibration method based on genetic algorithm to conventional non-linear camera calibration optimization methods was presented in this paper. Contrary to the classical non-linear photogrammetric approach, the proposed technique can correctly find the solution without the need of initial guesses. In this method, a more comprehensive camera model was adopted which based on the pinhole model, extended with radial distortions, decentering distortion and thin prism distortions correction. In the process of calibration, calibration precision is improved by imaging at different locations in the whole measurement space. The excellent performance of the proposed technique, such as convergence, accuracy and robustness, was demonstrated.