Abstract:

We propose a method of camera calibration based on genetic algorithms to improve the accuracy of camera calibration. The method extracts corners of chess board images and calculates the coordinates of the reference corners, then estimates intrinsic and external parameters of camera. Genetic algorithm is used to optimize the parameters extracted by the proposed method. Through evolution, selection and variation among generations, optimization is achieved by convergence. The result shows that the average projection deviation with the method is 0.255 9 pixel, and the standard deviation 0.139 3 pixel.

Key words: camera calibration, corner point extraction, genetic algorithms