Camera calibration based on hybrid differential evolution and crayfish optimization algorithm

This study proposes a hybrid differential evolution and crayfish optimization algorithm (HDECOA) for precise camera calibration . HDECOA synergizes differential evolution, enhanced by an adaptive parameter control strategy , with crayfish optimization algorithm within a parallel and competitive framework. The hybrid algorithm achieves an effective balance between exploration and exploitation by improving population diversity and optimizing evolutionary efficiency. Finally, HDECOA is applied to calibrate two cameras with distinct parameters. Experimental comparisons evaluate the mean reprojection error of the proposed method against those of methods employing crayfish optimization algorithm, particle swarm optimization , differential evolution, sparrow search algorithm, and Zhang's method. K -means cluster analysis is utilized to evaluate reprojection errors and relative reprojection errors are calculated under varying levels of Gaussian noise . The proposed method achieves mean reprojection errors of 0.054 pixels and 0.166 pixels for the two cameras, respectively. Comprehensive experimental results reveal rapid convergence, high accuracy, robust performance, and versatility of the proposed method, highlighting its superiority over the comparison methods. • A novel hybrid algorithm called HDECOA is proposed for camera calibration. • HDECOA uses parallel and competitive design for improved calibration performance. • Calibration results are analyzed using K -means clustering and back-projection. • HDECOA achieves fast convergence, high accuracy, and robustness.