Estimation of Camera Parameters via Gray Wolf Optimization Algorithm

In machine vision for industrial growth, camera calibration accuracy is essential. In this paper, we propose an algorithm for camera calibration that combines an enhanced version of the two-step approach with Gray Wolf Optimization algorithm (GWO). First, the two-step process is enhanced to lessen distortion's impact on camera calibration. After employing the principle of slope invariance to determine the camera's distortion parameters, we employ Zhang's calibration technique to acquire the camera's internal and exterior parameters. The above-mentioned camera parameters are then optimized using a GWO approach that combines a dynamic inertia weight and a variation idea. Finally, the proposed algorithm is subjected to a series of comparative experiments, the results of which demonstrate that it achieves higher accuracy, overcomes the problem of slipping into local optimal more effectively, and achieves faster convergence in parameter optimization.