No abstract

In camera auto calibration, the major goal is to discover intrinsic parameter values that minimize the cost function. This study proposes to implement Bat algorithm, a stochastic optimization technique, to determine the optimum intrinsic parameter values. Each bat in the Bat Algorithm represents a potential solution to the issue, and each dimension in the Bat Algorithm's search space represents one of the basic parameters: skew, focal length, and magnification factor. The Kruppa's equation is the basis for the cost function in this study. By studying the echolocation behavior of the microbats, the bats will try to improve the fitness with each iteration. The Bat Algorithm's performance is evaluated using a case study from a database from Le2i Universite de Bourgoune. This paper studies the correlation of different parameters selection in Bat Algorithm in solving the camera auto-calibration problem. Finding shows that Bat Algorithm produces output that as expected as theory of Computational Intelligence suggested.