A Preliminary Study on Camera Auto Calibration Problem Using Bat Algorithm

The main objective in camera auto calibration is to find intrinsic parameters values that minimize the cost function. This paper attempts to implement a stochastic optimization algorithm called Bat Algorithm in order to find optimal values of the intrinsic parameters. Each bat in Bat Algorithm represents a candidate solution for the problem and each dimension in the search space of the Bat Algorithm represents a parameter of intrinsic parameters: skew, focal length, and magnification factor. The cost function used in this paper is based on the Kruppa’s equation. For each iteration, the bats will try to improve its fitness by following the echolocation behavior of the microbats. A case study taken from database, provided by Le2i Universite de Bourgoune is used to evaluate the performance of the Bat Algorithm. The result obtained indicates potential application with further improvement required.