مرکز منطقه ای اطلاع رساني علوم و فناوري - The effectiveness of intelligent optimization techniques in camera calibration

Abstract :

In this paper, it is aimed to examine the effectiveness of the intelligent optimization algorithms to optimize the camera parameters with respect to the calibration method introduced by Luca Lucchese (LL). The motivation of the intelligent optimization algorithms is that they are so effective, flexible and easy adaptable for the real complex problems. The selected optimization algorithms are Artificial Bee Colony (ABC), Differential Evolution (DE), Genetic Algorithm (GA) and Particle Swarm (PSO). These algorithms except ABC have been used effectively for many complex problems. ABC has recently developed and its effectiveness has not been tested for a type of the camera calibration problem. But it is highly capable of generating good solutions for many benchmark functions such as Rosenbrock and Rastrigin with both low and very high dimensions. The other artificial intelligent optimization algorithms are also the first time being used in this camera calibration problem. In order to show the effectiveness of these intelligent optimization algorithms, their results have been compared with Levenberg-Marquardt (LM). the conventional derivative-based