Multipurpose optimization of camera position by modeling image noise in camera for robot vision

Autonomous robots have been used in the manufacturing sector worldwide. However, they cannot be used in random or unknown environments because most of them are controlled by only sequence controls. Two cameras are used as vision sensors that can recognize the outside environment to solve this problem. We know that the positions of the two cameras affect the camera calibration accuracies. In this research, we aim at improving the camera calibration accuracies and propose a multipurpose optimization method of the cameras´ positions. First of all, we examine the influences of the two cameras´ positions for the calibration accuracies. In addition, we try to determine the optimal camera position from a simulation based on the image noise models obtained through experimentation. After that, we verify the effectiveness of the optimal camera position by conducting a discrimination experiment. As a result, we can model the image noises of two cameras mathematically and determine the optimal camera position. The discrimination accuracies are more than 0.99 and we prove that the proposed method is effective.