확장 칼만 필터링의 비젼 시스템 모델을 이용한 물체 치수 측정에 관한 연구

Abstract

To reduce the processing time using the vision system in real-time is very important and difficult. To overcome above problem, we have to develop the high-speed image processing algorithm, and effective computational scheme in estimating the parameters involving in vision system. Unfortunately, the traditional batch estimation requires much computational time because it has iteration algorithm. Thus, the batch estimation has difficulty in measuring object's dimension in real time. On the other hand, the Extended Kalman Filtering(EKF) has much advantages in that it takes explicitly into account the measurement uncertainties, and is a simple and efficient recursive procedures. Thus, this thesis is to develop and high-speed image processing algorithm and the EKF algorithm to compute the parameters in the vision control scheme in real time. This vision system model involves the six parameters to account for the cameras inner(orientation, focal length etc) and outer parameters(the relative location between object's dimension and camera). Also the EKF is applied to estimate the object's dimension used for object's operation. The practicality of computer vision control scheme based on the EKF is verified experimently by performing the object's dimension determination.