An estimation of central points of circle markers in a vision system by using Kalman filter and complementary filter

In the paper, the vision system for tracking of a center point of color markers with disturbance is presented. The authors have proposed the complementary filter algorithm (CF) as well as Kalman Filter (KF) to estimate a central position of the markers. To compute markers central position from marker area two different methods were used. First one is based on measurement of markers center mass. Second method is a circle fitting method. These two methods give two measurements of central point of marker, however when disturbance is significant or some parts of markers are invisible, their results may have considerable errors. It is very important to introduce robust tracking method of markers position. In the paper, two estimators, namely KF and CF are used and evaluated in terms of accurancy and tracking abilties. The research proves outstanding performance of the Kalman Filter in real time application of markers position tracking.