A New Method for Uncertainty Evaluation of Centroid Detection

As a method of image feature extraction, centroid detection algorithm has been applied in many fields. Uncertainty evaluation of centroid detection is an important approach to evaluating the reliability of centroid detection. This paper presents a new method for uncertainty evaluation of centroid detection. The uncertainty associated with the intensity of a pixel, which involves in centroid detection, is firstly evaluated from the pixel intensity and the image gradient at the pixel. Then the uncertainties associated with centroid detection are evaluated by means of Monte Carlo Simulation. The mathematical model used to evaluate the uncertainty associated with pixel intensity is presented in the paper. The experiment results show that the statistical and evaluated uncertainties of centroid detection are quite close.