A comparison of visual target tracking methods in noisyenvironments

摘要

Motion detection and tracking are useful in computer vision. Thiswork represents a comparison of several methods for target tracking innoisy environments using motion-energy and template location basedapproaches. In order to remove the registration noise, a morphologicalfilter has been adopted. Since the scale and rotation of the targetchange, a template update method has been used. The performance ofdifferent methods to compare their respective speed and accuracy havebeen evaluated and the results obtained are presented