Dynamic object tracking based on marginal frequency group

A novel visual tracking algorithm based on Marginal frequency group is proposed in this paper. Firstly, we count the distribution area and gray levels of larger probability density in the Marginal frequency group. Then all the frequency of the Marginal frequency group is calculated, of which the weaker frequency gradient is removed by filtration. Finally, the stronger frequency gradient of Marginal frequency group that the distribution of high frequency information account for half node area is reserved. Experiments show that the proposed approach has good adaptability to variable structure tracking with real-time and robust tracking performance.