An Improved Resampling Algorithm for Particle Filtering in Small Target Tracking

Zhou, R. and Teng, J., 2015. An improved resampling algorithm for particle filtering in small target tracking.

The sample impoverishment after resampling is the main reason of the decrease of the estimation accuracy in particle filtering. To solve the problem, a new resampling algorithm is proposed in this paper. First, the existing resampling algorithm is analyzed and its defects are demonstrated; then an improved algorithm is introduced to overcome the defects by constructing the new particles based on Gaussian distribution. Compared to the existing resampling algorithm, the proposed algorithm can maintain the diversity of particles and avoid the sample impoverishment in particle filtering, thus increase the estimation accuracy. Simulation results show that the accuracy of tracking small target in low signal-to-noise ratio optimal image sequence is increased in terms of the quantitative criteria-RMSE.