Improved particle filter for object tracking

Robust real-time tracking of non-rigid objects is a challenging task. Color is a powerful feature for tracking deformable objects in image sequences with complex backgrounds. Color distribution is applied, as it is robust to partial occlusion, is rotation and scale invariant and computationally efficient. Particle filter has been proven very successful for non-linear and non-Gaussian estimation tracking problems. The article presents the integration of color distributions into particle filtering. A target is tracked with a particle filter by comparing its histogram with the histograms of the sample positions using the Bhattacharyya distance. Additionally, to solve the sample impoverishment (all particles collapse to a single point within a few iterations) in the particle-filter algorithm, a new resampling algorithm is proposed to tackle sample impoverishment. The performance of the proposed filter is evaluated qualitatively on various real-world video sequences. The experimental results show that the improved color-based particle filter algorithm can reduce sample impoverishment effectively and track the moving object very well.