Robust Object Tracking with Adaptive Fusion of Color and Edge Strength Local Mean Features Based on Particle Filter

Aiming at the problem of real time and robustness of visual object tracking in a clutter background, an adaptive fusion of color rectangle feature and edge strength local mean tracking algorithm based on particle filter is put forward. To improve the tracking speed and precision, integral image is used to quickly compute the color rectangle feature and edge strength local mean, besides fuzzy logic is applied to adaptively adjust each feature weight. The simulation results illustrate the algorithm is robust and effective.