Aiming at the problem of real time and reliability of visual object tracking in a clutter background

an adaptive fusion of color rectangle feature and Edge Orientation Histogram tracking algorithm based on Rao-Blackwellized Particle Filter is put forward. To improve the tracking speed and precision

first

Rao-Blackwellized Particle Filter is adopted to improve the filter performance

second

integral image is used to quickly compute the color rectangle feature and Edge Orientation Histogram

finally

fuzzy logic is applied to adaptively adjust each feature weight. The simulation result illustrates the algorithm is robust and effective at visible light and infrared object tracking under complex backgrounds.