The setting of inertia weight in the particle swarm optimization algorithm is crucial and has influence on the algorithm properties. In order to improve the operation efficiency of the algorithm, an improved particle swarm optimization is proposed and applied in the field of target tracking. First of all, the corresponding parameters of particle swarm optimization algorithm are initialized. Secondly, based on the introduction of the concept of particle optimization rate, inertia weight can be accurately adjusted according to the particle state in a timely manner, meanwhile, the speed and position of particles can be updated. Finally, there are optimization of objective similarity function and achievement of accurate positioning. Experimental results indicate that the method is well adapted to the situation when partial occlusion occurs in object tracking, and improves the operation efficiency with better realtime.