A particle filtering passenger target tracking method on Online Boosting belongs to target tracking technology or image processing technology. According to the method of the invention, through analyzing a particular condition in a tracking environment, a shielding matrix is established for mutual shielding in tracking passengers, and a weighing factor in a particle filtering model is formed, thereby effectively portraying the shielded passengers. Furthermore in an improved particle filtering model, an advantage of real-time portraying target updating by means of Online Boosting is utilized. Effective restriction is performed on an offset target in the tracking process, and a drift in the target tracking process can be effectively prevented, so that higher accuracy and higher robustness in tracking are realized. According to the method of the invention, effective information in a passenger motion process is sufficiently and effectively utilized. In passenger target tracking at next time, particle reconstruction is performed on an area around the target so that quick convergence of the model is realized, thereby realizing high calculating speed and high tracking accuracy. Furthermore the particle filtering passenger target tracking method has advantages of relatively high robustness to noise and relatively high algorithm stability.