Micro-object motion tracking based on the probability hypothesis density particle tracker

摘要: Tracking micro-objects in the noisy microscopy image sequences is important for analysis of dynamic processes biological objects. In this paper, an automated tracking framework proposed to extract trajectories micro-objects. This uses a probability hypothesis density particle filtering (PF-PHD) tracker implement recursive state estimation and association. order increase efficiency approach, elliptical target model presented describe using shape parameters instead point-like targets which may cause inaccurate tracking. A novel likelihood function, not only covering spatiotemporal distance but also dealing with geometric function based on Mahalanobis norm, improve accuracy weight update process PF-PHD tracker. Using framework, larger number tracks are obtained. The experiments performed simulated data microtubule movements real mouse stem cells. We compare nearest neighbor method multiple method. Our can simultaneously track hundreds sequence.