Tracking swimmers bacteria and pores within a biofilm

Along the food chain, most wet surfaces are covered by biological pellicles called biofilms. It has also been demonstrated that bacterial stealth swimmers create transient pores that increase macromolecular transfer within the biofilm. Therefore, it is interesting to investigate the behavior of swimmers bacteria within the biofilm. In this paper we propose a novel approach to track the swimmers as well as pores within the biofilm. A direct time-lapse confocal microscopy technique enables the visualization of swimmers as well as biofilm in two color images. In these two images swimmers and pores dynamics are correlated and we reinforce the tracking of swimmers with the tracking of pores and vice and versa. The proposed algorithm is built upon U-track algorithm we modify accordingly to this particular framework where some “tracks of the past” consisting in the pores are left after the swimming of bacteria within the biofilm. We track the trajectory of swimmers in one channel by the improved U-Track. While, in the other biofilm channel we detect pores visited by swimmers and then create the other kind of trajectory for the swimmer. Finally we combine and merge these two kinds of trajectories to get more accurate trajectories. The experiment on both simulated data and real sequence demonstrates the effectiveness of the method.