Multiple cells tracking by a novel raindrop ripple optimization algorithm

Abstract

Dynamics analysis of cells' movement is essential in medical research. Multiple cells tracking through fluorescence microscopy imaging is often challenged by many roadblocks including the severe image noise, adhesion among cells, and low resolution of image. To gain full dynamics of multiple cells, a novel raindrop and ripple optimization (RRO) algorithm is proposed for multiple cells tracking in low contrast image sequences in this paper. In the proposed RRO, concepts from the natural raindrop and ripple spreading phenomena, such as raindrop fall, ripple propagation and ripple decaying are employed metaphorically to design effective searching mechanisms for solving tracking problems. Simulation results verify the effectiveness of the methods.