Tracking interacting subcellular structures by sequential Monte Carlo method

With the wide application of green fluorescent protein (GFP) in the study of live cell, which leads to a better understanding of biochemical events at subcellular level, there is a surging need for the computer-aided analysis on the huge amount of image sequence data acquired by the advanced microscopy devices. One of such tasks is the motility analysis of the multiple subcellular structures. In this paper, an algorithm using sequential Monte Carlo (SMC) method for multiple interacting object tracking is proposed. We use joint state to represent all the objects together, and model the interaction between objects in the 2D plane by augmenting an extra dimension and evaluating their overlapping relationship in the 3D space. Markov chain Monte Carlo (MCMC) method with a novel height swap move is applied to sample the joint state distribution efficiently. To facilitate distinguishing between different objects, a new observation method is also proposed by matching the size and intensity profile of the object. The experimental results show that our method is promising.