A Novel Ant-Based Multiple Cells Tracking Approach with Cardinality Estimation

Cell migration is an important process in normal tissue, organ or entire organism development and disease. This paper proposes an ant algorithm based on cardinality estimation for clustered cells state and number estimator simultaneously. Cardinality prediction and updating model based on the existence probability of pheromone field are derived for effectively estimating the number of cells. In order to separate clusters cells, an ant work model based on the pheromone gradient information is developed to guide ants movement towards center of interested cells. Experiment results show that our algorithm could automatically track clustered cells in various scenarios, and, it is more accurate than other popular tracking methods.