A new model for tracking heterogeneous particle movement in dense environments
A major challenge in the tracking of multiple particle is the capture of heterogeneous movements in crowded scenes. Current models for motion prediction, adapted from aerospatial, are challenged by rapid and unpredictable changes in the nature of the dynamics of objects in motion in a dense intracellular environment. In a study recently published in IEEE Transactions on Image Processing, Roudot et al. propose a piecewise-stationary motion model tailored for intracellular dynamics and a new estimator that exploits multiple tracking rounds within an iterative multiple model smoother (PMMS). This method identifies meaningful dynamical switches in trajectories but also confers an increased robustness toward speed variation and lower acquisition speeds. These results indicate that PMMS is a potent solution to the problem of heterogeneous motion tracking in crowded particle fields which has been a notorious obstacle to high-fidelity analysis of object dynamics, especially in bioimaging applications.
The IEEE Transactions on Image Processing is a leading journal for imaging processing and computer vision. It covers novel theory, algorithms, and architectures for the formation, capture, processing, communication, analysis, and display of images, video, and multidimensional signals in a wide variety of applications.