Multiassignment for tracking a large number of overlapping objects

In this paper we present a new technique for data association using multias signment for tracking a large number of closely spaced (and overlapping) ob jects. The algorithm is illustrated on a biomedical problem, namely the tra cking of a group of fibroblast (tissue) cells from an image sequence, which motivated this work. Because of their proximity to one another and due to the difficulties in segmenting the images accurately from a poor-quality Im age sequence, the cells are effectively closely spaced objects (CSOs). The algorithm presents a novel dichotomous, iterated approach to multiassignmen t using successive one-to-one assignments of decreasing size with modified costs. The cost functions, which are adjusted depending on the "depth" of t he current assignment level and on the tracking results, are derived. The r esulting assignments are used to form, maintain and terminate tracks with a modified version of the probabilistic data association (PDA) filter, which can handle the contention for a single measurement among multiple tracks i n addition to the association of multiple measurements to a single track. E stimation results are given and compared with those of the standard 2-dimen sional one-to-one assignment algorithm. It is shown that iterated multiassi gnment results in superior measurement-to-track association. The algorithms presented here can be used for other general tracking problems, including dense air traffic surveillance and control.