A multi-seed dynamic local graph matching model for tracking of densely packed cells across unregistered microscopy image sequences

The tracking of plant cells in large-scale microscopy image sequences is very challenging, because plant cells are densely packed in a specific honeycomb structure, and the microscopy images can be randomly translated, rotated and scaled in the imaging process. This paper proposes a multi-seed dynamic local graph matching method to track the plant cells across unregistered microscopy image sequences, by exploiting the geometric structure and topology of cells' relative positions as contextual information. The proposed dynamic cell matching scheme always selects the most similar cell pair in the dynamically growing neighbor set of matched cells, so it tends to prevent the matching error accumulation during the cell correspondence growing process. Furthermore, the multi-seed-based majority voting scheme can automatically rectify the matching errors produced by one seed only. Last, the cells' lineage tracklets are associated by using the cells' spatial-temporal context to obtain long-term trajectories. Compared with the existing local graph matching method, the experimental results show that the proposed method improves the tracking accuracy rate by about 30% in the unregistered image sequences.