Robust plant cell tracking using local spatio-temporal context

Abstract

Automated segmentation and tracking of cells in actively developing plant tissues can provide high-throughput and quantitative spatio-temporal measurements of a range of cell behaviors. In this paper, we propose an automated segmentation and tracking method for the shoot apical meristem cells based on the cells׳ spatio-temporal contextual information. The cells are properly segmented and then tracked by using the proposed Triangle Neighborhood Structure matching method, which exploits the cells’ spatial context and turns out to be more robust than the other local graph matching methods. The tracking output acts as an indicator of the quality of segmentation and, in turn, the over-segmentation and under-segmentation errors are corrected by a local adjustment method, which is proved to be much more accurate and efficient than the global correction method. Furthermore, the cells׳ lineage tracklets are associated by using the cells׳ spatio–temporal contextual information to obtain long-term lineages. Our results on two datasets validate the effectiveness of the proposed method and we are able to track 99% of the plant cells across a long-term time period.