Automated Vision-based Tracking of Cell Movement and Deformation in Image Sequences
Automated cell segmentation and tracking are vital for quantitative analysis of cell behaviours such as migration, proliferation, apoptosis, etc. However, the low contrast cellular image quality, ill-defined cell boundaries, complex cell mo-tions, and diversity of cell shapes and large cell deformations all pose significant problems to the efficient and robust cell tracking in phase contrast cellular images. We present an auto-mated multiple cell tracking system, which can simultaneously deal with those challenging is-sues. It is based on a model-evolution tracking method: a localizing region based parametric active contour model. We demonstrate that the proposed method can accurately segment cell boundaries and track cell, even with relatively large cell displacements. Also, we exploit a confidence matching component that combines cell shape and intensity similarity measure-ments that can automatically estimate the tracking confidence.
Sha Yu Derek Molloy
National Biophotonics and Imaging Platform Ireland (NBIPI)Vision System Group, Dublin City Universit National Biophotonics and Imaging Platform Ireland (NBIPI) Vision System Group, Dublin City Universi
国际会议
武汉
英文
58-62
2010-10-10（万方平台首次上网日期，不代表论文的发表时间）