Active contours for the movement and motility analysis of biological objects

We present a complete framework for automating the objective and reliable assessment of biological objects motility. We address, in the context of one particular application, some of the main problems in dealing with the automatic tracking of biological objects, which are: (1) the robust segmentation of individual objects which is quite often hindered by the poor contrast of the imaging technique; (2) the proper initialisation of multiple objects description; and (3) the problem of temporary object aggregation, during which objects loose their individuality. We have implemented the active contour model based on the gradient vector field model (Xu and Prince, 1998) and adapted it in order to overcome some of its limitations in dealing with biological related objects. Results are presented of the analysis of video sequences of a living amoeba.