Título: A computer vision system for tracking proliferating cells /
Autores: Wheeler, Kim M. (Kim Margaret)
Fecha: 1992
Publicador: McGill University - MCGILL
Fuente:
Tipo: Electronic Thesis or Dissertation
Tema: Biology, Cell.
Computer Science.
Descripción: A computer vision system which tracks and analyses living cells and their offspring from sequential two-dimensional images is described. The cells are segmented using a method which combines global thresholding, component labeling and noise filtering. The shapes of the cells are then analyzed using a fast skeleton algorithm, and the states of the cells are determined as the cells change shape and move. This vision system recognizes the occurrence of cell mitosis and then tracks the offspring by evaluating the overall cell shape and position to determine the cell state. Both the cell shape and cell state are represented as fuzzy sets. The cell state takes into account previous shape and state knowledge as the mitotic process is described by a specific series of shape changes. This Mitosis Recognition System (MRS) has been developed for cell physiology research in the study of cell locomotory characteristics evolving over several generations.
Idioma: en