Image processing for combined bright-field and reflection interference contrast video microscopy

Abstract

Image processing algorithms for automatic extraction of cell body contours and cell-substratum contacts from video images, which were obtained by bright-field microscopy and reflection interference contrast microscopy, respectively, are described. Double-view imaging, which combines these two optical techniques, is used to investigate the relationship between cell-to-substratum adhesion and cell shape changes during locomotion of the amoeboid cells of Dictyostelium discoideum. Contact areas of cells are extracted from reflection interference contrast images via a routine which performs binarisation on the basis of threshold estimation, as it is calculated by a histogram minimum method. Boundaries of cells are extracted from bright-field images by utilising an algorithm that includes background subtraction and binarisation based on texture discrimination by means of a rank operator.