Center location and segmentation of cell two-photon microscopic images based on random sample consensus

Complex background, critical noisy and fuzzy boundary makes available methods of cell image segmentation disappointing. Thus, a new method that can locate and detect nucleus effectively is proposed in this paper. As a precursor to accurate segmentation, shape modeling of cells is required. Our method incorporates a priori knowledge about cell shape. That is, an elliptical cell contour model is introduced to describe the boundary of the cell. Our method consists of the following components: Firstly, the initial segmented regions are obtained by using C-means clustering algorithm. Secondly, locate the center of cells accurately through the RANSAC (random sample consensus) method. Finally, a reformed level set evolution is introduced to extract the edge of nucleus. The experiment result shows that, nucleus can be located accurately; even if the cell image has a complex background and is obscured by other objects. Moreover, the edge of nucleus extracted by this method has a higher accuracy.