Overlapping Cells Segmentation Algorithm Based on Curvature and Active Contour Model

Key words: 
• overlapping cells segmentation /
• curvature /
• active contour model /
• Otsu algorithm

Abstract: In this paper, an algorithm of overlapping cell segmentation based on curvature and active contour model has been proposed to solve the problem that it is difficult to automatically separate microscopic image boundaries of overlapping cells.The algorithm uses the Otsu algorithm and morphological filtering to get the whole outline of the overlapping cells, and according to its curvature information to locate the border contact points of the two cells.Then an active contour model is used to extract the contour of the overlapping region.Finally, according to the information of the border contact points, the algorithm obtains the single cell's contour by connecting the whole outline of the overlapping cells to the contour of the overlapping region.The experimental results show that the proposed algorithm can extract the contour of the overlapping cells, the integrity of the segmented cell is high, and the algorithm is robust to a certain extent.The algorithm for segmentation of overlapping cells is feasible and effective.