Cell Segmentation Using Ellipse Curve Segmentation and Classification
Xiaomin Li, Yuanyuan Wang, Yinhui Deng, Jinhua Yu · 2009
Cell segmentation from microscopic images is the first stage of the automatic biomedical image processing, which plays a crucial role in the study of cell behavior and cell structure. In this paper, a novel approach is proposed to segment cells which are characterized as elliptical objects. Cell segmentation is implemented using an ellipse detection algorithm based on ellipse curve segmentation and classification. In the method, the image preprocessing is firstly applied to extract object edges from the microscopic image. Then the curvature-based curve searching is proposed to segment edges into different curves followed by the process of curve's under-segmentation and the process of curve's over-segmentation. Finally, curve classification and ellipse fitting are used to obtain object cells. Experiment results demonstrated the effectivity of the proposed method.