The cell image segmentation and classification based on OTSU method and connected region labeling

Automatic analysis of medical cell images has been gaining more importance in pharmacology research. Segmentation and classification of cell images are the crucial steps in their analysis since results of segmentation and classification directly affect the following quantitative analysis steps. In this paper, we present a cell image segmentation and classification method based on OTSU method and connected region labeling. By K-L transform we choose one channel and ensure an automatic threshold value by OTSU method to make segmentation, then we adopt the method of connected region labeling designed by ourselves to make classification. The method has been applied to the segmentation and classification of several kinds of cell images. The experimental results show that the method is practical and effective; it has good robustness and ensures the accuracy.