Cell segmentation with median filter and mathematical morphology operation

This paper describes a part of current research work on an automated cell tracking (ACT) project with the aim of tracking the movement of representative cells in order to determine the activity of the cell once certain medical substances have been injected. A major requirement for this project is an efficient method to segment cell images. This work shows that the combination methods of the median filter and the mathematical morphology operation such as dilation and erosion are adaptable and can be optimized to improve the segmentation result. The experimental results show that this method is suitable for segmentation based on edge detection of melanomas (tumor cells) and lymphocytes (blood cells) and the performance of this method is comparable to several edge detection methods such as Sobel, Prewitt, Roberts and Laplacian of Gaussian