Cell Segmentation and Classification Using Digital Holographic Microscopy Images

Abstract

In this modern era Digital Holographic Microscopy is very much popular designed for cell imaging. The cellular imaging had developed in growing cell biology. Digital Holographic research (DHM) might be a label-free imaging technique which allows a noticeable illustration of clear cells with imaging cell culture plates. The most advantage is that, it doesn't offer the projected image of the thing however provides 3 dimensional data of the object’s optical thickness. Using DHM, machine learning approaches are used for the extraction purpose, variation and calculative blood statistics comparable to the Mean corpuscular Volume (MCV), the Red blood cell (RBC) count, Red blood cell Distribution width (RDW).Segmentation is that the method of partitioning a digital image in to multiple elements.  For cell segmentation, first the cell has to be detected. Then the detected cells are used to separate by means of layered segmentation. After the segmentation process, the blood cells are classified by means of K-NN classifier and SVM classifier. This shows the quantitative evidence that it works better than power watershed segmentation.