Separation and counting of blood cells using geometrical features and distance transformed watershed

In the field of biomedicine, blood cells are complex in nature; the microscopic images of a blood stream contains RBCs, WBCs and Platelets. Pathological inspection of an infected cell based on the disease, is solely dependent on subjective assessment which usually leads to significant inter-observer variation in grading and subsequently resulting in late diagnosis. However automatic assessment of required cell count still remains a challenging task as many of the cells are clumped in an image and proper segmentation is the primary aspect. This paper aims at segmentation of blood cells for counting. Auto threshold, Chessboard distance measure and watersheding are used for segmentation of blood cells.