Image Processing Techniques to Identify Red Blood Cells

This paper presents a method for the automatic identification and classification of red cells in different classes of interest for diagnosis using microscopic images of blood smear. The whole system uses different image processing techniques such as binarization, contrast enhancement, noise elimination, morphological operations (dilatation, erosion), labeling and extraction of some features of interest (area, perimeter, diameter). Using this information, some factors (form factor, circularity factor, and deviation factor) involved in the classification of red cells are calculated. The classification process has two phases: the first separates red cells in normal and abnormal type and the second classifies the abnormal in three subclasses. This system does not aim to replace the pathologist, but to assist him / her and to improve the execution time of these types of analyzes.