Microscopic Blood Smear RBC Classification using PCA and SVM based Machine Learning

The therapeutic analysis f microscopic blood smear begins with recognizing blood cells f various categories as well as estimating cell count in blood sample. The distinctive blood cell grading and counting furnish priceless knowledge to the pathologist about sundry infections. This exercise can be easily concluded if the shapes f blood cells are pinpointed first and using the shapes, we classify the blood cells. In this research work we build and test an automatic microscopic blood smear red blood cell (RBC) classification by using Principal Component Analysis (PCA) and Support Vector Machine (SVM) based machine learning. We train and test the statistical data models based n probabilistic pattern recognition to classify the blood smear RBC into Normal Cells, Echinocytes, Elliptocytes and Sickle cells. The H-minimum Transform (HmT) and Watershed Transform (WT) are used in pre-processing f images to increase the accuracy if segmentation shape extraction f the blood cells. Then the Bag f Features (BoF) created considering the 500 strongest features f each type f blood cell after K-Means clustering. Training takes place through Image Category Classifier (ICC) whose performance measured by using Mean Average Precision (mAP) justifies that SVM based classifiers provide audacious results.