Malaria Infected Erythrocyte Classification Based on the Histogram Features using Microscopic Images of Thin Blood Smear

Objectives: This paper aims to develop a system for malaria infected erythrocyte classification based on the histogram feature set. Method: The method consist of pre-processing, segmentation, feature extraction based on the histogram of different color channels, feature selection and malaria infected erythrocyte classification using Artificial Neural Networks (ANN), Support Vector Machine (SVM), k-Nearest Neighbor (k-NN) and Naive Bayes. Findings: The experimental analysis of all the classifiers with the different combinations of features has been carried out on clinical database. Based on the experimental results it may be concluded that ANN provides the higher classification rate in comparison to other classifiers which provides an overall accuracy of 96.32% and F-score of 85.32% respectively. Applications: The proposed system may be used for the automatic recognition of the malaria infected erythrocytes in the thin blood smears.