Automated malaria parasite detection in thin blood films:- A hybrid illumination and color constancy insensitive, morphological approach

摘要

This paper illustrates the automated diagnosis of malaria parasite (Plasmodium species) in microscopic images of Giemsa stained thin blood films. The procedure adapts a morphological approach for blood cell identification and uses the image features such as intensity, histogram, relative size and geometry for further analysis. Two methods of object classification have been described for parasite detection; one based on relative size and morphology and the other based on intensity variation. Furthermore, an analytical study on both methods has been performed in order to validate the accuracy of the methods.