Color Features Extraction and Classification of Digital Images of Erythrocytes Infected by Plasmodium berghei
The development of antimalarial drugs requires performing laboratory experiments that include the analysis of blood smears infected with Plasmodium berghei. Analyzing visually the resulting microscopy images is usually a slow and tedious task prone to errors due to fatigue and subjectivity of the analysts. These facts motivated the creation of digital image processing systems to automate the aforementioned analysis. We present in this work a computer vision solution which processes microscopy images of blood smears. This system performs tasks like illumination correction, color compensation, image segmentation including separation of clumped objects and the extraction and selection of color features. Then a set of classifiers was tested to find the best one in terms of classification results. Here a new feature named pixels fraction was introduced and a number of other color features were extracted, from which a subset was selected for the classification of the cells into either normal or infected. The classifiers tested for this application were: support vector machines (SVM), K-nearest neighbors (KNN), J48, Random Forest (RF), Naïve Bayes and linear discriminant analysis (LDA). All of them were evaluated in terms of their performance expressed as correct classification rate, sensitivity, specificity, F-measure and area under Receiver Operating Characteristic (ROC) curve (AUC). The usefulness of the pixels fraction as a new and effective feature was demonstrated by the experimental results. In regard of classifiers, J48 and Random Forest showed the best results.
KeywordsMalaria Image processing Computer vision Feature extraction Classifiers
The authors acknowledge the VLIR-UOS Project Cuba ICT Network for the financial support provided to this work.
- 1.World Health Organization. World Malaria Report (2017)Google Scholar
- 3.Abdul-Nasir, A.S., Mashor, M.Y., Mohamed, Z.: Colour image segmentation approach for detection of malaria parasites using various colour models and k-means clustering. WSEAS Trans. Biol. Biomed. 10(1), 41–55 (2013)Google Scholar
- 9.Chavan, S., Nagmode, M.: Malaria disease identification and analysis using image processing. Int. J. Latest Trends Eng. Technol. 3(3), 218–223 (2014)Google Scholar
- 11.Gonzalez, R.C., Woods, R.E.: Digital Image Processing, 3rd edn. Pearson Prentice Hall, Upper Saddle River (2008)Google Scholar
- 14.Bouckaert, R., Frank, E., Hall, M., et al.: WEKA Manual for Version 3-6-13. CreateSpace Independent Publishing Platform (2015)Google Scholar