Tuberculosis diagnosis through image processing

Abstract

Tuberculosis is most common contagious disease. Nowadays, millions of human beings of the world are suffering from it. We will use the most used worldwide method Ziehl-Neelsen stain (ZN-stain) to detect Tuberculosis which is based on sputum examination microscopically. This method needs expert human resources and implicit examination. The main constraints are expertise human,time and cost to implement our process. We will use Thresholding, multi-stage, color-based Bayesian segmentation identified possible ‘Tuberculosis objects’, removed artifacts by shape comparison and color-labeled objects as ‘definite’, ‘possible’ or ‘non-Tuberculosis’, bypassing photomicrographic calibration.In our work, we will use an algorithm based on image processing is developed for identification of Tuberculosis.