Classification of Mycobacterium Tuberculosis Based on Color Feature Extraction Using Adaptive Boosting Method

Mycobacterium Tuberculosis is acid -resistant bacteria found in the sputum. This bacterium has a special color like red to purple. Color is a specialist of clinical pathology may know that the bacteria Tuberculosis (TB) in the sputum and calculate the amount of TB bacteria. In this study, we used the Adaptive Boosting (Adahoost) method to identify TB bacteria. Before identification, filtering is carried out using the median filter and extraction of color features using HISV (Hue Saturation Value) and Adahoost with the decision tree classifier for identification. The target of this study was to determine the effect of color features in identifying TB bacteria. The results of this study indicate that the identification of TB bacteria using the extraction of I-NV color features on the Hue value can affect the accuracy value. In this study, we obtained the best accuracy value of the TB bacterial classification in testing process by using Adaboost method that was 81.7%when the hue in the color histogram was 64.