Detection of tuberculosis bacteria in sputum slide image using morphological features

Abstract

Automatic finding of tuberculosis bacteria is a medicinal imaging issue that includes the utilization of machine vision strategies. The manual technique for the detection of tuberculosis bacteria is costly, required much time and qualified persons to prevent errors. In this paper a new detection technique is presented. This technique is based on morphological features of bacteria object. The identification process is carried out using eccentricity, bounding box, area and aspect ratio abstract environment.