Abstract:
As we have a look on technology we can see that technology is becoming advanced day by day. But by having view on the health department we can see that the disease or also becoming very dangerous. Now a days a bacteria called mycobacterium is very threat full in the sense of human healthy life. This bacteria is the cause of tuberculosis which is infectious disease and is spreading gradually in world wide. According to research for the diagnosis of tuberculosis radiologist and physicians mostly recommend chest x-ray as compared to other technologies. Because it is easily available everywhere. But the diagnosis using x-ray is difficult because it’s a very noisy image and the quality of image is very poor. If the diagnosis of tuberculosis is not kept on time it will become a cause of death. While the so there should be an automatic system for tuberculosis detection. According to research automatic tuberculosis is not as accurate as human diagnosis. The main aim of this thesis is to improve the quality of automatic tuberculosis detection system using chest x-rays. In this thesis automatic tuberculosis detection is categorized into two categories 1st is segmentation and the 2nd one is classification. By segmenting the image it’s easy to detect the abnormality in the lung through classification. This process is done with image processing techniques.
