Segmentation on Chest Radiographs Using Otsu's and K-Means Clustering Methods

摘要

Lung infectiousdiseases which are commonly Tuberculosis TB, pneumonia, pleural effusion etc are generally affects the lungs. The primary identification of these infectious diseases is by Chest radiography (CXR). We usually use segmentation technique in identifying. Segmentation means subdividing an image into several segments to identify constituent parts in the image. There are many algorithms on segmentation of chest radiographs [1]-[3] etc. had been proposed till now. In this paper we will analyze segmentation on chest radiographs using Otsu's Threshold Method and K-means Clustering Method in terms of results and performance.