Tuberculosis is one of the major infectious diseases in many areas of the world. In order to prevent the tuberculosis, it must be detected. The most commonly used method for screening this disease is the Chest Radiography and the success of this method depends on the experience and interpretation of the skilled radiologist. In this paper, Computer Aided Diagnosis (CAD) system for detection of Tuberculosis using Legendre Moment Invariants is presented. The CAD method accelerates the process of active case finding as well as provides second medical opinion for radiologist. The derivation of Legendre moment invariants are presented in this paper. Legendre moments are based upon orthogonal Legendre polynomials and provide better representation capability of images and hence selected as features in this paper. Simulation results are carried out to solve three problems namely image reconstruction, invariance property and diagnosis of tuberculosis by considering standard Montgomery County X ray dataset and the results are compared with combined blur and affine moment Invariants. From the results, it is observed that the proposed method provided better reconstruction, invariance and accuracy.
Volume 11 | 04-Special Issue
Pages: 2219-2226