Performance Comparison of Discrete Orthonormal S-Transform for the Reconstruction of Medical Images

Abstract

This paper investigates the performance of the Discrete Orthonormal S-Transform (DOST) technique for the reconstruction of medical images. This technique is compared with the q-Recursive Zernike Moment (q-RZM) and Discrete Wavelet Transform (DWT) based on the performance measurement, namely Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) on medical images. The ultimate goal is to find the best technique that can extract the important feature of the medical image. Computer Tomography (CT) and Magnetic Resonance (MR) images are used to test the viability of the techniques to be used for image compression. The experiment is done using MATLAB tool. From the results, the DOST technique has high PSNR value and outperforms q-recursive Zernike Moments and Discrete Wavelet Transform Techniques.