M-SVD based Quality Measurement in Hybrid non-blind Watermarked Medical Images

Abstract

Image and video quality measurement is an important subject to find out distortion between multimedia elements such as images, videos, and animations. There are several methods implemented in image quality such as Quality Index (Q), Peak Signal Noise Ratio (PSNR), Mean Square Error (MSE) and Structural Similarity Index (SSIM). In this work, we proposed a hybrid image quality degree which is measure of singular value decomposition (M-SVD), it can express image and video quality for both grey and color images numerically and graphically in medical images. M-SVD performs better results in watermarked medical images even after common attacks. In this work, we embedded binary logo as a watermark into the medical images using Discrete Wavelet Transformation (DWT). To measure the quality of the watermarked images, we applied several image processing geometric and statistical attacks. Results show that MSVD based quality measurement gives very promising result in both grey scale and color images after common attacks.