Analysis of Image Watermarking Techniques against Various Noise Attacks for Medical Images using Genetic Algorithm (GA)

Reddy V

Abstract

In this paper, we have presented a dual secure digital image watermarking algorithm based on discrete wavelet transforms (DWT), lifting wavelet transforms (LWT) and singular value decomposition (SVD) using genetic algorithm (GA) for medical images. The natural image is chosen as cover image and is decomposed up to three levels using discrete wavelet transform. The medical image is chosen as a watermark. The genetic algorithm is used to embed and extract the watermark. The encryption is proposed to be effected using RSA and AES encryption algorithms. A graphical user interface (GUI) which enables the user to have an ease of operation in loading the image, watermark it, encrypt it and also retrieve the original image whenever necessary is also designed and presented in this paper. Peak signal to noise ratio (PSNR) and normalized correlation coefficient (NC) are computed to measure the image quality of the proposed technique. Experimental results show that the proposed watermarking technique has good imperceptibility and robustness against to image processing attacks such as sharpening, smoothening, motion blur, salt and pepper noise, gaussian noise, speckle noise and poisson noise.Cite this ArticleReddy and Siddaiah. Analysis of Image Watermarking Techniques againstVarious Noise Attacks for Medical Images using Genetic Algorithm(GA).Journal of Artificial Intelligence Research and Advances. 2015; 2(3):36–44p.

Abstract

Keywords