Robust and secure watermarking method through BEMD, SVD and Arnold transform in wavelet domain

Abstract

In this paper, an enhanced watermarking technique through discrete wavelet transform (DWT), singular value decomposition (SVD) and Bi- dimensional empirical mode decomposition (BEMD) is presented. During the embedding process, the first level of DWT has applied to decompose an original image into distinct sub-bands (LL, LH, HH, HL). Further, BEMD decomposition is applied to divide sub-bands of DWT according to the robust and least fragile bands. SVD is employed to compute the singular coefficient on a certain band for an embedding process. The proposed technique is implemented with haar wavelet. Further, an inverse process of IDWT, ISVD and IBEMD is applied to this method to get the watermarked image. Moreover, watermark images can be extracted by the extraction process. BEMD is used to get the multiple range representation in the form IMFs. Also, it is used to improve the visual quality of an image. Moreover, the fusion of SVD, DWT and arnold transform is employed to increase the security, robustness and imperceptibility of an image against various attack. Finally, the subjective measure is used to assess the visual quality of the watermarked image. The proposed method provides better imperceptibility, security and robustness against numerous geometrical and non- geometrical attacks such as salt & pepper, Gaussian attack, JPEG compression, median filter, shearing.