Asymmetric double image encryption, compression and watermarking scheme based on orthogonal-triangular decomposition with column pivoting

Abstract

A novel asymmetric scheme for double image encryption, compression and watermarking based on QR decomposition in the Fresnel domain has been presented. The QR decomposition provides a permutation matrix as a ciphertext and the product of orthogonal and triangular matrix as a key. The ciphertext obtained through this process is a sparse matrix that is compressed by the CSR method to give compressed encrypted data, which, when combined with a host image, gives a watermarked image. Thus, a cryptosystem that involves compression and watermarking is proposed. The proposed scheme is validated for grayscale images. To check the efficacy of the proposed scheme, histograms, statistical parameters, and key sensitivity are analyzed. The scheme is also tested against various attacks. Numerical simulations are performed to validate the security of the scheme.

Optical multi-image encryption based on chaotic fingerprint phase mask and multi-slice diffractive imaging

“…In addition to the single-image encryption schemes, researchers have also proposed dual-image encryption schemes [32][33][34][35][36][37] and multi-image encryption schemes [38][39][40][41][42] to improve encryption efficiency. Jiang et al [34] proposed an asymmetric double color image cryptosystem based on compression sensing (CS), chaos technology, and DRPE.…”

Asymmetric cryptosystem with random decomposition using fractional Fourier and Fourier transforms

“…In asymmetric scheme with the knowledge of only encryption key, it is challenging to retrieve the image. Therefore, other asymmetric schemes [11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26] were also proposed. The issue of optical axis alignment affects the traditional DRPE method.…”