A cryptosystem based on deterministic phase masks and fractional Fourier transform deploying singular value decomposition

• Published in: Optical and quantum electronics Vol. 50; no. 5; pp. 1 - 24
• Main Authors: Girija, R., Singh, Hukum
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.05.2018; Springer Nature B.V
• Subjects: Characterization and Evaluation of Materials; Computer Communication Networks; Correlation analysis; Correlation coefficients; Decomposition; Electrical Engineering; Encryption; Fourier transforms; Lasers; Masks; Occlusion; Optical Devices; Optics; Photonics; Physics; Physics and Astronomy; Reliability analysis; Reliability aspects; Security; Singular value decomposition; SVD; PSNR; DRPE; MSE; Deterministic phase masks; Fractional Fourier transform
• DOI: 10.1007/s11082-018-1472-6
• ISSN: 0306-8919

In this paper, an asymmetric cryptosystem has been proposed to enhance the security of DRPE. The traditional DRPE scheme is thus tweaked by using fractional Fourier transform (FrFT), a class of structured phase masks called as deterministic phase masks (DMKs) and deploying singular value decomposition (SVD). In specific, we propose to organise the encryption procedure by using two DMKs and FrFT, additionally deploying SVD. On the decryption front, the input image is recovered by utilising the inverse singular value decomposition (ISVD) and an angular portion of the deterministic phase masks. The use of FrFT for encryption and decryption would enhance the robustness of DRPE scheme. Deployment of SVD on our asymmetric cryptosystem provides three components for cipher image is yet another added feature that hardens the security of DRPE scheme. DMKs are formed by the deviation from conventional rectangular function and limited range values which delivers key components with reduced size, better performance and lower complexity. The capability study of defined method, includes analysis on SVD, histogram and correlation coefficient. Our system is subject to an occlusion attack and noise attack to evaluate its performance and reliability. Computational analysis outputs and security investigation are offered in aspect to determine the security potential of proposed system. Comparative results are shown for values of mean-square-error and peak-signal-to-noise ratio of DRPE schemes.