Single-pixel based double random-phase encoding technique

Abstract

A new encryption technique based on single-pixel compressive sensing along with a Double Random-Phase encoding (DRPE) is proposed. As compared with the conventional way of image compression where the image information is firstly capture and then compress, the single-pixel compressive sensing collects only a few large coefficients of the data information and throws out the remaining which gives scrambled effect on the image. Further, to enhance the complexity of the image data, the double random phase encoding along with a fractional Fourier transform is implemented to re-encrypt it. The single-pixel based compressive sensing, DRPE and fractional Fourier transform act as a secret keys. At the receiver end, the original image data is reconstructed by applying the inverse of double random phase process and an l1-minimization approach. The peak-to-peak signal-to-noise ratio and the minimum number of compressive sensing measurements to reconstruct the image are used to analyze the quality of the decryption image. The numerical results demonstrate the system to be highly complex, robust and secure.