Silhouette-free image encryption using interference in the multiple-parameter fractional Fourier transform domain.

Area of Science:

• Computer Science
• Cryptography
• Image Processing

Background:

• Traditional image encryption methods can be vulnerable to silhouette attacks.
• The fractional Fourier transform (FrFT) is a powerful tool in optical information processing and cryptography.
• Generalizing FrFT to multiple parameters offers potential for enhanced security.

Purpose of the Study:

• To propose a novel silhouette-free image encryption scheme.
• To leverage the discrete multiple-parameter fractional Fourier transform (DMPFrFT) for enhanced security.
• To develop a cryptosystem resistant to attacks using partial information.

Main Methods:

• The proposed method utilizes chaotic pixel scrambling (CPS) on the original image.
• The scrambled image is encoded into the real part of a complex signal.
• Three phase-only masks are generated in the DMPFrFT domain using interference principles.

Main Results:

• The encryption scheme ensures that the original image silhouette cannot be recovered using only one or two phase-only masks.
• The parameters of both CPS and DMPFrFT serve as encryption keys, significantly expanding the key space.
• Numerical simulations confirmed the feasibility and validity of the proposed encryption approach.

Conclusions:

• The developed image encryption technique effectively prevents silhouette extraction.
• The use of DMPFrFT and CPS provides a robust and secure cryptosystem with a large key space.
• This approach offers a promising solution for secure image data transmission and storage.