An optical multiple-image authentication based on computational ghost imaging and total-variation minimization

Area of Science:

• Optics and Photonics
• Image Processing
• Information Security

Background:

• Conventional image authentication often involves direct encryption, making it susceptible to detection.
• Embedding information discreetly within a cover image offers a more covert approach to authentication.

Purpose of the Study:

• To develop a secure and robust optical multiple-image authentication system.
• To embed multiple images into a cover image without drawing attention from potential eavesdroppers.

Main Methods:

• Utilizing computational ghost imaging to encode multiple images into real-valued sequences.
• Employing wavelet transform to decompose a cover image into sub-images for data embedding.
• Implementing total-variation minimization for high-fidelity image recovery.
• Enhancing security through chaotic sequence scrambling with a logistic map.

Main Results:

• Successful embedding and recovery of multiple images with high quality.
• Demonstrated robustness against potential eavesdropping attempts.
• Validation of the proposed method through optical experiments.

Conclusions:

• The proposed computational ghost imaging and total-variation minimization approach offers a secure and effective method for optical multiple-image authentication.
• This technique provides a robust solution for covertly embedding and recovering image data, outperforming conventional methods.