Counterfeiting attacks on oblivious block-wise independent invisible watermarking schemes

Area of Science:

• Digital image forensics
• Information security
• Cryptography

Background:

• Oblivious watermarking embeds data without altering the original image.
• Block-based schemes divide images into blocks for independent processing.
• Existing oblivious watermarking methods may have security vulnerabilities.

Purpose of the Study:

• To identify and analyze attacks on block-based oblivious watermarking.
• To demonstrate the feasibility of counterfeiting attacks on these schemes.
• To propose a method for mitigating identified vulnerabilities.

Main Methods:

• Analysis of block-wise independent embedding in oblivious watermarking.
• Development and implementation of a counterfeiting attack.
• Evaluation of attack success on published watermarking techniques.
• Formulation of a countermeasure against block-wise independence.

Main Results:

• Demonstrated a counterfeiting attack on block-based oblivious watermarking.
• Showcased watermark forgery without secret keys or explicit watermark knowledge.
• Successfully implemented attacks on several literature-proposed watermarking schemes.
• Identified a solution to prevent block-wise independence vulnerabilities.

Conclusions:

• Block-wise independent oblivious watermarking schemes are susceptible to counterfeiting.
• The proposed solution effectively addresses the vulnerability of block-wise independence.
• Enhanced security measures are crucial for robust digital watermarking systems.