Steganographic coding scheme based on dither convolutional trellis under resampling mechanism.

Area of Science:

• Digital steganography
• Information security
• Multimedia security

Background:

• Digital media files are common carriers for steganography in mobile applications.
• Media servers may resample files, potentially corrupting embedded secret messages.
• Existing steganographic methods struggle with resampling transformations.

Purpose of the Study:

• To develop a steganographic coding scheme resilient to media resampling.
• To reduce embedding distortion while maintaining secret message extraction accuracy.
• To address both one-dimensional and two-dimensional resampling scenarios.

Main Methods:

• A steganographic coding scheme based on dither convolutional trellis is proposed.
• Resampling mapping is estimated using finite sample pairs.
• Syndrome-Trellis Codes (STCs) are used to generate stego media files.
• A dither convolutional trellis is constructed for one-dimensional resampling.
• The scheme is extended to two-dimensional resampling, like image scaling.

Main Results:

• The proposed scheme achieves reduced embedding distortion.
• Accurate extraction of secret messages is ensured under resampling.
• The method demonstrates effectiveness for both 1D and 2D resampling.
• Experimental results validate the scheme's performance.

Conclusions:

• The dither convolutional trellis-based steganographic scheme effectively handles media resampling.
• It offers a robust solution for embedding secret data in digital media subject to transformations.
• This approach enhances the security and reliability of steganography in mobile environments.