Robust and secure image steganography with recurrent neural network and fuzzy logic integration

Area of Science:

• Computer Science
• Information Security
• Artificial Intelligence

Background:

• Digital communication necessitates robust data security and privacy measures.
• Image steganography offers a method for secure data transmission by concealing messages within images.
• Existing methods face challenges in maintaining data integrity against various attacks.

Purpose of the Study:

• To develop a novel, highly secure, and robust image watermarking technique.
• To enhance the protection and transmission rate of hidden information in digital images.
• To advance the field of secure data concealment through an intelligent framework.

Main Methods:

• Implementation of a two-layer system combining fuzzy logic and Recurrent Neural Networks (RNNs).
• Utilizing RNNs for optimizing the embedding process and managing uncertainty.
• Employing fuzzy logic for improved decision-making and handling noise, compression, and attacks.

Main Results:

• The proposed method demonstrates enhanced security and robustness against various attacks.
• Achieved high inaudibility and invisibility of steganographic data.
• Outperformed existing steganography techniques in embedding efficiency and attack resistance.

Conclusions:

• The integration of fuzzy logic and RNNs provides an efficient and scalable framework for secure image steganography.
• The developed method offers superior protection for hidden data in digital images.
• This approach represents a significant advancement in safeguarding data privacy and integrity in digital communications.