The review delves into steganography and steganalysis, emphasizing the significance of detecting hidden information in digital media. It discusses the use of deep learning techniques, such as CNN, LSTM, DNN, RNN, DBN, and GNN, across various data types like images, audio, video, and text. The paper provides insights into recent advancements in steganalysis systems using deep learning methods and addresses challenges and future research directions in this field.
Over the years, technology advancements have led to increased multimedia usage for data transfer over insecure networks. Data encryption is a primary solution to mitigate security risks associated with data transfer. Steganography has emerged as an alternative method to securely transfer data without raising suspicion.
Steganalysis plays a crucial role in detecting secret messages embedded in digital media using steganography techniques. The review covers various DL-based steganalysis algorithms used for different types of carriers like images, audio, video, and text. It also discusses the impact of adversarial attacks on DL-based steganalysis models and explores areas for further research.
The paper evaluates current DL techniques for steganalysis across diverse data types and assesses their effectiveness in detecting hidden messages. It examines metrics commonly used to evaluate steganalytic methods and presents a taxonomy of DL algorithms used in steganalysis based on different carrier types.
Egy másik nyelvre
a forrásanyagból
arxiv.org
Mélyebb kérdések