Survey in Image and Audio Steganography by using the Deep Learning Methods

Abstract

Steganography is the practise of secretly encoding information into another medium (called a cover media) such that its existence cannot be identified. Steganography is the method that is utilised to successfully do this task. Through the use of digital image steganography, one is able to achieve the capacity of secure communication, which is vital in the bulk of the applications that are now being created. The use of steganography is applicable in a wide variety of advantageous settings. Its ascent to the forefront of current security systems may be attributed to a confluence of factors, including a phenomenal rise in processing power and a heightened awareness among individuals of the need of safeguarding their personal information. When it comes to the creation of a steganographic method, the most significant challenge that must be surmounted is finding the optimal balance between quality and quantity. The method is distinguished from linked systems like encryption and watermarking by its ability to embed information, as well as by the fact that it cannot be seen by the human eye. This article presents a thorough examination and analysis of some contemporary steganographic methods that are regarded as state-of-the-art approaches. In addition, we have included an in-depth discussion of the most important steganographic technologies now available. Recent steganographic techniques that are based on deep learning have been enhanced, which has enabled the resolution of issues that were previously present

Keywords

• Image Steganography
• Deep Learning
• GAN
• Audio Steganography