Contemporary Approaches On Reversible Data Hiding In Iris Image Using Deep Learning Techniques

The concept of reversible data concealing is crucial for secret communication. After the extraction stages, the protected information can be completely rebuilt in addition to the hidden data being removed. Pixel, Block and Interpolation based algorithms have seen substantial development over the past decade, employing the spatial-frequency domain and other related techniques. The goal of this research work is to categorize and describe the various methods, inner groups and research papers related to the idea of Reversible Data Hiding (RDH). In order to protect privacy, this research suggests a revolutionary iris image data concealing strategy. Here, private information is incorporated into the iris image in a way that has the least negative effect on iris recognition. The Syndrome Trellis Coding (STC) architecture is applied, limiting changes to the embedded data and to the regions that infrequently affect iris identification. A new distortion function is also proposed to enumerate the impression of data implanting on recognition of Iris, which assigns high embedding cost to locations with strong iris features. According to experimental findings, enough information can be incorporated into the iris image while sustaining excellent identification accuracy by employing the proposed technique.