An Ensemble Learning Approach for Reversible Data Hiding in Encrypted Images with Fibonacci Transform

Reversible data hiding (RDH) is an active area of research in the field information security. In RDH, a secret can be embedded inside cover medium. Unlike other data-hiding schemes, RDH becomes important applications that demand recovery without any deformation, along with hidden secret. this paper, new scheme proposed for performing reversible encrypted images using Fibonacci transform ensemble learning method. scheme, hider encrypts original image and performs further hiding. During hiding, partitioned into non-overlapping blocks, each block considered one-by-one. The selected undergoes series transforms during number required on determined by integer value wants to embed. On receiver side, message extraction restoration are performed help will try perform all possible decrypt blocks. recovered identified trained machine-learning models. novelty lies (1) retaining pixel intensities unaltered while data. Almost every described literature alters embed data, which represents security concern encryption algorithm; (2) Introducing efficient means through model framework. majority votes from different models guarantee correct image. enables reduction bit error rate contributes ensuring suitability areas such as medical transmission cloud computing. results obtained experiments undertaken show was able attain improved payload capacity 0.0625 bits per pixel, outperforming many related schemes complete reversibility.