Accurate Gradient Selective Based Histogram Bin Shifting for Reversible Data Hiding

A good prediction Error Expansion (PEE) based reversible data hiding method demands an effective pixel prediction strategy. There are numerous pixel prediction techniques in the literature. The current pixel can be predicted using gradients. In order to integrate more data in the current pixel, several researchers have suggested using gradient estimation to predict the current pixel. In this paper First, a 3 × 4 neighbourhood context-based Accurate Gradient Selective Prediction with gradients of the image has been proposed. Later, to more accurately anticipate the current pixel, the Causal pixel method has been utilised in place of linear predictors. In accordance with the pixel’s local complexity, adaptive histogram bin shifting has been employed to insert more data with less distortion. TThe experimental study demonstrates that the suggested strategy performs better than a number of other published methods.