Gradient Adaptive Planar Prediction for High-Fidelity Images

A good pixel prediction strategy is essential to have a good Prediction Error- Expansion (PEE) based reversible data hiding (RDH) technique. Several pixel prediction strategies exist in literature. Gradients are used to predict the current pixel. Several Researchers have been proposed based on the gradient estimation to predict the current pixel for embedding more data in the current pixel. In this paper Threshold controlled gradient adaptive planar prediction with gradients of the image using context on 3 x 3 neighbourhood has been proposed for better predicting the current pixel. Adaptive histogram bin shifting has been used to insert more data with less distortion, depending on the local complexity of the pixel. The experimental study shows that the proposed technique performs better than some existing methods.