Combination schemes Reversible Data Hiding for medical images

The conventional schemes of data hiding such as Least Significant Bit (LSB) or frequency domain using the Discrete Cosine Transform (DCT), Discrete Fourier Transform (DFT) or Discrete Wavelet Transform (DWT) on digital images can cause a permanent distortion after the extraction process. In medical images, distortion in the image can cause misdiagnosis. To overcome this problem, a special scheme is required for the embedding data in the medical image such as reversible data hiding. One of the schemes in reversible data hiding is Difference Expansion (DE), which is able to restore the image without any distortion after the extraction process. In this paper, a new scheme based on DE has been proposed to enhance capacity and visual quality. One of the characteristics of medical images is the large smooth block areas, while the original DE scheme is applied to the non-smooth areas image. Eight different medical images were used to evaluate the proposed scheme. The result showed that the proposed scheme has a high capacity and very good visual quality compared to the original scheme and similar schemes that have ever been previously proposed.