This paper proposes a lossless data embedding scheme that exploits the difference expansion of the pixels to conceal large amount of message data in a digital image. The proposed scheme takes into consideration the correlation between the pixel and its surrounding pixels to determine the degree of the difference expansion for message data embedding. The performance has been evaluated in terms of image distortion, payload capacity, as well as embedding rate. The experimental results show that the scheme is capable of providing a great payload capacity, and the image quality of the embedded image is better than that of Tian?s and Celik?s schemes for a gray-level image. What is more, for a color image, the proposed scheme outperforms Alattar?s scheme at low PSNR. In addition, the proposed scheme can completely restore the original image after data extraction.
