The Robust Watermarking Algorithms For Medical Volume Data In The Encrypted Domain

In recent years, the development of the Internet and the medical information technology has facilitated the doctors'diagnosis and communication immensely, as well as solved the problems of uneven distribution of medical resources. However, it has brought more threats to the patients'private information. Unprocessed medical images are exposed to something dangerous when they are transmitted in the Internet or stored in the cloud. Therefore, it is urgent to protect the security of medical images.The usual way to protect the security of images is to encrypt the plain images. The encrypted images are in a chaotic and unintelligible state. Unauthorized users can't see the original content of the encrypted images, unless they have keys. Digital watermarking technology can be used to hide patients'information or doctors'diagnosis in the encrypted images, which can ensure information communication as well as the security of patients' information.At present, the purpose of the most of the information hiding schemes in the encrypted domain is to improve the embedding capacity and the PSNR value of the recovered image. There is short of research on robust watermarking in the encrypted domain, especially for three-dimensional images. While there existing a large number of 3D medical images, like CT-. MRI images. Therefore, it is significant to do research on robust watermarking schemes in encrypted domain for 3D medical images.This paper proposed a robust watermarking algorithm for encrypted medical volume data based on DCT. This algorithm uses Logistic map to encrypt the medical volume data in the DCT domain. Then the third party extracts the feature vector of the encrypted image to combine with the watermark using zero-watermarking method. The watermark embedding doesn't change the image quality. The experimental results show that the proposed algorithm is robust to some normal and geometric attacks.This paper proposed a robust watermarking algorithm in encrypted domain for medical volume data based on DFT. This algorithm uses Chen chaotic system to achieve more secure encryption effects as it has larger key space and faster diffusion rate, besides it can overcome the drawbacks of Logistic map. After encryption, the feature vector of the encrypted medical volume data is extracted in the DFT domain and then calculated with the watermark by exclusive OR. This algorithm can achieve blind watermark extraction. The experimental results show that the proposed algorithm has good performance against some normal and geometric attacks.This paper proposed a robust watermarking algorithm in encrypted domain for medical volume data based on DWT. This algorithm utilizes the multi-resolution property of DWT to extract the feature vector of the encrypted medical volume data in the DWT-DCT domain. This algorithm doesn't limit the number of the embedded watermarks. The experimental results show that the proposed algorithm has good practicality and good robustness against some normal and geometric attacks.