Zero Watermarking For Medical Image

The digital representation of medical data has lots of advantages in data storage and transmission. But these advances have introduced new problems of medical information security. Compared with the traditional image, it will have a high risk of medical image which is used as diagnostic evidence circulating in open networks. If the information of medical images is poached, it cannot be used as the evidence in the medical accident, even it will harm individual’s privacy. Currently, encryption and access control technologies are difficult to meet the requirements of the medical image’s information security. It is an urgent need of security measures in medical information system. Using watermark’s properties of invisibility and robustness, a landmark information can be put as a digital watermarking hidden in medical images. Even though the medical images have been attacked in the process of information exchange, it can still completely extract the watermarking which can protect the patient’s information of medical image. It can be effectively solve the problem of tampering, poaching in transmission. Because of medical image and the corresponding medical report were stored separately in present, the watermarking technology can be an effective way to settle the issues of the information disordering in the management of medical record.In this dissertation, medical images are taken as my research object. By utilizing the theory of image processing and watermarking technology, a feature vector which is robust to geometrical distortion can be used in the process of the watermarking embedding and extracting. These algorithms combine the visual feature vector of images, encryption technology with the third party authentication, and are resistant to geometric attacks.The innovations of this dissertation are as follows:1. Proposing a DWT-DCT based multiple watermarks for medical image algorithm robust to geometric attacks. A novel algorithm is presented. A medical image feature vector is acquired in transform domain, and the watermarking can be embedded and extracted by using it. The algorithm has desired robustness against common attacks and geometric attacks. It can easily embed multiple watermarks which can effectively solve the problem of the electronic signature of multiple medical experts.2. Presenting a DCT based zero-watermarking for medical image encryption algorithm.This algorithm is grounded on DWT and DCT. Before the watermarking image is embedded into the original medical image, it is scrambled by Logistic transform to enhance the watermarking scheme secure. The algorithm innovatively uses the concept of zero-watermarking. Hence, the embedded watermarking has a very strong robustness against different attacks without affecting on the lesion area of the original medical image. Meanwhile, the watermarking scheme does not need the original medical image for watermarking extraction, so it is a blind watermarking.3. Creating a3D medical image watermarking algorithm. Firstly, this algorithm achieves a medical volume by3D reconstruction of2D medical image. Then, the phase character is utilized to embed and exact watermarking in the medical volume. It can effectively solve the visualization of3D medical volume, and enhance medical image security. The result of experiment indicates that the scheme is resistant to common attacks, and it has desired robustness against geometric attacks, such as JEPG, rotation, scaling and translation attacks, etc. Moreover, the content of medical image remains unchanged with the algorithm which is one kind of the zero-watermarking technology. This method has no restriction of the capture of watermarking.In summary, some creative watermarking algorithms for medical image are studied in this dissertation. By utilizing a feature vector of medical image, the watermarking can be embedded and extracted. The algorithm combines the visual feature vector of images, encryption technology, the third party authentication, with the zero-watermarking technology and has desired robustness against geometric attacks. They can effectively solve the problems of medical multiple watermarks, and can also protect the medical data. Meanwhile, they can effectively solve the problem of the visualization of3D medical volume.