Research On Digital Watermarking Technology For Medical Image

With the development of the Internet and the construction of medical informatization,the hospital produces a large number of digital medical images,and the storage of patient information is also converted from the traditional paper form to the electronic form.Digitalization technology brings convenience to information exchange among hospitals and cooperative institutions,as well facing the risk of illegal copying,stealing and disclosure of information during network transmission.Researchers have proposed a variety of methods to protect medical images and patient privacy information,in which,digital watermarking is an important technology that is often used.Existing medical image watermarking methods have problems of poor universality and low embedding capacity.In order to solve the above problems,the paper research digital watermarking technologies based on transform domain,the main work is as the following:For the problem of poor universality of medical image digital watermarking based on ROI,and the problem that the carrier can't block fitly in the block-based watermarking algorithms,an adaptive full-blind medical image watermarking algorithm based on sub-block segmentation(AMWASS)is proposed.Before embedding the watermarking,the RGB channel is separated from the color carrier image.The first watermark is the copyright image,which is fine-tuned according to the size of the carrier.The separated R channel is evenly partitioned and transformed by DCT,and the copyright image is encrypted using the feature of the image block.The encrypted information is embedded into each sub-block in R channel image by adjusting the coefficients in DCT domain.The second watermark is the length and width of the original copyright image,which is embedded into the DWT domain of G-channel image by SVD.When extracting watermark,the second watermark is processed firstly,and this information is used to determine the block ing of R-channel image and restore the extracted copyright image.The simulation results show that this algorithm has good invisibility and robustness,it can resist various attacks such as noise,filtering and so onFor the hiding problem of patient information,watermarking algorithms based on SVD have capacity limitation,and mismatch between carrier and watermark.This paper presents an adaptive medical image watermarking algorithm based on DWT-SVD and live code(AWDLC).Firstly,the copyright identification and patient information are converted into a live code,then the live code is adjusted according to the carrier.Finally,pre-processed live code is embedded into the singular value matrix of the second-level DWT domain of carrier,using the method of additive embedding.Experimental results show that the algorithm can resist common attacks and can be decode accurately.In this paper,a kind of watermarking algorithm based on sub-block is designed for the problem of copyright protection in medical images,which improves the limitations of the existing watermarking algorithms and reducing the expense of the system.At the same time,the embedding of rich patients' information is realized using live code.