A Robust Watermarking For Medical Volume Data Based On Three Dimensional Wavelet Transform And Arnold Scrambling

With the digital information system is widely used in the modern medicine, how to avoid tampering and thieving medical information on the internet、how to ensure the information safety and protect the copyright of medical images, has already attracted great attention in academia. Therefore, considering many aspects such as the simple algorithm, the watermarking robustness and information security, the paper provides a robust algorithm for medical volume data based on Arnold scrambling and the three dimensional wavelet transform.Firstly, the3D medical volume data as the research object, basing on the3D-DWT principle, the pixel array with strong correlation is mapped the low frequency and high frequency components, the energy is highly concentrated in low frequency that represents the main information, it has strong correlation coefficient and large coefficients, the rest high frequency region, in addition to the detail and edge, has many small coefficients that zero and close to zero, it is unimportant detail component. In addition, considering the human visual system (HVS), the low signal has a greater impact on human vision, so choosing the low-frequency portion to realize the watermarking embedding and extract some low frequency coefficients for three-dimensional reconstruction, in order to remove spatial redundancy and compress the medical volume data. Finally, using the storage space efficiently, transmits data quickly, and enhance the stability and security of the watermark information. Afterwards, the paper uses Hash function to process information, therefore, no matter how length the carried information, after mapping information into a hash value through Hash function, the real value of the embedded watermark information is not so much, so the algorithm has lower computational complexity and can flexibly control the hidden information; This paper focuses on the watermarking algorithm based on three-dimensional wavelet transform, however, the3D-DWT can be not studied alone and must combine with other transform, because it has no geometric invariance. Firstly, studying the3D-DFT and3D-DCT transform respectively, so as to observe the robustness, then combining the3D-DWT transform with above two transforms, finally, analyzing the simulation results and robustness, thus more confirm that the algorithm based on three-dimensional wavelet transform has quickly computing speed, high precision, good compatibility, strong ability to against the attacks, etc. It is a good way to satisfy robustness, abstraction and security.