Partial Video Encryption Technology Based On H.265

Visualization is the most important way of human perception of the world,and digital video follows the tide of IT and gets very rapid development.However,in order to obtain practical applications,as a carrier of very large amount of data,video must be compressed and coded.H.265/HEVC,a more mature coding standards,is jointly issued by ITU-T and ISO / IEC.Compared with the previous generation H.264 / AVC,H.265 can save 50% bit rate at the same coding quality,it also has a parallel processing design,and its hardware and software implementation will have more practicality,so its appearance will quickly replace H.264 / AVC,meanwhile,it is widely sought after by video business.But its appearance did not change the video information was frequently malicious attacks and destruction of the fate,so based on H.265 / HEVC video encryption technology is paying more and more attention.In recent years,research scholars have designed a number of encryption algorithms,the full encryption method simply treated the video data as a series of ordinary binary stream and then encrypted,although its security is high,it makes its usability weak because of its huge computation.And then there are encryptions based on DCT coefficient and entropy coding process,which is combined with the encoding and decoding process.this algorithm can greatly reduce the amount of encrypted data,but for the security of the algorithm is relatively reduced.In order to reduce the amount of computation while having a considerable level of encryption,and without affecting the compression efficiency,so in this paper,we proposed a new encryption technology,its work content and purpose are as follows:Understand and analysis prediction,transformation,quantization and other parts in H.265,meanwhile,learn some of the existing encryption algorithm,compare its advantages and disadvantages and computational complexity,and then We want to design a video encryption algorithm to achieve the video part of the encryption which is based on the whole things we have learned,and at the same time the image can be rebuilt even without the secret key.Compared to the already existing encryption method in the entropy coding stage or the bit stream,we propose encryption in the transformation phase:according to the change of the rotation angle of the second stage in the butterfly,a series of transformation matrices are designed and verified to prove whether these matrices will affect the coding efficiency.Then,a series of new transformation matrices are selected by the RC4 algorithm.and then the new transform would repace the original DCT.So that if the decoding end does not know the key,we can achieve the purpose of the partial encryption of the video,because which particular transformation matrix is used is unknown.Apply our algorithm to a specific video sequence,and then analyze the video quality of the decoding side with respect to the different degrees of key information,and evaluate the encryption level.