Research On Video Encryption Technology Based On H.264 Coding Standard

With the continuous development and innovation of computer network and wireless mobile communication technology,digital media data(image,sound,video,etc.)are increasingly penetrating into people's lives.Therefore,multimedia video applications based on network transmission are more and more diversified,such as related video conference,video on demand,real-time video monitoring system,etc.Similar to other data transmitted through the network,unencrypted video data is vulnerable to data information interception,data tampering,illegal copying and piracy in open network transmission.The security of video data is very important for the application of multimedia,and video encryption technology is the most effective tool for the security during the protection period.At present,most encryption schemes use selective encryption to encrypt some important and sensitive video information to ensure real-time and encryption efficiency.The concept of selective video encryption has been introduced in the past few years.Different from the traditional completely secret video encryption,selective video encryption does not require that the encrypted video signal be robust to all complex and time-consuming security attacks.The main purpose is to provide a certain degree of protection for video signal by using a fast and effective algorithm.Therefore,on the basis of h.264/avc standard,this paper first proposes that encryption can be carried out in the conversion stage of video coding process.Firstly,some unitary transformations based on plane rotation are designed in the famous DCT flow graph structure,and it is proved that these new transformations are as effective as DCT in the coding of predicted residual signals.According to the secret key generated by the properties of multiplication group in prime field,a new encryption system is constructed by selecting one from several transformations,and the validity of the algorithm is verified.In order to further reduce the quality of video viewing,this paper also modifies the syntax elements in the CAVLC process and the motion vectors in the inter prediction process,which provides sufficient protection for video visual information and ensures the compatibility of H.264 video format.Because the multiplication group of prime field has no floating-point operation,the operation efficiency is greatly improved,and it can meet the real-time demand of video transmission in the Internet.Finally,the paper gives the video quality evaluation results of PSNR and SSIM,and proves the effectiveness of the algorithm.The hardware and software platform are designed to analyze the security of the algorithm.The experimental results show that the proposed algorithm is highly sensitive to the key.