Perceptual Attack Resistance Selective Encryption after Video Compression H.264/AVC for Real Time Applications

Abstract

Selective encryption is a technique to save computational complexity or enable new system functionality by only encrypting a portion of a compressed bitstream of H.264/AVC. Selective encryptions are used for real time video streaming applications like, video conferencing, Live TV etc. The applications also need the perceptual encryption. During AVC encoding, such sensitive data as NZs in CAVLC, intra-prediction slice and motion vector are encrypted with the use of standard AES algorithm. This encryption scheme keeps secure against brute-force attack or any other classical attacks, combines encryption process with compression process with low cost, and keeps the file format unchanged (video codec compliance) with some direct operations (such as displaying, time seeking, copying, cutting, etc. means data operability). Many perceptual encryptions provide the lower encryption time, which are more suitable for real time applications. However, it cannot provide perception security. It is easy for cryptanalyst to reconstruct the video by applying perceptual attack and get better quality video without using key. This paper summarizes the result of these methods and proposed the better encryption method to resist the perceptual attack. All simulations are carried out in Intel Core i5 processor having 2.4GHz frequency, 4GB RAM with 64-bit window 7 operating system.