Research Of Vedio Information Hiding Based On H.264

Since1994in which the first papers of information hiding was delivered. Thetechnology of information hiding wins a worldwide attention. Especially after theEvents of September11of2001, Information Hiding Technology shows a booming trend.By hiding secrets into multimedia carrier, covert communications, copyright protectionas well as some other uses is achieved. So far binary data such as: images, audio, video,3D models as well as programmer can used as a carrier in Information HidingTechnology. Among all of this multimedia, Images have the most paper in this area.Video follows. And Video has the least. This phenomenon is against the size andcomplexity of these media file. For this reason, the more complexity the carrier is, themore factors must be considered a hiding secrets to ensure its robustness.The main focuses in video-based Information Hiding filed are on original as well ascompressed video. However, Algorithm used for hiding secrets on original video hasbad robustness. That’s why in recent years, most researches of Information Hiding onVideo, usually taking video coding technology into consideration, become popular.The draft of H.264offers the model of video encoder as well as decoder. Theencoding process contains the inner frame coding and intra frame coding. During theinner frame coding, different sizes of macro blocks is used to store frame data andmotivation vector according to the relationship between the frame of the video prior andafterwards. And the sizes of macro block are decided by the Lagrange Distortion-RateAlgorithm. Afterwards DCT transform, quantizing and coding are performed.An information hiding algorithm based on compressed-video by modifying theDistortion-Rate Algorithm is proposed in this thesis according to the process above.Characteristics of this algorithm are showed below:1. Compared with other hiding algorithm, secret is not hidden into the content butthe format of the video. Therefore no data is loosed, and no visual effects will beaffected.2. The position of secret can be varied. Some hidden algorithm uses the LSB bit or the low-frequency region of DCT to hide secret. With fixed position, secret can bedetected easily. However the algorithm in this thesis can determines whether a macroblock to use or not randomly. From the decoding process, this does not take anyaffection on the recovery of secret.3. With well robustness. By compatible with the MPEG-4format, this algorithmnatively robust on MPEG compression. In addition, if the secret information is hiddeninto the video repeatedly, an ability to resist the attack of the video shearing, cutting, etcwill be seen.Finally a test based on the JM18.0project is done to test the ability of thisalgorithm and the results will be analyzed.