Research And Implementation Of Large Capacity Information Hiding Technology Based On Video Format

With the development of computer technology and network technology, data exchanging and sharing through the network become frequently. In the process of storaging and exchanging for data, it often appears illegal phenomenon such as illegal access, replication, unauthorized use, change, and so on. If the secret of higher level is stolen, it will cause serious problem of leak, which may cause incalculable damage to the interests of the state. Information Hiding Technology conceals the existence of secret information, and hides secret information to the general file without being discovered, thus ensuring the content and behavior of invisibility. To ensure data security, we use information hiding technology to achieve distribution of data through networks. As a next-generation video code standard, H.264 has been rapidly gaining popularity and widespread application. Therefore, the combination with H.264 standard and video information hiding has some practical significance.This paper analyzes and summarizes recent technique of H.264 video information hiding based on the researched results. The thesis researches the shortcomings and tendency of existing H.264 video information hiding technology, combining the needs of geospatial data network distribution, this article makes a deep research on H.264 video steganography algorithms. The main works are as the following:1. When using Sobel edge detection operator,it is only sensitive to vertical and horizontal direction and can’t make some edge detection,which limit the detection performance of the Sobel algorithm. According to this problem, this article presented an egde detection algorithm based on odd Gabor filter. Firstly, Gabor transform should be carried out on the image, to find out the different coefficient of human eye sensitivity is low, and detect the edge of the transformed image with Sobel algorithm. Finally, we plus it with the image which detected by Sobel algorithm directly. Experimental results show that this algorithm modify Sobel algorithm’s deficiency of singleness in edge detection, containing more testing information.2. Consider of the characteristics of quantity of remote sensing image data, in order to transport and storage them easily, the text analysis and research some compression technology about remote sensing image, and make compares among them, then propose an optimized method based on integer wavelet transform about image compression. This method encodes the lowest frequency subband and high frequency subband separately, after multilevel wavelet transform. The experiment has shown the efficiency of the technique.3. Study on the embedding and processing of map data as an example, according to the problem of hiding capacity for video compressed domain is low, we put forward a method of open-loop hiding algorithm based on video. The algorithm makes use of visual identification threshold to determine maximum carrier coefficient of tolerate errors, to determine the information embedding strength, and improve the steganographic capacity in the guarantee of human visual perception tolerance range as far as possible. Combining the Watson sensing model is used to control the information embedding area. The experimental results show that, this method effectively improves the capacity of steganography, imperceptibility and robustness.4. Using different types of spatial data and video data, we give prototype system verification to important indicators. Results show that the algorithms effectively improve the main index, and they have applicability.