Research And Implementation Of Jitter Video Compression Based On

With the development of information technology, video applications are inceasingly widespread in various fields of life, social security, military and so on. Because of the huge amount of data, video need to be compressed in their process of storage and transmission. The compression processing can save video’s storage space and transmission bandwidth. When shooting video, camera’s irregular movement will make the video random jitter. Compared to the stable video, the jitter video will not only badly affect the subjective visual effect, and will increase the interframe difference of the video. This will increase the difference between coding frame and reference frame and result in a large amount of residual data when compressing the jitter video. As a result it increases the compression code flow and reduces the efficiency of video compression. This paper makes a series of research to the jitter video compression.Firstly, a video stabilization method based on feature points matching is proposed. In view of the phenomenon that the feature points distribution is too close, a method which reverses maximum response feature points in a area to homogenize the points is put forward. With the problem that mismatched points and correct matching points existed on the foreground image will reduce the global motion estimation’s precision, a method is proposed to eliminate the false matches. This method is basing on the relations of the matching points’distance. The comparing experiment shows that compared with the traditional method basing on RANSAC algorithm, the proposed method is faster.Then, this paper studied the key technologies of the H.264 video coding standard. The interframe predictive coding process is analyzed in detail, the block mode selection strategy in this process is studied. Through the simulation experiment, two kinds of different block selection strategy RDO mode and normal mode is compared. The detailed procedure in coding the syntactic elements of the best block pattern, the number of reference frame, the motion vector and residual data is shown.Finally, the method of electronic image stabilization is introduced into video compression to compress jitter video. The experimental data show that when compressing jitter video, the proposed method can reduce the final data and improve the compression ratio.