Research On Compressing Algorithm For Video Coding Based On 3D-Wavelet Transform

As modern communication and network technology developed rapidly, the technique of video compression has showed more important, which played a necessary role in video service, such as Multimedia, High-Definition TV ( HDTV ) , video telephone, Video-On-Demand(VOD), conference TV and so on. AS a result it has wide application foreground and huge commercial value.Wavelet transformation is a very effective method in video coding, which is characteristic of well spatial and frequency multi-resolution. This method can catch unbalanced stochastic signal so that it can obtain higher compression ratio, also its multi-resolution trait approach the vision feature of HVS(Human Visual System). What's more important is that it has capacity of Scalable Video coding, which can prevent from the problem of error codes transmission and reduce complication..This paper not only introduces some video compressing algorithm using 3D-Wavelet Transforms, but also presents an algorithm for video coding based on 3D-wavelet transformation with temporal enhancement. Compared to normal 3D-wavelet transformation, the algorithm quantized various data according to human visual system. Especially, the quantizing method using MSE ( Mean Square Error ) as criterion to quantize temporal high-frequency, solving the matter that the large number of data is created when motion is acuity .Furthermore, the approach will not be concerned with motion estimation and motion compensation. The simulation experiment shows that it is validity and feasibility.