Research On Error Concealment Technology Based On H.264 Decoder

With the development of network and video compression technology, the application of transporting video signal through wireless channel and Internet has been more and more common. The compressed video signal will be susceptible to the channel disturbance, and cause error, resulting decline in the quality of video. Error concealment is an important technology to improve the video quality on the decoder, without increasing the transmission channel bandwidth, or the amount of information. According to repair theory, error concealment can be divided into spatial domain and time domain two categories, each with different applications.This paper describes the H.264 video coding standard in all aspects, then the existing spatial and temporal error masking algorithm carried out a detailed study of the integration of existing technology strengths and weaknesses of the I and P-frames, respectively, the error masking proposed to improve the algorithm. For I-frame error masking, the traditional linear interpolation for smooth video images cover a good effect, but correlation in the frame is weak recovery is difficult to achieve the desired quality. In this paper, based on traditional airspace shadowing errors, combined with time-domain error obscured the advantages of using inter-frame correlation is proposed based on video scene change carried adaptive I-frame error masking mode selection algorithm. The algorithm at the decoder according to the changes in intensity of the video scene to provide adaptive I-frame error recovery strategies, and to match during the macro block adjacent macroblocks according to the state to provide adaptive search strategy. Error for the P-frame shelter, taking into account the zero motion vector mask does not apply to campaign more intense images, boundary matching translational motion in the image content more severe cases, obtain good results, but the non-translational motion estimation is not precise enough, this an improved P-frame error masking algorithm, matching the best use of the traditional boundary matching macro-block, then use the arrow matching the macroblock boundaries in different directions for the second match, which is equivalent to the best reference for a macroblock rotation angle of arbitration, to improve the image in the event of non-translational motion recovery in case of video quality.