The Key Technologies Of H.264 Video Coding Based On Visual Perception

Conventional video coding systems derive from the rate distortion theory which focuses on eliminating the temporal, spatial and codeword redundancies of the video data. H.264/AVC is a landmark of these conventional systems because it achieves a significant coding gain. However, how to efficiently analyze the video content is not fully studied, yet so H.264/AVC is not capable enough to compress the visual redundancy. If the video coder can simulate the Human visual System and understand the video content,video compression will be established on a visual observation basis, and thus the coding framework and coding efficiency will be significantly improved. This paper is based on Just-Noticeable-Distortion(JND) model and H.264/AVC standard, mainly focus on the JND model concerned motion factors and the rate-distortion theory revised by JND model.Firstly, by analysing the existing DCT-based JND model, which only considering the CFS, the background effect and the block-based effect, this paper proposed the motion-based dynamic JND model adopting to the video rather than the still picture. According to the human visual system property that the motion intensity is inversely to JND model, the algorithm using motion factor to amend existing JND model which is complicated. Secondly, considering the Lagrangian optimization techniques applied in rate control algorithm for H.264, this paper amend the Lagrangian model by integrating the property of human visual system, doing a lot of experiments and combining the visual weight in each macroblocks. Meanwhile, this paper proposes a fast selecting mode algorithm by using JND model and fixing rate distortion formula.Experiments shows that the proposed algorithm can monitor objects accurately and have a better performance both in subjective quality and coding speed.