Fast Mode Decision Using Fuzzy Reasoning for Motion Estimation in H.264/AVC Video Coding

terframe motion estimation/compensation can select the most suitable encoding mode from seven modes (16×16, 16×8, 8×16, 8×8, 8×4, 4×8 and 4×4) by the method of Lagrangian rate-distortion (RD) optimization. The Lagrangian function is shown in eq.(1), To improve the coding efficiency, many optional macroblock encoding modes have been provided in H.264/AVC video coding standard. Exhaustively searching over all possible modes to choose the best coding mode for each macroblock is the most reliable one, but it causes a lot of computational complexity. This paper proposes a fast algorithm to select the coding mode such that the number of candidate modes to be searched in motion estimation is significantly reduced and the coding picture quality is almost remained. Fuzzy logic technique with four inputs, one motion degree and three edge energies distribution of the image is used in the proposed algorithm to select the most appropriate mode for the motion estimation in H.264/AVC video coding. The simulation results show that the encoding time in the proposed algorithm is reduced to 55.7%~70% of the time searching for all coding modes exhaustively and the coded picture quality is not affected almost. ( , , | , ) ( , , | ) ( , , | ), Mode Mode J s c Mode QP SSD s c Mode QP R s c Mode QP λ λ =