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 signifi-cantly reduced and the coding picture quality is al-most remained. Fuzzy logic technique with four in-puts, one motion degree and three edge energies dis-tribution of the image is used in the proposed algo-rithm 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.