With the surge in sequence data of biological sequence database, developing a algorithm which has the high biology sensitivity and efficiency is very urgent. Based on the deep analysis on the Needleman-Wunsch and Smith-Waterman Algorithm of bio-sequence alignment, the author enhances the Smith-Waterman algorithm as well as proves its accuracy through a series of experiments in this paper. Comparing between the Smith-Waterman algorithm and the improved one, the author analyzes the performance of the two algorithms. Experimental results show that the newly improved algorithm can optimize the number of local optimal solutions for pairwise sequence, reduce the complexity in time and space of bio-sequence alignment algorithms, and increase the scores and accuracy of sequence alignments.