After the reference genomes of many organisms are sequenced in this post-genetic era, it has become an extremely important issue that how to do the re-sequencing and assembly for individual genomes from very large amount of reads. In this paper, we will present a re-sequencing tool designed for the Next Generation Sequencing (NGS) data. And these data are composed of a huge amount of short reads which will be aligned onto a reference genome. We modified and implemented the algorithm of Burrows-Wheeler Transform and FM-index to build the genome index of human, and proposed an idea to segment each short read into multiple non-overlapping seeds, which let us align short reads with large Hamming distance. Finally, we used 4 real data sets with different lengths form 1000 Genome Project to demonstrate the performance of our tool with a personal computer, and compared the results with a widely used tool, bowtie.