为了对福建罗源湾互花米草进行监测研究，以2000年和2006年的Landsat TM/ETM+影像为数据源，在对不同地物进行光谱分析的基础上，结合进行了影像归一化植被指数和缨帽变换处理，建立了利用决策树分类法提取罗源湾互花米草的遥感解译技术。结果表明基于决策树分类法能准确快速地对研究区互花米草的生长和分布状况进行动态监测。互花米草主要分布在环绕罗源湾的滩涂上，在2000年～2006年的6年中互花米草的面积增加了17.657 km2，年增长率为10.74%。特别值得关注的是，罗源湾的米草正处于迅速扩张的阶段，严重影响了当地的渔业和生态资源，有关部门需要引起重视。

In order to investigate and monitor spartina alterniflora resources in the Luoyuan Bay, this paper illustrates the image process and interpretation based on the Landsat TM/ETM+ remote-sensing data of 2000 and 2006. On the base of the spectral analysis of different ground objects, NDVI and tasseled cap transform adopted on multi-spectral images, the algorithms for extracting and rectifying the spartina alterniflora from other landscape types were determined by decision tress classify in combination with visual interpretation. The results showed clearly and truly that the distribution and expansion trend of spartina alterniflora and prove that the remote sensing approach was useful for monitoring smooth cordgrass invasion in Luoyuan Bay. The spartina alterniflora was distributed in coastal wetland around the Luoyuan Bay. The areas of spartina alterniflora were increased 17.657 km2 from 2000 to 2006, and the yearly increasing rate was 10.74%. Meanwhile, the spartina alterniflora was experiencing a quickly expand period and was required to treat with by local government.