According to the characteristics of wetland vegetation,the paper used the soil line automatic identification algorithm and NIR-Red dimensional feature space to extract soil line for SPOT 5 data acquired on May 7,2005.The paper also calculated perpendicular vegetation index to extract vegetation information in the Jiuduansha wetland,and the classification accuracy reached 86.5 %.Compared with the classification result of the maximum likelihood classification,the accuracy increased 5.7%.%针对潮滩湿地植被的特点,利用2005年5月7日的SPOT 5数据,结合自动提取土壤线算法和近红外一红外二维光谱特征空间,获取土壤线参数,并利用垂直植被指数对九段沙湿地典型试验区进行植被信息提取.计算结果表明,土壤线计算方法简单可靠,利用垂直植被指数进行分类其精度达到86.5%,相比最大似然分类方法提高了5.7%.
展开▼