改进线性光谱混合分解模型湿地信息提取_吴见.pdfVIP

2011, 16( 3) : 140- 144 Journal of China Agricultural Universit y 吴见 彭道黎 ( , 100083) 由于干旱区农牧交错地带湿地面积小且地类混杂, 混合像元现象严重, 使得该区湿地信息 自动提取难 度较大针对湿地遥感信息提取 特点和难点, 采用 NDVI( normalized different vegetation index) 阈值法提取水 体, 应用支持向量机模型( SVM) 提取去除水体后 湿地信息, 并以修正线性光谱混合分解模型分解 草甸分量, 进 一步提取高盖度中盖度和低盖度草甸信息试验验证结果表明: 提取结果 总体精度为 88% , Kappa 系数为 083该方法可为其他光谱特征混杂地区湿地遥感信息 提取提供参考 湿地; 线性光谱混合分解模型; 遥感; 农牧交错地带; 植被覆盖度 T P 79 1007-4333(2011) 03-0140-05 A Wetland information extraction based on improved linear spectral mixture model * WU Jian, PENG Dao- li ( The Ke y Laboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing 100083, China) Abstract Automatic extraction of w etland information in farming-pastoral regions is very difficult, becaus e land us e types are complex and etland area is small w hich lead to erious mixed pixel phenomenon. Aming at the characteristics and difficulties of wetland remote sensing information extraction in these regions, the NDVI ( Normalized Different Vegetation Index) threshold method w as used to extract w ater body first.Afterwards, the support vector machine( SVM) was selected to extract wetland information which has removed water. Finally, the meadow component which w as decomposed by linear spectral mixture model w as amended by the high-resolution satellite image to further extract high coverage, medium coverage and low coverage meadow.The experimental results show ed that the overall accuracy w as 88% and the Kappa coefficient was 0. 83. This method can provide a reference of remote s ensing information extract