论文部分内容阅读
利用2015年4~11月8景C波段Rardarsat-2影像SGX产品数据和2015年9月10日Landsat 7的7个波段的反射率数据,采用面向对象分类方法,获得8个日期的扎龙湿地淹水区,计算出淹水频次;选取的19个特征变量分别为Landsat 7影像的1波段~7波段的反射率、归一化植被指数、二阶距、对比度、相关性、差异性、能量、同质性、中值、HH和HV极化波段的后向散射系数、淹水范围和淹水频率,采用随机森林算法,提取扎龙湿地的土地利用类型信息。研究结果表明,有淹水频率参与的分类精度为91.73%,无淹水频率参与的分类精度为76.49%,精度提高主要体现在沼泽湿地与草地的分类上。本研究为准确地更新湿地基础信息提供了方法示范。
Based on the data of 8 C-band Rardarsat-2 image SGX products from April to November 2015 and the reflectance data of 7 bands from Landsat 7 on September 10, 2015, an object-oriented classification method was used to obtain 8 date Zhalong wetlands The frequency of flooding was calculated. The 19 eigenvalues selected were the reflectance from 1 band to 7 band of Landsat 7 image, normalized vegetation index, second-order distance, contrast, correlation, difference, energy , Homogeneity, median, backscattering coefficient of HH and HV polarization bands, flooding range and flooding frequency, random forest algorithm was used to extract the land use type information of Zhalong wetland. The results show that the classification accuracy of flooding frequency participation is 91.73% and the classification accuracy of no flooding frequency is 76.49%. The improvement of accuracy is mainly reflected in the classification of marsh and grassland. This study provides a demonstration of how to accurately update wetland basic information.