基于HJ-1影像与地形因子的竹资源专题信息提取研究
|
摘要
以永安市的DEM数据和HJ-1多光谱影像为数据源,研究不同坡度、坡向、海拔等地形条件下竹资源的光谱反射率曲线及辅助于地形因子的竹资源专题信息提取效果。结果表明,海拔、坡向对竹资源的光谱反射率影响大,而坡度影响小,在海拔400m~800m和800m~1200m范围内阳坡、半阳坡、阴坡、半阴坡的竹资源的光谱反射率差异最为明显。采用最大似然法结合地形因子辅助分类提取的竹资源专题信息精度为74.50%,比单纯依靠光谱信息分类的精度高14个百分点,说明地形因子对竹资源专题信息提取具有很大帮助,可以辅助解决"同物异谱"和"异物同谱"问题。
The effect of the bamboo thematic information extraction assisted by the terrain factors and the spectral reflectance curve of bamboo resources in different terrain conditions(slope,aspect,elevation)were analyzed with the DEM data and RS data in Yong'an City.The results show that the most impact terrain factors on the spectral reflectance curve of bamboo resources are altitude and aspect,the less impact factors is slope.There are obvious differences on the spectral reflectance curve of bamboo resources between the altitude of 400m~800m and 800m~1200m under the sunny aspect,semi-sunny aspect,shady aspect and semi-shady aspect.The accuracy of bamboo thematic information extraction by the maximum likelihood method with the help of the terrain factors is 74.50% which is 14 point higher than the classification based on spectral information only.It shows that the terrain factors could be great help for the bamboo thematic information extraction and could solve problems in the images about"the same object with different spectrums"and"the same spectrum for different objects".
The effect of the bamboo thematic information extraction assisted by the terrain factors and the spectral reflectance curve of bamboo resources in different terrain conditions(slope,aspect,elevation)were analyzed with the DEM data and RS data in Yong'an City.The results show that the most impact terrain factors on the spectral reflectance curve of bamboo resources are altitude and aspect,the less impact factors is slope.There are obvious differences on the spectral reflectance curve of bamboo resources between the altitude of 400m~800m and 800m~1200m under the sunny aspect,semi-sunny aspect,shady aspect and semi-shady aspect.The accuracy of bamboo thematic information extraction by the maximum likelihood method with the help of the terrain factors is 74.50% which is 14 point higher than the classification based on spectral information only.It shows that the terrain factors could be great help for the bamboo thematic information extraction and could solve problems in the images about"the same object with different spectrums"and"the same spectrum for different objects".
引文
[1]赵英时.遥感应用分析原理与方法[M].北京:科学出版社,2003.
[2]丁丽霞,王祖良,周国模,等.天目山国家级自然保护区毛竹林扩张遥感监测[J].浙江林学院学报,2006,23(3):297-300.
[3]颜梅春,张友静,鲍艳松.基于灰度共生矩阵法的IKONOS影像中竹林信息提取[J].遥感信息,2004,(2):31-34.
[4]余坤勇,许章华,刘健,等.“基于片层-面向类”的竹林信息提取算法与应用分析[J].中山大学学报(自然科学版),2012,51(1):89-95.
[5]刘健,余坤勇,许章华,等.竹资源专题信息提取纹理特征量构建研究[J].遥感信息,2010,(6):87-94.
[6]梁益同,万君.基于HJ-1A/B-CCD影像的湖北省冬小麦和油菜分布信息的提取方法[J].中国农业气象,2012,33(4):573-578.
[7]王羲程,曹广超,张富玲,等.结合TM影像和DEM两种数据源的青藏高原湿地提取方法[J].测绘科学,2013,38(4):40-42.
[8]冯益明,智长贵,姚爱冬.基于决策树的戈壁信息提取研究[J].干旱区地理,2013,36(1):125-130.
[9]乔艳雯,臧淑英,那晓东.基于决策树方法的淡水沼泽湿地信息提取——以扎龙湿地为例[J].中国农学通报,2013,29(8):169-174.
[10]吴见,刘民士,李伟涛.不同地形条件下植被盖度信息提取技术研究[J].植物生态学报,2013,37(1):18-25.
[11]冯朝阳,于勇,高吉喜,等.地形因子对京西门头沟区土地利用/覆盖变化的影响[J].山地学报,2007,25(3):274-279.
[12]黎景良,后斌,危双峰,等.基于DEM的广东省山区土地利用变化分析[J].测绘通报,2007,(6):53-57.
[13]HELMI Z M,AFFENDI S,SHATTRI M.The performance of maximum likelihood,spectral angle mapper,neural network and decision tree classifiers in hyperspectral image analysis[J].Journal of Computer Science,2007,3(6):419-423.
[14]余坤勇,刘健,许章华,等.南方地区竹资源专题信息提取研究[J].遥感技术与应用,2009,24(4):449-455.
[2]丁丽霞,王祖良,周国模,等.天目山国家级自然保护区毛竹林扩张遥感监测[J].浙江林学院学报,2006,23(3):297-300.
[3]颜梅春,张友静,鲍艳松.基于灰度共生矩阵法的IKONOS影像中竹林信息提取[J].遥感信息,2004,(2):31-34.
[4]余坤勇,许章华,刘健,等.“基于片层-面向类”的竹林信息提取算法与应用分析[J].中山大学学报(自然科学版),2012,51(1):89-95.
[5]刘健,余坤勇,许章华,等.竹资源专题信息提取纹理特征量构建研究[J].遥感信息,2010,(6):87-94.
[6]梁益同,万君.基于HJ-1A/B-CCD影像的湖北省冬小麦和油菜分布信息的提取方法[J].中国农业气象,2012,33(4):573-578.
[7]王羲程,曹广超,张富玲,等.结合TM影像和DEM两种数据源的青藏高原湿地提取方法[J].测绘科学,2013,38(4):40-42.
[8]冯益明,智长贵,姚爱冬.基于决策树的戈壁信息提取研究[J].干旱区地理,2013,36(1):125-130.
[9]乔艳雯,臧淑英,那晓东.基于决策树方法的淡水沼泽湿地信息提取——以扎龙湿地为例[J].中国农学通报,2013,29(8):169-174.
[10]吴见,刘民士,李伟涛.不同地形条件下植被盖度信息提取技术研究[J].植物生态学报,2013,37(1):18-25.
[11]冯朝阳,于勇,高吉喜,等.地形因子对京西门头沟区土地利用/覆盖变化的影响[J].山地学报,2007,25(3):274-279.
[12]黎景良,后斌,危双峰,等.基于DEM的广东省山区土地利用变化分析[J].测绘通报,2007,(6):53-57.
[13]HELMI Z M,AFFENDI S,SHATTRI M.The performance of maximum likelihood,spectral angle mapper,neural network and decision tree classifiers in hyperspectral image analysis[J].Journal of Computer Science,2007,3(6):419-423.
[14]余坤勇,刘健,许章华,等.南方地区竹资源专题信息提取研究[J].遥感技术与应用,2009,24(4):449-455.