基于SEaTH的决策树方法在区域尺度土地覆被分类中的应用
|
摘要
利用最大值合成法处理洮河流域2000年MODIS数据,得到归一化植被指数年最大值栅格数据,结合该区数字地面模型构造土地覆被分类数据集,采用SEaTH算法提取不同地类的特征阈值,构建决策树,对洮河青藏片区和黄土片区进行土地覆被分类,与NLCD-2000数据相对比进行精度评价.结果表明:决策树法能够较好实现洮河流域主要地物的识别并反映其宏观分布格局.青藏片区地物分类的总体精度为74%, Kappa系数为0.71;黄土片区地物分类的总体精度为63.8%, Kappa系数为0.57;青藏片区的分类效果总体要好于黄土片区.与最大似然法相比,决策树法在青藏片区的分类精度提高约10%,黄土片区分类精度提高约5%.
The maximum value composites, method was used for filtering the maximum raster normalized difference vegetation index(NDVI) value of Tao River Basin in 2000. Combined with the digital terrain model dataset, the SEaTH algorithm was adopted to determine thresholds and to form decision trees for land cover classification in the two diverse regions of the Eastern Tibetan Plateau(ETP) and Southwestern Loess Plateau(SLP) occupying the basin. Results of the classification were calibrated and validated by a comparison between the NLCD-2000 data and land types retrieved by the decision trees, showing that the decision trees could discern major land cover types and reflect the macro distribution pattern of the basin's land cover. The overall classification accuracy of ETP was validated to 74%and the Kappa coefficient was 0.71, whereas in the SLP, the overall classification accuracy of ETP was63.8% and the Kappa coefficient was 0.57. Compared with the classification results by the Maximum Likelihood method, decision trees' implementation increased the classification precision by about 10%and 5% in the ETP and the SLP, respectively.
The maximum value composites, method was used for filtering the maximum raster normalized difference vegetation index(NDVI) value of Tao River Basin in 2000. Combined with the digital terrain model dataset, the SEaTH algorithm was adopted to determine thresholds and to form decision trees for land cover classification in the two diverse regions of the Eastern Tibetan Plateau(ETP) and Southwestern Loess Plateau(SLP) occupying the basin. Results of the classification were calibrated and validated by a comparison between the NLCD-2000 data and land types retrieved by the decision trees, showing that the decision trees could discern major land cover types and reflect the macro distribution pattern of the basin's land cover. The overall classification accuracy of ETP was validated to 74%and the Kappa coefficient was 0.71, whereas in the SLP, the overall classification accuracy of ETP was63.8% and the Kappa coefficient was 0.57. Compared with the classification results by the Maximum Likelihood method, decision trees' implementation increased the classification precision by about 10%and 5% in the ETP and the SLP, respectively.
引文
[1]李常斌,杨林山,杨文瑾,等.洮河流域土地利用/土地覆被变化及其驱动机制研究[J].地理科学, 2014, 34(7):848-855.
[2]安美玲,张勃,孙力炜,等.黑河上游土地利用动态变化及影响因素的定量分析[J].冰川冻土, 2013, 35(2):355-363.
[3] Lunetta R S, Knight J F, Ediriwickrema J, et al. Land-cover change detection using multi-temporal MODIS NDVI data[J]. Remote Sensing of Environment, 2006, 105(2):142-154.
[4] Wardlow B D, Egbert S L, Kastens J H. Analysis of time-series MODIS 250m vegetation index data for crop classification in the US Central Great Plains[J]. Remote Sensing of Environment, 2007, 108(3):290-310.
[5]李耀辉,韩涛.基于EOS/MODIS资料的中国黄土高原西部土地覆盖分类[J].高原气象, 2008, 27(3):538-543.
[6]孙艳玲,杨小唤,王新生,等.基于决策树和MODIS数据的土地利用分类[J].资源科学, 2007, 29(5):169-174.
[7]张雪蕾,王万瑞,王刘明,等.石羊河流域干旱变化趋势及气候影响因素[J].兰州大学学报:自然科学版, 2017,53(5):593-597.
[8]顾娟,李新,黄春林.基于时序MODIS NDVI的黑河流域土地覆盖分类研究[J].地球科学进展, 2010, 25(3):317-326.
[9]药静宇,王国印,李晨蕊,等.不同区域典型植被类型的净生态系统碳交换特征[J].兰州大学学报:自然科学版, 2017, 53(5):622-627.
[10]李奇虎,陈亚宁. 1981-2006年西北干旱区NDVI时空分布变化对水热条件的响应[J].冰川冻土, 2014, 36(2):327-334.
[11] Li Chang-bin, Qi Jia-guo, Yang Lin-shan, et al. Regional vegetation dynamics and its response to climate change:a case study in the Tao River Basin in Northwestern China[J]. Environmental Research Letters, 2014, 9(12):653-658.
[12]李常斌,王帅兵,杨林山,等. 1951-2010年洮河流域水文气象要素变化的时空特征[J].冰川冻土, 2013, 35(5):1259-1266.
[13]李常斌,杨林山,王帅兵,等.洮河流域潜在蒸散发估算:模式遴选及应用[J].兰州大学学报:自然科学版,2013, 49(6):727-732.
[14]宋富强,康慕谊,郑壮丽,等.陕北黄土高原地区土地利用/覆被分类及验证[J].农业工程学报, 2011, 27(3):316-324.
[15]王志慧,李世明,刘良云,等.基于MODIS NDVI时间序列的土地覆盖分层分类方法研究[J].遥感技术与应用, 2013, 28(5):910-919.
[16] Nussbaum S, Niemeyer I, Canty M J. Feature recognition in the context of automated object-oriented analysis of remote sensing data monitoring the Iranian nuclear sites[J]. Proceeding of SPIE, 2005, 5988:5-9.
[17]余晓敏,湛飞并,廖明生,等.利用改进SEaTH算法的面向对象分类特征选择方法[J].武汉大学学报:信息科学版, 2012, 37(8):921-924.
[18] Johansen K, Bartolo R, Phinn S. Special feature:geographic object-based image analysis[J]. Journal of Spatial Science, 2010, 55(1):3-7.
[19]刘勇洪,牛铮,王长耀.基于MODIS数据的决策树分类方法研究与应用[J].遥感学报, 2005, 9(4):405-412.
[2]安美玲,张勃,孙力炜,等.黑河上游土地利用动态变化及影响因素的定量分析[J].冰川冻土, 2013, 35(2):355-363.
[3] Lunetta R S, Knight J F, Ediriwickrema J, et al. Land-cover change detection using multi-temporal MODIS NDVI data[J]. Remote Sensing of Environment, 2006, 105(2):142-154.
[4] Wardlow B D, Egbert S L, Kastens J H. Analysis of time-series MODIS 250m vegetation index data for crop classification in the US Central Great Plains[J]. Remote Sensing of Environment, 2007, 108(3):290-310.
[5]李耀辉,韩涛.基于EOS/MODIS资料的中国黄土高原西部土地覆盖分类[J].高原气象, 2008, 27(3):538-543.
[6]孙艳玲,杨小唤,王新生,等.基于决策树和MODIS数据的土地利用分类[J].资源科学, 2007, 29(5):169-174.
[7]张雪蕾,王万瑞,王刘明,等.石羊河流域干旱变化趋势及气候影响因素[J].兰州大学学报:自然科学版, 2017,53(5):593-597.
[8]顾娟,李新,黄春林.基于时序MODIS NDVI的黑河流域土地覆盖分类研究[J].地球科学进展, 2010, 25(3):317-326.
[9]药静宇,王国印,李晨蕊,等.不同区域典型植被类型的净生态系统碳交换特征[J].兰州大学学报:自然科学版, 2017, 53(5):622-627.
[10]李奇虎,陈亚宁. 1981-2006年西北干旱区NDVI时空分布变化对水热条件的响应[J].冰川冻土, 2014, 36(2):327-334.
[11] Li Chang-bin, Qi Jia-guo, Yang Lin-shan, et al. Regional vegetation dynamics and its response to climate change:a case study in the Tao River Basin in Northwestern China[J]. Environmental Research Letters, 2014, 9(12):653-658.
[12]李常斌,王帅兵,杨林山,等. 1951-2010年洮河流域水文气象要素变化的时空特征[J].冰川冻土, 2013, 35(5):1259-1266.
[13]李常斌,杨林山,王帅兵,等.洮河流域潜在蒸散发估算:模式遴选及应用[J].兰州大学学报:自然科学版,2013, 49(6):727-732.
[14]宋富强,康慕谊,郑壮丽,等.陕北黄土高原地区土地利用/覆被分类及验证[J].农业工程学报, 2011, 27(3):316-324.
[15]王志慧,李世明,刘良云,等.基于MODIS NDVI时间序列的土地覆盖分层分类方法研究[J].遥感技术与应用, 2013, 28(5):910-919.
[16] Nussbaum S, Niemeyer I, Canty M J. Feature recognition in the context of automated object-oriented analysis of remote sensing data monitoring the Iranian nuclear sites[J]. Proceeding of SPIE, 2005, 5988:5-9.
[17]余晓敏,湛飞并,廖明生,等.利用改进SEaTH算法的面向对象分类特征选择方法[J].武汉大学学报:信息科学版, 2012, 37(8):921-924.
[18] Johansen K, Bartolo R, Phinn S. Special feature:geographic object-based image analysis[J]. Journal of Spatial Science, 2010, 55(1):3-7.
[19]刘勇洪,牛铮,王长耀.基于MODIS数据的决策树分类方法研究与应用[J].遥感学报, 2005, 9(4):405-412.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |