云检测阈值自动生成(CDAG)算法是一种自动生成阈值的云检测方法,这种方法是使用预先准确确定出云和晴空像元的AVIRIS高光谱数据模拟出不同传感器的云和晴空像元数据,据此生成它们的云检测阈值。这种方法具有自动化程度高、云检测效果好...云检测阈值自动生成(CDAG)算法是一种自动生成阈值的云检测方法,这种方法是使用预先准确确定出云和晴空像元的AVIRIS高光谱数据模拟出不同传感器的云和晴空像元数据,据此生成它们的云检测阈值。这种方法具有自动化程度高、云检测效果好等优点。但是在原数据库的构建中,由于AVIRIS数据集中雪像元较少,导致云和雪的识别存在较大的误差。为了解决该问题,从待测数据中提取大量不同类型的雪像元,加入模拟得到的多光谱数据集中,生成新的云检测阈值,阈值生成中使用了单波段反射率、多波段反射率组合参数等作为云检测算法的输入参数。使用2013—2017年青藏高原地区多景典型Landsat8陆地成像仪(Operational Land Imager,OLI)影像进行实验验证。结果显示,在积雪覆盖的区域平均云像元识别正确率达到93.11%,对包括雪在内的晴空像元的漏判率降低到4.35%,表明此方法能有效避免雪的影响,实现高精度的云检测。展开更多
The biosphere of the Earth is essential to human survival and development.The services of ecosystems are critical to the functioning of the Earth’s life-support system.They contribute to human welfare both directly a...The biosphere of the Earth is essential to human survival and development.The services of ecosystems are critical to the functioning of the Earth’s life-support system.They contribute to human welfare both directly and indirectly.Ecological capital refers to the sum of the direct biological resources value and the indirect ecosystem services value.It is necessary to estimate the ecological capital in order to bring it to the society and market economic system,and draw the social attention to ecological environment constructions.An estimation model for eco-logical capital based on remote sensing is presented in this paper.The parameters in the model are quantitatively measured using NOAA/AVHRR and other ancillary data,including the land cover types,the vegetation coverage,and the vegetation net primary productivity(NPP)of ter-restrial ecosystem.Based on the economic parameters in previously published studies and a few original calculations,the annual ecological capital of the entire terrestrial ecosystem of China is quantitatively estimated at 6.44 trillion(10^(12))yuan(RMB),and the spatial distribution of the eco-logical capital is also analyzed.Traditional ecological methods to ecological capital measurement are based on homogeneous plot scales,and the regional scaling is a key problem in their appli-cations.As the proposed remote sensing approach,it provides a new method to ecological capital measurement completely based on observation data.It can not only overcome the re-gional scaling problem easily,but also allows the ecological capital to be estimated objectively and spatial-explicitly.展开更多
文摘云检测阈值自动生成(CDAG)算法是一种自动生成阈值的云检测方法,这种方法是使用预先准确确定出云和晴空像元的AVIRIS高光谱数据模拟出不同传感器的云和晴空像元数据,据此生成它们的云检测阈值。这种方法具有自动化程度高、云检测效果好等优点。但是在原数据库的构建中,由于AVIRIS数据集中雪像元较少,导致云和雪的识别存在较大的误差。为了解决该问题,从待测数据中提取大量不同类型的雪像元,加入模拟得到的多光谱数据集中,生成新的云检测阈值,阈值生成中使用了单波段反射率、多波段反射率组合参数等作为云检测算法的输入参数。使用2013—2017年青藏高原地区多景典型Landsat8陆地成像仪(Operational Land Imager,OLI)影像进行实验验证。结果显示,在积雪覆盖的区域平均云像元识别正确率达到93.11%,对包括雪在内的晴空像元的漏判率降低到4.35%,表明此方法能有效避免雪的影响,实现高精度的云检测。
基金the National Key Bas ic Research&Deve lopment Program(Grant No.G2000018604)the National High TechnologyResearch and Development Program of China(Grant No.2002AA133060).
文摘The biosphere of the Earth is essential to human survival and development.The services of ecosystems are critical to the functioning of the Earth’s life-support system.They contribute to human welfare both directly and indirectly.Ecological capital refers to the sum of the direct biological resources value and the indirect ecosystem services value.It is necessary to estimate the ecological capital in order to bring it to the society and market economic system,and draw the social attention to ecological environment constructions.An estimation model for eco-logical capital based on remote sensing is presented in this paper.The parameters in the model are quantitatively measured using NOAA/AVHRR and other ancillary data,including the land cover types,the vegetation coverage,and the vegetation net primary productivity(NPP)of ter-restrial ecosystem.Based on the economic parameters in previously published studies and a few original calculations,the annual ecological capital of the entire terrestrial ecosystem of China is quantitatively estimated at 6.44 trillion(10^(12))yuan(RMB),and the spatial distribution of the eco-logical capital is also analyzed.Traditional ecological methods to ecological capital measurement are based on homogeneous plot scales,and the regional scaling is a key problem in their appli-cations.As the proposed remote sensing approach,it provides a new method to ecological capital measurement completely based on observation data.It can not only overcome the re-gional scaling problem easily,but also allows the ecological capital to be estimated objectively and spatial-explicitly.