热红外遥感图像模拟中异质异构地表像元辐射计算方法研究

项目名称： 热红外遥感图像模拟中异质异构地表像元辐射计算方法研究

项目编号： No.41501402

项目类型： 青年科学基金项目

立项/批准年度： 2016

项目学科： 天文学、地球科学

项目作者： 赵利民

作者单位： 中国科学院遥感与数字地球研究所

项目金额： 20万元

中文摘要： 热红外遥感图像模拟是提高红外载荷应用效能的重要先导性研究，难点在于像元尺度异质异构地表有效辐射强度的精确计算。项目基于能量平衡、辐射换热和光谱混合分析理论，构建像元尺度热辐射计算方法，提高复杂地表热红外遥感图像模拟的精度与可靠性。以城市下垫面为研究对象，分析地表覆盖模式，通过可见-近红外多光谱图像解混以提取组分异质信息，利用结构参数计算异构组分间辐射换热可视因子。继之，由地表能量平衡方程分析地表热辐射变化特征，并建立基于可视因子的热红外多次散射估算方法，实现组分有效辐射精确计算。在此基础上，通过组分有效辐射的线性组合，建立像元尺度地表辐射计算模型，藉此模拟不同时相地表热红外辐射图像，并利用热红外航空遥感数据分析评价模拟精度。通过项目研究，将厘清异质异构地表热辐射变化与相互作用特征，建立物理原理可靠、模拟精度高、计算简便的复杂地表像元辐射计算方法，为热红外遥感图像模拟提供准确有效的技术支持。

中文关键词： 热红外遥感；城市热环境；遥感数据模拟；计算机仿真模拟

英文摘要： Thermal infrared (TIR) remote sensing imagery simulation is a significant pre-guiding procedure required by the future TIR sensors’ prototyping and performance evaluation. Precisely estimation of heterogeneous land surface thermal radiance (LSTR) is a key factor that determines the TIR image simulation. This proposal, based on the combination of techniques of the land surface energy balance (SEB), the heat transfer and the spectral mixing analysis, the origination and the mixing mechanism of LSTR at pixel scale are analyzed. Firstly, the feature distributions of land thermophysical property (LTPD) are retrieved from the visible and near infrared (VNIR) multispectral remote sensing imageries, by extracting the exponent of endmembers of typical materials and calculating the coverage ratio of endmembers in each pixel. The shape factor of heterogeneous components is calculated using statistical parameters of land buildings. Secondly, the dynamic variations of LSTR is simulated by establishing a surface temperature model using the SEB equation and the LTPD. The multiple scattering radiance within heterogeneous non-isothermal components is calculated using the shape factor. The surface radiance at pixel scale is subsequently simulated through linear combination of the effective radiance of each component. Finally, the proposed approach is validated and evaluated using the acquired airborne TIR remote sensing imageries. This research is expected to build an effective approach for the LSTR simulation over urban surfaces, and to further improve the accuracy and reliability of TIR image simulation. Of greater importance, the results of this research will provide the sensor designers and the end-users with future images in an operational context.

英文关键词： thermal infrared remote sensing;urban thermal environment;remote sensing data simulation;computer simulation