【摘要】 基于透视原理、地面试验中对于较高目标的观测存在着一定的偏差。这种偏差随传感器高度、观测角度、视场角大小、观测位置等多个因素改变。由于垄行作物空间结构和温度分布的复杂性 ,在采用较大视场角测量方向亮温的地面实验中 ,将不可避免地存在着误差。采用一个简化的三分量二维结构模型对这种误差进行初步的分析与估算。亮温三分量分别为植被、被阳光照到的亮土和植被阴影下的暗土。作物的结构简化为剖面为矩形的无限长平行体。通过对这三个分量在传感器视场中面积权重的计算来模拟目标结构、传感器高度、位置、视场角大小、观测角度等因素对测量结果产生的影响。模拟结果表明 ,在垂直观测中 ,视场中的植被权重往往被高估 ,偏差随传感器高度的降低急剧增加。在倾斜观测中 ,由于一种互补效应的产生 ,偏差被限制在一个较低的范围内。经过分析 ,减小误差的最有效办法是提高传感器高度。最后 ,实验数据与模拟结果进行了比较。恰当地选取模型输入 ,两种数据能非常好的吻合。更多还原

【Abstract】 Composite scene of row crops induced an unavoidable error in ground observations due to the use of wide field of view (FOV) in the measurements. The measurements vary with sample size and position, detector height and view direction, and bias due to project principle, which is called FOV effect. This study focused on the estimation of FOV effect on the measurements of maize canopy directional brightness temperature (DBT) using a computational geometric 2D model. The model was developed to simulate the fractional variations of canopy brightness temperature components. In this research, the maize canopy was classified into three brightness temperature components: sunlit soil, shaded soil and vegetation, each component has a unique brightness temperature value. The simulation results revealed that the errors caused by wide FOV have complex features due to canopy geometry and measurement geometry. Generally, vegetation fraction is always over counted in the nadir, errors increase dramatically with the decrease of detector height as well as the enlargement of sample size, and the deviation of the error corresponding to sample position is small. In oblique view, the errors are limited to a low level due to the compensation effect. The best approach to reduce this kind of error is to set the detector to a higher altitude as the model suggested.更多还原