Leave Area Index (LAI) is one of the most basic parameters to describe the geometric structure of plant canopies. It is also important input data for climatic model and interaction model between Earth surface and atmo...Leave Area Index (LAI) is one of the most basic parameters to describe the geometric structure of plant canopies. It is also important input data for climatic model and interaction model between Earth surface and atmosphere, and some other things. The spatial scaling of retrieved LAI has been widely studied in recent years. Based on the new canopy reflectance model, the mechanism of the scaling effect of con- tinuous canopy Leaf Area Index is studied, and the scaling transform formula among different scales is found. Both the numerical simulation and the field validation show that the scale transform formula is reliable.展开更多
基于图像的三维建模中图像的弱纹理区域因其颜色单一和局部反光现象,使得该区域内难以有效的检测和匹配特征点以及进行鲁棒的三维点云扩展,容易产生空洞现象和大量的噪声三维点,影响三维建模的精度和完整性。运用松弛变量约束对误匹配...基于图像的三维建模中图像的弱纹理区域因其颜色单一和局部反光现象,使得该区域内难以有效的检测和匹配特征点以及进行鲁棒的三维点云扩展,容易产生空洞现象和大量的噪声三维点,影响三维建模的精度和完整性。运用松弛变量约束对误匹配特征点滤波并优化相机矩阵,在稠密点云扩展阶段运用张量投票原理,滤波点云扩展中与周围三维点法向不一致的噪声点,运用多尺度离散-连续深度图法重构三维模型。实验结果表明:提出的方法与PMVS(patch based multi view stereo),MVE(multi view environment),Mesh Recon等重构方法相比,具有更好的建模精度和完整度。展开更多
基金Supported by National Basic Research Program of China (Grant No. 2007CB714402)National Natural Science Foundation of China (Grant Nos. 40401036, 40734025 and 40401036)
文摘Leave Area Index (LAI) is one of the most basic parameters to describe the geometric structure of plant canopies. It is also important input data for climatic model and interaction model between Earth surface and atmosphere, and some other things. The spatial scaling of retrieved LAI has been widely studied in recent years. Based on the new canopy reflectance model, the mechanism of the scaling effect of con- tinuous canopy Leaf Area Index is studied, and the scaling transform formula among different scales is found. Both the numerical simulation and the field validation show that the scale transform formula is reliable.
文摘基于图像的三维建模中图像的弱纹理区域因其颜色单一和局部反光现象,使得该区域内难以有效的检测和匹配特征点以及进行鲁棒的三维点云扩展,容易产生空洞现象和大量的噪声三维点,影响三维建模的精度和完整性。运用松弛变量约束对误匹配特征点滤波并优化相机矩阵,在稠密点云扩展阶段运用张量投票原理,滤波点云扩展中与周围三维点法向不一致的噪声点,运用多尺度离散-连续深度图法重构三维模型。实验结果表明:提出的方法与PMVS(patch based multi view stereo),MVE(multi view environment),Mesh Recon等重构方法相比,具有更好的建模精度和完整度。