Numerical recipes in C (2nd ed.): the art of scientific computing
Numerical recipes in C (2nd ed.): the art of scientific computing
| Hi-index | 0.00 |
Most land-cover types are anisotropic; that is, they do not reflect solar radiation uniformly in all directions. Characterizing the bidirectional reflectance distribution function of the earth's surface is critical in understanding surface anisotropy. Although several methods exist for retrieving the BRDF of various land-cover types, most of them have been applied over small data sets collected either on the ground or from aircraft at limited spatial and temporal scales. Satellite-based sensors can provide the data necessary for large-scale studies of BRDF. However, most sensor systems (for example, AVHRR-advanced very-high-resolution radiometer) take directional spectral measurements. Land-surface anisotropy causes variations in surface reflectances when measured under different illumination and view angles. So, these measurements (for example, reflectance) are valid only for a particular sensor- illumination geometry. In this article, we'll describe three models for deriving global BRDF that compensate for the directional limitations of satellite sensors. We've devised algorithms that implement these models on a high-performance computing system, using an efficient method to handle the large data set involved. Our implementations optimize I/O access time and efficiently balance computations across the nodes.