We first present a brief overview of the classical description of the scattering problem in radiative transfer theory followed by our perspective

2017 
Electro- Electro-optical terrain reflectance modeling is usually taken in the specific sense to mean the calculation of the bidirectional reflectance distribution function for a terres­ trial medium based on the biogeophy s ica 1 attributes of that medium. Typical terrestrial media envisioned consist of water, snow, soil, and vegetation canopies or composites of such categories. For remote sensing applications it is often also necessary to consider the subsequent propagation through the intervening atmospheric medium. Physical scales may range from a few centimeters to several meters although the relative parameter of impor­ tance is the ratio of the resolution cell size of the measurement device, i.e. sensor, to the individual elements considered in the sensor f ie Id-of-view . These elements are abstractions of objects in a scene and are characterized by properties or distributions. Elements may also be nested; that is, elements, along with their properties or distribu­ tions may be used to derive the properties or distributions of larger elements that are aggregated from the smaller ones. The definition of elements or classes of elements is somewhat arbitrary depending upon a modeler's abstraction, but in the examples presented here we will consider that the sensor f ie 1 d-of-view does not resolve the individual ele­ ments. For example, if the classes of interest are vegetation canopies, it is assumed that individual leaves are not resolved; conversely, if the individual leaf is sensed, its micros tructure is not. Similarly, for a soil surface neither the micro s true ture , defined with respect to the sensor f ie Id-of-view, nor individual soil grains would be resolved. For simplicity, we will also assume the temporal scale to be such that both the incident and exitant flux distributions may be taken to be stationary over the instrument response time . A review of electro-optical terrain reflectance models developed during the past decade and our current state-of-the-art to predict electromagnetic scattering of terrestrial media at optical wavelengths can be found in Smith and Bunnik . The relevance of such modeling to remote sensing applications is discussed in Smith , Strahler, et al . , and Murphy and Deering . Similarly, the set of questions and accompanying suite of associated issues dealing with model validations, comparisons, sensitivities, and experimental requirements is discussed in Smith , e.g. Duggin , and Slater . The focus of this paper is to consider the use or combination of models across media domains and to highlight the advantages that accrue to investigators if a more formal approach is taken to the terrain reflectance problem. We first present a brief overview of the classical description of the scattering problem in radiative transfer theory followed by our perspective of the problem with special refer­ ence to terrain media. Finally, two examples are given to illustrate our perspective that it is as equally important for modelers to report their formulations for their biogeophysi- cal description of media attributes in standard radiative transfer terms as it is to actu­ ally calculate the media bidirectional reflectance distribution functions.
    • Correction
    • Cite
    • Save
    • Machine Reading By IdeaReader
    17
    References
    0
    Citations
    NaN
    KQI
    []
    Baidu
    map