An overview of Spatial-Spectral Unmixing in Satellite Imagery

Merging information acquired by sensor systems providing image data with different resolutions may in many cases be a valuable tool to the analyst. The multiresolution image fusion techniques merge the spatial information from a high-resolution image with the radiometric information from a low-resolution image. On the other hand, one of the physical quantities that could be acquired from remote sensing data is surface reflectance in different regions of electromagnetic reflected spectra. There are several important questions related to the spectral properties of urban materials e.g. how do those materials differ in their spectral response? what are the most suitable spectral bands for mapping urban land cover? what are the spectral limitations of current high spatial resolution remote sensing systems in terms of mapping urban land cover?, how different data fusion could be used in improving classification accuracy and urban spectral reflectance modeling? …. Almost all the urban pixels that imaged by low/medium spatial resolution sensor systems represent a composite radiance field emanating from several distinct features with different reflectance's within the sensor’s field of view. Some factors like existence of mixed pixels, non-Lambertian behavior of urban and material aging, complicates urban environments. A general approach for describing land covers is using classification methods and unmixing models. One of the useful unmixing models is spectral mixture analysis in which the mixed pixel reflectance is considered as a linear summation of the reflectance and fractions corresponding to land cover types within the pixel. The aim of this study was to analyze the methods of determining spectral reflectance of important land cover materials through remote sensing data and spatial unmixing model taking in to account the variability of the land covers by using high spatial resolution and hyperspectral imagery and applications. Results show an innovation in this field.

Keywords:

Spatial , Spectral , Unmixing , Mixture , Multiresolution