Image Processing Reference
In-Depth Information
only marginally resolved the complex spectral characteristics of urban
environments, especially for built surface types. The use of three-dimensional
information from LIDAR can help to overcome some of these spectral
limitations. Important for mapping urban land cover from remote sensing are
data with high spatial resolution. However, hyperspectral data provide the
most stable spectral mapping accuracies for coarser spatial resolution data.
LEARNING ACTIVITIES
•
Review the main terms and laws related to electromagnetic radiation:
Plancks law, Boltzmanns equation, Wiens law
Exitance, irradiance, radiance, reflectance and their units
Atmosphere: transmission and absorption processes, scattering (Rayleigh and
Mie), atmospheric windows
Surface interactions: reflection and transmission, absorption (electronic and
vibrational processes), Snells law, Fresnels equation
•
Download and explore some example spectra (vegetation/soil/build surfaces)
from existing spectral libraries:
Santa Barbara urban spectral library: www.geogr.uni-jena.de/~c5hema/spec/
ieee_fig2.zip
USGS spectroscopy lab: http://speclab.cr.usgs.gov/
JPL/ASTER spectral library including man made materials: http://speclib.jpl.
nasa.gov/
•
IKONOS and Landsat (E)TM are commonly used for mapping urban areas.
Recapitulate their capabilities and limitations for urban remote sensing for spe-
cific applications. In this context discuss multispectral versus hyperspectral
remote sensing of urban areas.
Learn more about concepts and applications of image analysis techniques for
•
hyperspectral data. A suitable software system to do so is ENVI:
Spectral angular mapper (SAM)
Spectral mixture analysis (SMA)
Matched filter analysis (MFA)
•
Review major fields of application in detailed urban mapping:
Impervious surfaces for flood and urban water quality management
Infrastructure and land use for urban planning and management
