Image Processing Reference
In-Depth Information
transportation analysis. Demographic, population, and socio-economic estimates
may be related to morphological composition. Health conditions relate to air
quality patterns which are influenced by V-I-S cover composition. Finally, the
payoff for remote sensing applied to urban/peri-urban areas leads to a refinement
of scientific problems and provides a basis for environmental management and
urban planning.
6.3
The Influence of Resolution on V-I-S Analysis
Previous chapters in this topic have treated resolution - spectral, spatial, and
temporal - in general application to urban remote sensing. This section establishes
some principles of resolution that influence the effective-
ness of the V-I-S model, and sets a stage for the following
sections summarizing various applications of the model.
Clearly, adequate spectral and spatial resolutions are vital
to detection, mapping, and quantification of vegetation,
impervious surface, and exposed soil, as well as water. The
need for improvement in resolution increases whenever the
objective calls for dividing each of the V-I-S groups into
sub-categories. As resolution increases, the opportunity for
further subdivision improves.
Herold et al. (
2003
) identify 14 spectral bands most suitable for separability of
urban land cover classes (also see Chapter 4). Those bands are distributed across
the visible (VIS: 400-700 nm) and near infrared (NIR: 700-2,500 nm) spectral
range. That is paramount because healthy green photosynthesizing vegetation is dis-
tinctly identifiable in the 700-1,000 nm range, and environmental moisture is par-
ticularly deflectable in the short wave infrared (SWIR) range from 1,000 to 2,500
nm. To have spectral sensors strategically placed in all three of those zones is vital
to detection of vegetation, impervious surface, and soil. Landsat thematic mapper
(TM) is designed to take advantage of this spectral range with its six reflective
bands: three in the visible range (blue 450-520 nm, green 520-600 nm, and red
630-690 nm); one in the NIR (760-900 nm) region and two in the SWIR region
(1,550-1,750 and 2,080-2,350 nm). These are, respectively TM bands 1, 2, 3, 4, 5,
and 7 (see Fig. 4.6). TM band 7 was added for its particular sensitivity to certain
minerals some of which are significant in the urban environment for detecting vari-
ous soils and concrete, for example. This band was added after the thermal band
(10,400-12,500 nm), band 6. The reflective bands are set at a nominal 30 m spatial reso-
lution. The thermal band at 120 m has limited utility in urban analysis. Furthermore,
at the time of Landsat over flight, in mid-morning, temperatures are crossing over
from morning cool to mid-day heating.
The earliest earth remote sensing satellite system, Landsat 1, with its multispectral
scanner (MSS) data, 1972, had only four spectral bands, in the general range of TM
bands 2, 3, and 4. The minimal 80m IFOV (instantaneous field of view) further
effectiveness of
the V-I-S model is
directly
dependent upon
resolution -
spatial, spectral,
temporal, and
radiometric
