|
STATEMENT
OF THE PROBLEM
Monitoring of existing construction sites within the limits of the
City of Columbia is a requirement of the city government for:
1) Control of construction site permits issued by the City,
2) Location of construction sites where the natural ground cover
or pre-existing structures were removed and bare soil is exposed
and becomes a potential source of sediments that will lower water
quality in local watersheds,
3) Monitoring of construction progress at those sites
PRESENT SITUATION
The City of Columbia management would like to have a tool that would
complemented its present monitoring of construction sites to provide
accurate updates on the progress of building activities and would
effectively show the amount of bare ground exposed to erosion.
POTENTIAL BENEFITS
Monitoring of construction sites in the City of Columbia will bring
benefits to city management in reference to urban development and
compliance with more strict EPA regulation of water quality in urban
watersheds.
TECHNICAL APPROACH TAKEN
The project was carried out using
as data source the four IKONOS-2 multispectral bands (0.45 –
0.53 mm, 0.52 – 0.61mm, 0.64 – 0.72 mm, 0.77 –
0.88 mm) with a 4 m spatial resolution. The IKONOS-2 data was obtained
on 30 April 2000 at 10:48 AM. Sun azimuth was 138.74 degrees and
sun elevation was 60.74 degrees. The nominal data collection azimuth
was 352.42 degrees and the nominal data collection elevation was
62.89 degrees
Figure 1. Color display of IKONOS
multispectral image with 4-meter spatial resolution (RGB = 3,2,1)
sharpened with IKONOS 1-meter panchromatic image showing the outline
of City of Columbia
A Supervised Maximum Likelihood Classifier was
used to map active construction sites where bare ground was exposed.
Training sites were defined for two types of bare ground (associated
with different exposed soil horizons and poorly developed grass)
and eight other major land cover classes within the project area.
Other training classes included two types of grass (well and fairly
well developed grassland), two types of water surface (separated
by presence or absence of sediments in suspension), two classes
of concrete surfaces (road and parking areas), asphalt surfaces
and woodlands. Single-pixel training sets (Gong and Howarth, 1990)
with approximately 100 pixels in each were established for each
class.
 
A B
Figure 2. Some of the training sites
used for supervised classification. A. Area where some of the bare
soil types #1 and #2 and impervious surfaces (concrete) were selected.
B. Area with some of the impervious surfaces (asphalt) training pixels
Figure 3. Color IKONOS multispectral
image (RGB = 3, 2,1; 4-m spatial resolution) of an area in southern
Columbia showing several construction sites
Figure 4. Color IKONOS multispectral
image (RGB = 4, 3, 2; 4-m spatial resolution) of the same area in
Figure 2 in southern Columbia showing several construction sites.
The classification results were combined
into three major classes where two bare soil classes could be discriminated.
One of them could be associated with urban and rural areas where the
ground surface was in part covered by vegetation consisting of a mixture
of poorly developed grass and stressed or dead crops and the other
was associated with construction sites alone. The third class contained
all other initially mapped classes.
Figure 5:
Result of supervised maximum likelihood classification with six
classes. Areas with active construction sites shown by arrows.
Accuracy assessment was based on a priori selected
known pure pixels selected in the image. The results indicated that
an accuracy level of 94% had been reached for the entire area. For
display purposes the final map has been filtered to show only those
areas with more than one acre.
Table 1 . Confusion or error
matrix for the supervised classification of construction sites indicated
by bare soil exposures in the City of Columbia.
At this time the City is only interested in tracking
down those construction sites that have at least one-acre area.
A display was produced (Fig. 6) showing only those areas with exposed
soil with area equal or larger than 1-acre. This approach eliminated
from the classification results those small areas that had exposed
soil but do not represent construction sites.
Figure 6. Location of construction
in the City of Columbia and surrounding areas on April30, 2000 based
on the supervised classification of IKONOS multispectral data (4m
spatial resolution).
ANTICIPATED
IMPLEMENTATION PROBLEMS IN URBAN GOVERNMENT
The government agency interested in implementing this approach to
be used on a routine basis should have the appropriate image processing
hardware and software in place. Once the training and testing sites
for supervised classification have been defined for an appropriate
range of soil types in the area, it will be possible to produce
land use maps showing active construction areas from any other IKONOS
multispectral data sets without technical supervision.
ADDITIONAL WORK
The information contained in IKONOS pan band (0.45 – 0.90
mm) still has to be evaluated in combination with the multispectral
bands
PROJECT PARTICIPANTS
Dr. Aderbal C. Corrêa and Mr. Janggam Adhityawarma of the
Center for Environmental Technology (CENTECH), Department of Civil
and Environmental Engineering, carried out the project for ICREST.
|
