International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
of no less than 3.
The weights in Eq. 1 can be approximated by dividing the sum
of values at that row by the total sum (i.e., the shaded cell in
Table 3). Saaty (1980) determined the weights using the
analytic hierarchy process, which makes a series of pair-wise
comparisons to determine the relative importance and ensures
consistency between all the factors in a multi-criteria evaluation.
In Table 3, a pair-wise comparison matrix is constructed, where
each factor is compared with the other factor, relative to its
importance, on a scale from 1 to 9. The empirical values about
the comparative importance between every two factors are
shown in Table 3. The weights are obtained by scaling the
principal eigenvector of the matrix, that is, (0.13, 0.23, 0.05,
0.59). For example, the mountain summits sited at the
northwest corner are about 1700 m from the HIS and protrude
the C/OHATS (see Figure 10b). Their risk index is equal to
2.09 as calculated using Eq. | with weights substituted.
Risk Row
R, R; Ry R4 Weight
factors sum
Rı | H2 3 1/5 4,70 0.14
R2 2 1 5 1/3 8.33 0.26
R; 3 1/3 1 1/9 1.65 0.05
R4 5 3 9 | 18.00 0.35
Sum 32.68 1.00
Table 3. Comparison of the relative importance of factors
Saaty (1980) calculates a consistency ratio (CR) to check the
probability that the ratings are randomly generated. The CR is
defined by Eq. 2, where X4, is the principal eigenvalue of the
matrix; n is the number of factors. A matrix with the CR value
greater than 0.1 should be re-evaluated, and the process is
repeated until the CR is less than this threshold. The CR is
0.0123 for the matrix in Table 3.
CR = Paras E n) / (n E 1) (2)
A risk level for each obstruction is assessed, and a part of the
risk-rating map is shown in Figure 11. The high-risk
obstructions pose a severe threat to aircrafts and should be
removed to conform to the AID. The median-risk obstructions
may be kept, but should be inspected periodically.
Figure 11. Obstructions risk-rating map
124
S. CONCLUDING REMARKS
By combining lidar data processing techniques with
photogrammetric mapping services, new toolsets help airfield
monitors solve problems and make decisions. In this paper, we
present an approach for identifying airfield obstructions
according to the new safety requirements in the Airfield
Initiative Document published by NIMA. The obstructions
include all kinds of physical features, and airport manager can
directly select all the dangerous obstructions from the
risk-rating mapping results. Next, we will setup an automatic
model on the OIS creation and object extraction. This will meet
the extremely urgent requirements of obstruction identification
from 7,200 airports in the US and an undetermined number of
airports, worldwide.
6. REFERENCES
Burrough, P, McDonnell, R.A, 1998. Principles of
Geographical Information Systems. Oxford University Press.
Chen, L., Lee, L., 1992. Progressive Generation of Control
Frameworks for Image Registration. Photogrammetric
Engineering & Remote Sensing, 58(12): 1321-1328.
Hu, Y., Tao, V., 2004a. Lidar Expert: a toolkit for information
extraction from airborne lidar data. Proc. of the ASPRS Annual
Conference, 23-28 May, Denver, 9 p.
Hu, Y., Tao, V., 2004b. Hierarchical recovery of digital terrain
models from single and multiple returns lidar data. Proc. of the
ASPRS Annual Conference, 23-28 May, Denver, 12 p.
Jensen, J.R., 1996. Introductory Image Processing: A Remote
Sensing Perspective. Prentice Hall, 156 p.
Lo, C.P., Yeung, AK.W., 2002. Concepts and Techniques of
Geographic Information System. Prentice Hall, 152 p.
Michael N.D., 2000. Fundamentals of Geographic Information
Systems (2nd Edition). John Wiley & Sons, Inc.
Morain, S.A., 2001. Remote Sensing for Transportation Safety,
Hazards, and Disaster Assessment. Urban Geoinformatics,
16-19 October, Wuhan, 6 p.
National Imagery Mapping Agency (NIMA), 2001. Airfield
Initiative Document, 50 p. URL:
http//www2.nima.mil/products/rbai/AIDOCwww.zip.
Saaty, T.L., 1980. The Analytical Hierarchy Process: Planning,
Priority Setting, Resource Allocation. McGraw-Hill
International Book Co., New York, 287 p.
TRB, 2002. Airfield inactive remote sensing technologies
evaluation project. Remote Sensing Conference, December
2002, 8 p.
Acknowledgements:
The authors would appreciate great assistance from Dr. Richard
Watson, Earth Data Analysis Center (EDAC), University of
New Mexico, for providing Santa Barbara Airport datasets. The
project is partially supported by US DOT/DASH consortium.
KEY WO
ABSTRA!
In this woi
being usec
software f
et al (199
distributio
scale fact
However €
factors, th:
were, ther
compariso
Brazil pos
control or
motivate 1
detection,
combat to
the contro
Initially c
monitorin
Automatic
years sate
the detect;
(1978) de
theoretical
images. H
temporary
operationa
With the
aperture re
technolog:
the most :
fact made
processing
extration €
to the inhe
factor tha
whose red
cause the
demonstra
images fo
that the de
of factors,
the ship, r
of the rade
to the dire
