Professional Documents
Culture Documents
A Knowledge-Based Method For Road Damage Detection Using High-Resolution Remote Sensing Image
A Knowledge-Based Method For Road Damage Detection Using High-Resolution Remote Sensing Image
3565
pixels). l represents the searching line which moves along 3. RESULTS
the road centerline and the length is D ( D wroad ). The
procedure for extracting the post-disaster road segments is 3.1. Road Damage Detection
as follows:
(1) l is bisected vertically by the road centerline. Starting In this paper, the study area is located in the Wenchuan
from p1 , the road objects intersected by l at p1 are County, Sichuan Province, China. The experimental data is a
examined by the hypothesis model. panchromatic remote sensing image which was acquired by
(2) If they pass the hypothesis model, they are classified WorldView-1 on 15 May 2008. The detection of damaged
as hypothetic road segments and vice versa. road is based on post-disaster road recognition. The process
(3) If the hypothetic road segments are identified, they are mainly includes road centerline extraction combining with
to be verified by the verification model and the false road road seed points, post-disaster road extraction based on
objects will be removed, otherwise the next pixel p2 is knowledge and damage detection using spatial analysis.
analyzed. The road centerline is extracted firstly. The result of
(4) After traversing all the pixels in P, the extracted extracted road centerline (red line) is shown in Figure 3
results will be post-processed by the closing operation of where road seed points are shown as yellow crosses. Then,
mathematical morphology. In this way, the post-disaster the post-disaster road segments are extracted based on the
road segments are extracted. hypothesis model and the verification mode (Figure 4).
Y Finally, the buffer with the radius of wroad / 2 is generated
P = {p1 ,p 2 ,...,pn } by the road centerline and the erase analysis is used to detect
Post-disaster
road
the damaged regions of the road. The detection results of
damaged roads are shown in Figure 5.
D
yp
Damaged
wroad l road
Road
centerline
o xp
X
Erase
3566
3.2. Accuracy Evaluation great value for damaged road detection in disaster areas
especially where pre-disaster data are hard to obtain.
Within the study area, the damaged road segments were also Considering the influence of mixed pixels, the research on
manually delineated. The width, length and area of the target enhancing algorithms to increase the contrast between
manually delineated results were calculated and used as the road and other land markers worth a further exploration.
references to assess the accuracy of the automatic damaged
detection. Table 1 presents the accuracy of road damage 5. ACKNOWLEDGEMENTS
detection.
This study was funded by the National High Technology
Table 1. Accuracy evaluation of road damage detection Research and Development Program ("863"Program) of
China (Grant Number 2012AA121305)
Real Detected Correctly
Indicators damaged damaged detected PA(%) UA(%) 6. REFERENCES
road road road
Width 8m 7.5m 7.5m 93.75 100.00 [1] Lan R., Feng B., Wang Z., “Study on the fast assessment of
traffic capacity of highway bridges after strong earthquakes,”
Length 353m 392m 323m 91.50 82.40 World Earthquake Engineering, 2009, 25(2): 81-87.
Area 2661 m2 2809 m2 2322 m2 87.26 82.66 [2] Trinder J., Wang Y., “Knowledge-Based Road Interpretation
in Aerial Images [J],” In International Archives of
PA: producer’s accuracy; UA: user’s accuracy Photogrammetry and Remote Sensing, 1998, 32, 635-640.
It can be seen from Table 2 that the detection results are [3] Senthilnath J., Rajeshwari M., Omkar S., “Automatic road
fairly accurate, which justifies the effectiveness of this extraction using high resolution satellite image based on
method. The PA and UA can reach, on average, to 90% and texture progressive analysis and normalized cut method,” J.
85% respectively. The newly proposed method does not Indian Soc. Remote Sens., 2009, 37, 351-361.
require the pre-disaster data and achieves comparable results
[4] Singh P., Garg R., “Automatic Road Extraction from High
with existing methods, in which high accuracies were
Resolution Satellite Image Using Adaptive Global
acquired from the pre-disaster data and post-disaster image Thresholding and Morphological Operations,” Indian Society
[8, 9]. However, some errors still exist and may be attributed of Remote Sensing, 2013, 03, 631-640.
to two reasons. One is the influence of mixed pixels. Mixed
pixels blur the road boundary which further affects the [5] Samadzadegan F., Zarrinpanjeh N., “Earthquake destruction
accuracy of damage detection. The other is the influence of assessment of urban roads network using satellite imagery
trees, vehicles and other landmarks which present different and fuzzy inference systems,” Proc. The International
spectra as road and may falsely be classified as the damaged Archives of the Photogrammetry, Remote Sensing and
region. Spatial Information Sciences. 2008, pp. 409-414,
3567