Access to this full-text is provided by MDPI.
Content available from Remote Sensing
This content is subject to copyright.
Citation: Liu, R.; Zhu, W.; Yang, X.
Screening Image Features of
Collapsed Buildings for Operational
and Rapid Remote Sensing
Identification. Remote Sens. 2023,15,
5747. https://doi.org/10.3390/
rs15245747
Academic Editor: Gemine Vivone
Received: 28 October 2023
Revised: 12 December 2023
Accepted: 13 December 2023
Published: 15 December 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
remote sensing
Article
Screening Image Features of Collapsed Buildings for
Operational and Rapid Remote Sensing Identification
Ruoyang Liu 1,2 , Wenquan Zhu 1,2,* and Xinyi Yang 1,2
1State Key Laboratory of Remote Sensing Science, Beijing Normal University, Beijing 100875, China;
liury@mail.bnu.edu.cn (R.L.); 202221051111@mail.bnu.edu.cn (X.Y.)
2Beijing Engineering Research Center for Global Land Remote Sensing Products, Faculty of Geographical
Science, Beijing Normal University, Beijing 100875, China
*Correspondence: zhuwq75@bnu.edu.cn; Tel.: +86-10-58807053
Abstract: The accurate detection of collapsed buildings is of great significance for post-disaster
rescue and reconstruction. High-resolution optical images are important data sources for identifying
collapsed buildings, and the identification accuracy mainly depends on the features extracted from
the images. However, existing research lacks a comprehensive screening and general evaluation of
the ability of remote sensing features to detect collapsed buildings, and there is still a considerable
gap in the operational process of rapid identification of collapsed buildings in remote sensing. Based
on 2630 pairs of building samples distributed in 6 regions worldwide, this study evaluated the
ability of 25 remote sensing features (including spectral and spatial features) to detect collapsed
buildings and select the most capable ones. Then, we test the application effect of selected features in
identifying collapsed buildings on large-scale remote sensing images. Based on the two experiments
above, an operational process for rapid identification of collapsed buildings was suggested. The
result shows that Homogeneity, Energy, Local Entropy, Local Standard Deviation, and Gradient
can effectively and stably distinguish collapsed buildings from non-collapsed buildings (Jeffries-
Matusita distances are greater than 1.59 and Transformed Divergences are greater than 1.60) and have
high recognition accuracy for collapsed buildings on large-scale remote sensing images (
F1-scores
are 0.71–0.94). In addition, Contrast, Local Coefficient of Variation, Edge Density, and Global
Entropy can also distinguish collapsed buildings from non-collapsed buildings at a normal level
(Jeffries-Matusita distances are 1.14–1.28, and Transformed Divergences are 1.24–1.48), while Gradient
Orientation Entropy, Fractal Dimension, Local Binary Patterns, Edge, Local Mean, Correlation,
Gradient Orientation Standard Deviation, Global Coefficient of Variation, Gabor feature, Local
Moran’I, and six spectral features have relatively weak abilities (Jeffries-Matusita distances are less
than 0.73, and Transformed Divergences are less than 1.07). The selected remote sensing features
can support rapid identification of potential collapsed building areas from post-disaster remote
sensing images.
Keywords: collapsed buildings; remote sensing features; rapid identification; operational processing
1. Introduction
Building collapse caused by disasters is an important damage to lives and properties.
Rapid identification of collapsed buildings is of great significance for post-disaster rescue [
1
]
and reconstruction [
2
]. In recent years, remote sensing has become an important method
of obtaining building damage information because of its advantages of non-contact, low
cost, wide field of view, and fast response [
3
]. Due to the urgency of post-disaster rescue, it
is necessary to make the process of remote sensing identification of collapsed buildings
operationalized so that we can rapidly identify collapsed buildings while maintaining high
recognition accuracy.
Remote sensing identification essentially involves the following processes: data collec-
tion
→
feature extraction
→
feature selection
→
target identification using an appropriate
Remote Sens. 2023,15, 5747. https://doi.org/10.3390/rs15245747 https://www.mdpi.com/journal/remotesensing
Remote Sens. 2023,15, 5747 2 of 18
classification method. The current methods for remote sensing identification mainly in-
clude traditional machine learning methods and deep learning methods, both of which
essentially follow the process mentioned above. The deep learning methods avoid manual
feature extraction and feature selection [
4
–
6
], but they require a large number of samples
for training [
7
], which are limited in the case of collapsed buildings [
8
]. Although methods
like data augmentation [
9
], transfer learning [
8
,
10
], or pseudo-labeled training sample pro-
duction [
11
] could alleviate this problem, they all rely on the samples from the study area.
Due to the small sample volume, it is easy to suffer from the uneven distribution of training
samples, which affects the training accuracy of the model [
8
]. Therefore, current deep learn-
ing methods are still unable to meet the operationalization needs of rapid identification of
collapsed buildings after disasters. In contrast, traditional machine learning methods do
not require as many samples, and the effectiveness of their identification largely depends
on the features used. The enhancement in either feature extraction or feature selection can
contribute to improving the accuracy of collapsed building recognition. However, this
study primarily focuses on feature selection.
There have been many features used for remote sensing recognition of collapsed
buildings in previous studies. For example, Ye et al. [
12
] used texture and local spatial
statistical features to identify collapsed buildings in the post-earthquake images of the 2015
Nepal earthquake; Zeng et al. [
13
] used grayscale mean and inverse different moment to
identify collapsed buildings in post-earthquake images of Haiti and Yushu; Ye et al. [
14
]
used local Moran’I, texture, and gradient to identify collapsed buildings in post-earthquake
images of the Yushu earthquake; Sumer et al. [
15
] used grayscale values and gradient to
identify collapsed buildings in post-earthquake images of the Kocaeli earthquake. The
features used for collapsed building identification in previous studies include spectral
features [
16
,
17
], Entropy, Coefficient of Variation, Local Mean, Local Standard Deviation,
Local Moran’I [
12
,
14
], Gradient, Gradient Orientation [
14
,
18
], Edge, Edge Density [
19
],
Gray Level Co-Occurrence Matrix (GLCM) [
19
,
20
], Local Binary Patterns (LBP), Fractal
feature [
21
], and Gabor feature [
21
], etc. These features can be summarized into spectral
features and spatial features (Figure 1), where spectral features include original bands and
their derived features, and spatial features include spatial statistical features, gradient and
edge features, and texture features. The research mentioned above has achieved good
recognition results for collapsed buildings in specific areas, but there has been no study
on how to select the optimal features. Instead, several features were directly used for
identification based on prior knowledge, making the individual roles of these features
unclear in the process. Furthermore, existing studies are all case studies conducted in
a certain area, so it is still unknown whether the features can be equally applicable in
other regions. In summary, although many features have been used for the recognition
of collapsed buildings, their recognition ability and transferability have not been widely
evaluated, and there is a lack of theoretical guidance on which features should be used for
practical identification work. Therefore, there is still a gap in the operational process of
rapid recognition of collapsed buildings in remote sensing.
This article is dedicated to screening out remote sensing features of collapsed buildings.
By using a large building sample set collected from different regions around the world,
this study evaluated the capability of 25 different features in detecting collapsed buildings
and explored the optimal parameters for the features with good performance to make the
best use of the features. For the features with good distinguishing ability for collapsed
buildings, we test their effectiveness on the application test image set.
Remote Sens. 2023,15, 5747 3 of 18
RemoteSens.2023,15,xFORPEERREVIEW3of18
Figure1.Featuresofopticalremotesensingimagesforidentifyingcollapsedbuildings.
Thisarticleisdedicatedtoscreeningoutremotesensingfeaturesofcollapsedbuild-
ings.Byusingalargebuildingsamplesetcollectedfromdifferentregionsaroundthe
world,thisstudyevaluatedthecapabilityof25differentfeaturesindetectingcollapsed
buildingsandexploredtheoptimalparametersforthefeatureswithgoodperformanceto
makethebestuseofthefeatures.Forthefeatureswithgooddistinguishingabilityfor
collapsedbuildings,wetesttheireffectivenessontheapplicationtestimageset.
2.DataandMethodology
2.1.Data
2.1.1.BuildingSampleSet
Thebuildingsamplesetofthisstudyincludes2630pairsofbuildingsamplesdistrib-
utedin6regionsworldwide.Eachpairofbuildingsamplesincludesthepre-andpost-
collapseimagesofthesamebuilding.Thesamplesareallvisible-lightimageswithaspa-
tialresolutionofabout0.5m.Eachsampleisapproximately100rows×50columnsof
pixelsinsize.ThesampleswereselectedfromthexBDdataset[22]andtheOpenData
Programdataset(hps://www.digitalglobe.com/ecosystem/open-data,accessedon3
March2023)ofMaxar.ThexBDdatasetcontainsover850,000pairsofpre-andpost-col-
lapsebuildingimagesaswellasboundaryvectordataforpre-collapsebuildings.Inthis
study,atotalof2508pairsofsampleswereselectedthroughvisualinterpretationfrom
thexBDdataset.TheOpenDataProgramdatasetprovidesremotesensingimagesbefore
andafterdisasterevents.Atotalof122pairsofpre-andpost-collapsebuildingsamples
wereselectedthroughvisualinterpretation,andthevectorboundariesofpre-collapse
Figure 1. Features of optical remote sensing images for identifying collapsed buildings.
2. Data and Methodology
2.1. Data
2.1.1. Building Sample Set
The building sample set of this study includes 2630 pairs of building samples dis-
tributed in 6 regions worldwide. Each pair of building samples includes the pre- and
post-collapse images of the same building. The samples are all visible-light images with a
spatial resolution of about 0.5 m. Each sample is approximately 100 rows
×
50 columns
of pixels in size. The samples were selected from the xBD dataset [
22
] and the Open Data
Program dataset (https://www.digitalglobe.com/ecosystem/open-data, accessed on 3
March 2023) of Maxar. The xBD dataset contains over 850,000 pairs of pre- and post-collapse
building images as well as boundary vector data for pre-collapse buildings. In this study,
a total of 2508 pairs of samples were selected through visual interpretation from the xBD
dataset. The Open Data Program dataset provides remote sensing images before and after
disaster events. A total of 122 pairs of pre- and post-collapse building samples were selected
through visual interpretation, and the vector boundaries of pre-collapse buildings were
manually outlined. Figure 2shows image examples of some samples, and the detailed
information of all samples is shown in Table 1.
Remote Sens. 2023,15, 5747 4 of 18
RemoteSens.2023,15,xFORPEERREVIEW4of18
buildingsweremanuallyoutlined.Figure2showsimageexamplesofsomesamples,and
thedetailedinformationofallsamplesisshowninTab l e 1.
Figure2.Imageexamplesofsomesamples.Note:Thesamplesarewithintheblueboundaries.The
pre-collapsesamplesareontheleft,andthecorrespondingpost-collapsesamplesareontheright.
Tab l e 1.Detailedinformationonallbuildingsamples.
ImageShooting
Location
ReasonsandTimingof
BuildingCollapseImageShootingTime
Numberof
Samplesfor
BuildingPairs
Bata
(EquatorialGuinea—Litoral,
1.87°N,9.77°E)
explosion
(7March2021)
imagebeforecollapse:7August2020
imageaftercollapse:9March202192
Beirut
(Lebanon—Beirut,
33.87°N,35.70°E)
explosion
(4August2020)
imagebeforecollapse:31July2020
imageaftercollapse:5August202030
Joplin
(USA—Missouri,
37.10°N,94.58°W)
tornado
(22March2011)
imagebeforecollapse:8August2009
imageaftercollapse:29May20111558
Moore
(USA—Oklahoma,
35.33°N,97.50°W)
tornado
(20May2013)
imagebeforecollapse:17February
2013
imageaftercollapse:22May2013
594
Tuscaloosa
(USA—Alabama,
33.2°N,87.51°W)
tornado
(27April2011)
imagebeforecollapse:21June2006
imageaftercollapse:19May2011277
Woolsey
(USA—California,
34.06°N,118.76°W)
wildfire
(8November2018)
imagebeforecollapse:23October
2018
imageaftercollapse:18November
2018
79
Figure 2. Image examples of some samples. Note: The samples are within the blue boundaries. The
pre-collapse samples are on the left, and the corresponding post-collapse samples are on the right.
Table 1. Detailed information on all building samples.
Image Shooting
Location
Reasons and Timing of
Building Collapse Image Shooting Time Number of Samples for
Building Pairs
Bata
(Equatorial Guinea—Litoral,
1.87◦N, 9.77◦E)
explosion
(7 March 2021)
image before collapse: 7 August 2020
image after collapse: 9 March 2021 92
Beirut
(Lebanon—Beirut,
33.87◦N, 35.70◦E)
explosion
(4 August 2020)
image before collapse: 31 July 2020
image after collapse: 5 August 2020 30
Joplin
(USA—Missouri,
37.10◦N, 94.58◦W)
tornado
(22 March 2011)
image before collapse: 8 August 2009
image after collapse: 29 May 2011 1558
Moore
(USA—Oklahoma,
35.33◦N, 97.50◦W)
tornado
(20 May 2013)
image before collapse: 17 February 2013
image after collapse: 22 May 2013 594
Tuscaloosa
(USA—Alabama,
33.2◦N, 87.51◦W)
tornado
(27 April 2011)
image before collapse: 21 June 2006
image after collapse: 19 May 2011 277
Woolsey
(USA—California,
34.06◦N, 118.76◦W)
wild fire
(8 November 2018)
image before collapse: 23 October 2018
image after collapse: 18 November 2018 79
2.1.2. Application Test Image Set
The application test image set consists of three large-scale remote sensing images,
namely post-disaster remote sensing images from Joplin (6408 rows
×
4734 columns),
Moore (13,560 rows
×
10,019 columns), and Tuscaloosa (8431 rows
×
5941 columns). These
large-scale images are also visible light images derived from the Open Data Program
dataset, with a spatial resolution of approximately 0.5 m. The application test image set is
used to test the practical application ability of selected features in this study.
Remote Sens. 2023,15, 5747 5 of 18
2.2. Methodology
Based on the building sample set, this study evaluated the ability of 25 features to
distinguish collapsed buildings from non-collapsed buildings by two indicators (Jeffries-
Matusita distance and Transformed Divergence) and explored the optimal parameters for
the features with good performance. For the features with good distinguishing ability for
collapsed buildings, we test their effectiveness on the application test image set.
2.2.1. Ability Evaluation of Features
The indicators used to evaluate the ability of features are Jeffries-Matusita distance
(J-M distance) and Transformed Divergence (TD), which can measure the distance between
collapsed building samples and non-collapsed building samples in the feature space. Both
J-M distance and TD are in the range of 0 to 2, and the higher value indicates a better
distinguishing ability of the feature.
The calculation formula for the J-M distance is shown in Equations (1) and (2):
Jij =2×1−e−B(1)
B=Ui−UjT×∑i+∑j
2−1×Ui−Uj
8+
ln"1
2×|∑i+∑j|
q(|∑i|×|∑j|)#
2(2)
where Uis the mean vector of the sample,
∑
is the covariance matrix, and iand jare
non-collapsed and collapsed building samples, respectively.
The calculation formula for TD is shown in Equation (3):
TDi j =2×1−e−Di j /8(3)
where D
ij
is the divergence between non-collapsed and collapsed buildings, and its calcula-
tion formula is shown in Equation (4):
Dij =trh∑i−∑j×∑−1
j−∑−1
ii
2+trh∑−1
i−∑−1
j×Ui−Uj×Ui−UjTi
2(4)
where Uis the mean vector of the sample,
∑
is the covariance matrix, t
r
[A] is the sum
of diagonal elements in matrix A, and iand jare non-collapsed and collapsed building
samples, respectively.
The extraction methods for various features are shown in Table 2.
2.2.2. Optimization of Parameters in Feature Extraction
The Local Mean, Local Standard Deviation, Local Entropy, and Local Coefficient of
Variation are easily affected by the window size. If the window size is too small, the feature
cannot reflect the fragmentation of collapsed buildings, while the feature calculated by an
oversized window may exaggerate the grayscale variations of non-collapsed buildings,
resulting in a larger feature value than it should be. Therefore, an inappropriate window
size may lead to a decrease in the ability of the feature to identify collapsed buildings.
To explore the optimal window size, we calculated the above features under a series of
window sizes and used J-M distance and TD as the measures of distinguishing ability.
Features calculated based on GLCM, such as Contrast, Correlation, Energy, and Ho-
mogeneity, are not only easily influenced by the window size but also by the gray level. If
the gray level is not large enough, the difference between different fragments of collapsed
buildings in grayscale will be diminished, so that the distinguishing ability for collapsed
buildings of the feature can be weakened. If the gray level is huge, it can not only lead
to excessive computation but also exaggerate the grayscale differences of non-collapsed
buildings, and the extracted texture feature is more susceptible to noise. To explore the
Remote Sens. 2023,15, 5747 6 of 18
optimal combination of window size and gray level, we calculate the above features under
a series combination and use J-M distance and TD as measures of distinguishing ability.
Table 2. Feature Extraction methods.
Features Calculation Methods
Red/Green/Blue Calculate the mean of the red/green/blue band values of all pixels as the Red/Green/Blue
feature of the building sample
Hue/Saturation/Intensity Calculate the mean of hue/saturation/intensity values of the color space transformed sample
image of all pixels as the Hue/Saturation/Intensity feature of the building sample
Global Entropy Calculate the entropy of grayscale values of all pixels as the Global Entropy feature of the
building sample
Global Coefficient of Variation
Calculate the coefficient of variation of grayscale values of all pixels as the Global Coefficient of
Variation feature of the building sample
Local Mean
Using a specific-sized sliding window, calculate the mean/the standard deviation/the
entropy/the coefficient of variation of the grayscale values in the window as the mean/the
standard deviation/the entropy/the coefficient of variation value of the central pixel, and take
the mean of the calculated values of all pixels as the Local Mean/the Local Standard
Deviation/the Local Entropy/the Local Coefficient of Variation feature of the building sample
Local Standard
Deviation
Local Entropy
Local Coefficient of
Variation
Local Moran’I Calculate the local Moran’I of each pixel, and take the mean of the local Moran’I values of all
pixels as the Local Moran’I feature of the building sample
Gradient Calculate the gradient value for every pixel by the Sobel operator, and take the mean of the
gradient values of all pixels as the gradient feature of the building sample
Gradient Orientation Entropy
Calculate the gradient orientation for every pixel by the Sobel operator, and take the entropy for
the gradient orientation values of all pixels as the Gradient Orientation Entropy feature of the
building sample
Gradient Orientation Standard
Deviation
Calculate the gradient orientation for every pixel by the Sobel operator, and take the standard
deviation for the gradient orientation values of all pixels as the Gradient Orientation Standard
Deviation feature of the building sample
Edge Perform a convolution operation using the Laplacian operator, and take the mean value as the
edge feature of the building sample
Edge Density
Perform a convolution operation using the Laplacian operator and apply thresholding
segmentation to obtain edges. Calculate the density of edges in each pixel’s neighborhood,
pixel by pixel. Afterwards, take the mean of the edge density values of all pixels as the edge
density feature of the building sample
Contrast
Based on GLCM, calculate the contrast/correlation/energy/homogeneity value at each pixel
and take the mean of the contrast/correlation/energy/homogeneity values of all pixels as the
Contrast/Correlation/Energy/Homogeneity feature of the building sample
Correlation
Energy
Homogeneity
Local Binary Patterns (LBP) Calculate LBP for every pixel with a certain radius, and take the mean of the LBP values of all
pixels as the LBP feature of the building sample
Gabor Feature Calculate the Gabor value for every pixel by Gabor filtering, and take the mean of the Gabor
values of all pixels as the Gabor feature of the building sample
Fractal Dimension Calculate the fractal dimension using the Box Counting Method as the Fractal Dimension
feature of the building sample
2.2.3. Assessment of the Application Effects of Selected Features
J-M distance and TD only illustrate the ability of the selected features to distinguish
collapsed buildings from non-collapsed buildings in feature space, so we test the application
effect of the features by using the images in the application test image set as an example. We
Remote Sens. 2023,15, 5747 7 of 18
used the thresholding method as the algorithm for extracting collapsed buildings instead of
using a more complex machine learning algorithm, as it is the simplest and most intuitive
method. Using the thresholding method can demonstrate that the extraction effect of
collapsed buildings is derived from the contribution of the features rather than from the
contribution of the classification algorithm. The technical flowchart for the assessment is
shown in Figure 3.
RemoteSens.2023,15,xFORPEERREVIEW7of18
theoptimalcombinationofwindowsizeandgraylevel,wecalculatetheabovefeatures
underaseriescombinationanduseJ-MdistanceandTDasmeasuresofdistinguishing
ability.
2.2.3.AssessmentoftheApplicationEffectsofSelectedFeatures
J-MdistanceandTDonlyillustratetheabilityoftheselectedfeaturestodistinguish
collapsedbuildingsfromnon-collapsedbuildingsinfeaturespace,sowetesttheapplica-
tioneffectofthefeaturesbyusingtheimagesintheapplicationtestimagesetasanexam-
ple.Weusedthethresholdingmethodasthealgorithmforextractingcollapsedbuildings
insteadofusingamorecomplexmachinelearningalgorithm,asitisthesimplestand
mostintuitivemethod.Usingthethresholdingmethodcandemonstratethattheextrac-
tioneffectofcollapsedbuildingsisderivedfromthecontributionofthefeaturesrather
thanfromthecontributionoftheclassificationalgorithm.Thetechnicalflowchartforthe
assessmentisshowninFigure3.
Figure3.Atechnicalflowchartforassessingtheapplicationeffectsofselectedfeatures.
(1) Collapsedbuildingidentification
TakeLocalEntropyasanexampletoillustratetheprocessofcollapsedbuildingiden-
tification.First,thevegetationindex(thecalculationformulaisshowninEquation(5)[23])
wasusedtomaskvegetationareasinthetestimage.Then,theLocalEntropyofevery
pixelwascalculatedinthemaskedimage,andthethresholdmethodwasusedtoextract
collapsedbuildings.Forthesegmentedbinaryimage,majorityanalysisandopeningop-
erationswereusedtoeliminatesmallandmeaninglesspatches,andthefinalidentification
resultwasobtained.𝑉𝐼 2 𝐺 𝑅 𝐵(5)
whereVIrepresentsthevegetationindex,andR,G,andBrepresentthepixelvaluesinthe
red,green,andbluebandsoftheimage.
(2) Accuracyevaluation
Theconfusionmatrixcalculatedbyrandomlygeneratedpixelsisusedtoevaluatethe
accuracyoftheidentificationofcollapsedbuildings.Approximately1000pixelswereran-
domlygeneratedfromthenon-collapsedandcollapsedbuildingsoftheclassificationre-
sult,respectively.Then,aconfusionmatrixiscalculatedbyvisuallyinterpretingevery
randompixel.Basedontheconfusionmatrix,precision,recall,andF1-scorewerecalcu-
latedastherecognitionaccuracyindicators.
Precisionreflectstheidentificationmethod’sabilitytocorrectlyidentifycollapsed
buildings.Recall,ontheotherhand,reflectstheidentificationmethod’sabilitytoavoid
misidentifyingnon-collapsedbuildings.Thecalculationformulasforprecisionandrecall
areshowninEquations(6)and(7).𝑝𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛
(6)
𝑟𝑒𝑐𝑎𝑙𝑙
(7)
whereTPisthenumberoftruepositivesamples,whichisthenumberofthesamples
classifiedascollapsedbuildingsand,infact,alsocollapsedbuildings;FPisthenumberof
falsepositivesamples,whichisthenumberofthesamplesclassifiedascollapsedbuild-
ingsbutactuallynon-collapsedbuildings;FNisthenumberoffalsenegativesamples,
Figure 3. A technical flowchart for assessing the application effects of selected features.
(1)
Collapsed building identification
Take Local Entropy as an example to illustrate the process of collapsed building identi-
fication. First, the vegetation index (the calculation formula is shown in Equation (5) [
23
])
was used to mask vegetation areas in the test image. Then, the Local Entropy of every
pixel was calculated in the masked image, and the threshold method was used to extract
collapsed buildings. For the segmented binary image, majority analysis and opening oper-
ations were used to eliminate small and meaningless patches, and the final identification
result was obtained.
VI =2×G−R−B(5)
where VI represents the vegetation index, and R,G, and Brepresent the pixel values in the
red, green, and blue bands of the image.
(2)
Accuracy evaluation
The confusion matrix calculated by randomly generated pixels is used to evaluate
the accuracy of the identification of collapsed buildings. Approximately 1000 pixels were
randomly generated from the non-collapsed and collapsed buildings of the classification
result, respectively. Then, a confusion matrix is calculated by visually interpreting every
random pixel. Based on the confusion matrix, precision, recall, and F
1
-score were calculated
as the recognition accuracy indicators.
Precision reflects the identification method’s ability to correctly identify collapsed
buildings. Recall, on the other hand, reflects the identification method’s ability to avoid
misidentifying non-collapsed buildings. The calculation formulas for precision and recall
are shown in Equations (6) and (7).
precision =TP
TP +FP (6)
recall =TP
TP +F N (7)
where TP is the number of true positive samples, which is the number of the samples
classified as collapsed buildings and, in fact, also collapsed buildings; FP is the number
of false positive samples, which is the number of the samples classified as collapsed
buildings but actually non-collapsed buildings; FN is the number of false negative samples,
which refers to the number of samples classified as non-collapsed buildings but actually
collapsed buildings.
The F
1
-score takes both precision and recall into account and can comprehensively
represent the accuracy of the classification results. The calculation formula is shown in
Equation (8).
F1=2×precision ×recal l
precision +recal l (8)
Remote Sens. 2023,15, 5747 8 of 18
3. Results
3.1. Ability of 25 Features to Distinguish Collapsed Buildings from Non-Collapsed Buildings
There are significant differences in the ability to distinguish collapsed buildings from
non-collapsed buildings of different features (Figure 4). Although the results of J-M distance
and TD are not entirely consistent, they generally show the same patterns. The result
shows Homogeneity, Energy, Local Entropy, Local Standard Deviation, and Gradient are
more capable of distinguishing collapsed buildings from non-collapsed buildings (J-M
distances are greater than 1.59, TDs are greater than 1.60), followed by Contrast, Local
Coefficient of Variation, Edge Density, and Global Entropy (J-M distances are 1.14–1.28,
TDs are
1.24–1.48
). However, Gradient Orientation Entropy, Fractal Dimension, LBP, Edge,
Local Mean, Correlation, Gradient Orientation Standard Deviation, Global Coefficient of
Variation, Gabor feature, Local Moran’I, and six spectral features perform relatively poorly
(J-M distances are less than 0.73, TDs are less than 1.07). There are also differences in the
extraction times of different features; the results are presented in Appendix A.
RemoteSens.2023,15,xFORPEERREVIEW8of18
whichreferstothenumberofsamplesclassifiedasnon-collapsedbuildingsbutactually
collapsedbuildings.
TheF
1
-scoretakesbothprecisionandrecallintoaccountandcancomprehensivelyrep-
resenttheaccuracyoftheclassificationresults.ThecalculationformulaisshowninEqua-
tion(8).
F1 2 𝑝𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 𝑟𝑒𝑐𝑎𝑙𝑙
𝑝𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 𝑟𝑒𝑐𝑎𝑙𝑙(8)
3.Results
3.1.Abilityof25FeaturestoDistinguishCollapsedBuildingsfromNon-CollapsedBuildings
Therearesignificantdifferencesintheabilitytodistinguishcollapsedbuildingsfrom
non-collapsedbuildingsofdifferentfeatures(Figure4).AlthoughtheresultsofJ-Mdis-
tanceandTDarenotentirelyconsistent,theygenerallyshowthesamepaerns.Theresult
showsHomogeneity,Energy,LocalEntropy,LocalStandardDeviation,andGradientare
morecapableofdistinguishingcollapsedbuildingsfromnon-collapsedbuildings(J-M
distancesaregreaterthan1.59,TDsaregreaterthan1.60),followedbyContrast,Local
CoefficientofVariati o n , EdgeDensity,andGlobalEntropy(J-Mdistancesare1.14–1.28,
TDsare1.24–1.48).However,GradientOrientationEntropy,FractalDimension,LBP,
Edge,LocalMean,Correlation,GradientOrientationStandardDeviation,GlobalCoeffi-
cientofVariat i on ,Gaborfeature,LocalMoran’I,andsixspectralfeaturesperformrela-
tivelypoorly(J-Mdistancesarelessthan0.73,TDsarelessthan1.07).Therearealsodif-
ferencesintheextractiontimesofdifferentfeatures;theresultsarepresentedinAppendix
A.
Figure4.(a)J-Mdistanceand(b)TDof25featuresfornon-collapsedandcollapsedbuildingsam-
ples.Note:LocalstdisLocalStandardDeviation,LocalCVisLocalCoefficientofVar i a t i o n , GO
Figure 4. (a) J-M distance and (b) TD of 25 features for non-collapsed and collapsed building samples.
Note: Local std is Local Standard Deviation, Local CV is Local Coefficient of Variation, GO Entropy is
Gradient Orientation Entropy, LBP is Local Binary Patterns, GO Std is Gradient Orientation Standard
Deviation, and Global CV is Global Coefficient of Variation.
3.2. Optimal Parameters for Feature Extraction
The ability of Local Mean, Local Entropy, Local Coefficient of Variation, and Local
Standard Deviation to distinguish collapsed buildings from non-collapsed buildings varies
under different window sizes (Figure 5). Due to the same trend of variation in J-M distance
and TD under different window sizes, only the results of J-M distance are shown here (the
results of TD are presented in Appendix B). The Local Mean is not a well-performed feature
and is not sensitive to changes in window size; therefore, it will not be discussed here. The
Remote Sens. 2023,15, 5747 9 of 18
J-M distances of the other three features exhibit the characteristic of initially increasing
and then decreasing as the window size increases. Under the premise of a sample spatial
resolution of around 0.5 m, the Local Entropy is optimal when the window size is 7 (i.e.,
around 3.5 m), and the Local Coefficient of Variation and Local Standard Deviation are
optimal when the window size is 5 (i.e., around 2.5 m).
RemoteSens.2023,15,xFORPEERREVIEW9of18
EntropyisGradientOrientationEntropy,LBPisLocalBinaryPaerns,GOStdisGradientOrienta-
tionStandardDeviation,andGlobalCVisGlobalCoefficientofVariation.
3.2.OptimalParametersforFeatureExtraction
TheabilityofLocalMean,LocalEntropy,LocalCoefficientofVar i at io n, andLocal
StandardDeviationtodistinguishcollapsedbuildingsfromnon-collapsedbuildingsvar-
iesunderdifferentwindowsizes(Figure5).DuetothesametrendofvariationinJ-M
distanceandTDunderdifferentwindowsizes,onlytheresultsofJ-Mdistanceareshown
here(theresultsofTDarepresentedinAppendixB).TheLocalMeanisnotawell-per-
formedfeatureandisnotsensitivetochangesinwindowsize;therefore,itwillnotbe
discussedhere.TheJ-Mdistancesoftheotherthreefeaturesexhibitthecharacteristicof
initiallyincreasingandthendecreasingasthewindowsizeincreases.Underthepremise
ofasamplespatialresolutionofaround0.5m,theLocalEntropyisoptimalwhenthe
windowsizeis7(i.e.,around3.5m),andtheLocalCoefficientofVariationandLocal
StandardDeviationareoptimalwhenthewindowsizeis5(i.e.,around2.5m).
Figure5.J-Mdistanceof(a)LocalMean,(b)LocalEntropy,(c)LocalCoefficientofVar i a t i o n , and
(d)LocalStandardDeviationunderdifferentwindowsizes.
Contrast,Correlation,Energy,andHomogeneityarecalculatedbasedontheGLCM;
theirabilityiseasilyinfluencedbywindowsizeandgraylevel(Figure6).Thevariation
trendsofJ-MdistanceandTDarethesameunderdifferentwindowsizesandslightly
differentunderdifferentgraylevels,butthisdoesnotaffecttheoptimalparameterresults
ofeachfeature.Therefore,onlytheresultsofJ-Mdistanceareshownhere(theresultsof
TDarepresentedinAppendixB).ExceptforCorrelation,whichisnotawell-performed
feature,theJ-MdistancesofContrast,Energy,andHomogeneityallshowacharacteristic
ofinitiallyincreasingandthendecreasingasthewindowsizeincreases.Theoptimal
Figure 5. J-M distance of (a) Local Mean, (b) Local Entropy, (c) Local Coefficient of Variation, and
(d) Local Standard Deviation under different window sizes.
Contrast, Correlation, Energy, and Homogeneity are calculated based on the GLCM;
their ability is easily influenced by window size and gray level (Figure 6). The variation
trends of J-M distance and TD are the same under different window sizes and slightly
different under different gray levels, but this does not affect the optimal parameter results
of each feature. Therefore, only the results of J-M distance are shown here (the results of
TD are presented in Appendix B). Except for Correlation, which is not a well-performed
feature, the J-M distances of Contrast, Energy, and Homogeneity all show a characteristic of
initially increasing and then decreasing as the window size increases. The optimal window
size is 5 or 7 (i.e., 2–4 m). The J-M distance of Contrast, Energy, and Homogeneity also
shows a characteristic of initially increasing and then decreasing with the increase in gray
level. The optimal gray level for Contrast is 8, for Energy is 16, and for Homogeneity is 32.
Remote Sens. 2023,15, 5747 10 of 18
RemoteSens.2023,15,xFORPEERREVIEW10of18
windowsizeis5or7(i.e.,2–4m).TheJ-MdistanceofContrast,Energy,andHomogeneity
alsoshowsacharacteristicofinitiallyincreasingandthendecreasingwiththeincreasein
graylevel.TheoptimalgraylevelforContrastis8,forEnergyis16,andforHomogeneity
is32.
Figure6.J-Mdistanceof(a)Contrast,(b)Correlation,(c)Energy,and(d)Homogeneityunderdif-
ferentwindowsizesandgraylevels.
3.3.ApplicationAbilityofSelectedFeatures
Inpracticalapplicationtesting,theselectedfeatures(LocalEntropy,Homogeneity,
Energy,LocalStandardDeviation,andGradient)caneffectivelyidentifycollapsedbuild-
ings(Table3).Therecallratesforidentifyingcollapsedbuildingsfromtestimagesareall
above90%,whiletheprecisionratesarerelativelylow,rangingfrom55%to95%.TheF
1
scoresarebetween0.71and0.94.Figure7showstheresultofcollapsedbuildingsidenti-
fiedinJoplin,andthefeatureusedhereisGradient.Itcanbeseenthatalmostallcollapsed
buildingshavebeenextracted,withalmostnomissingareas.However,therearesome
non-collapsedareasmisidentifiedascollapsedareas.
Tab l e 3.Theaccuracyofcollapsedbuildingrecognitionwhenapplyingselectedfeaturestolarge-
scaleremotesensingimages.
Features
JoplinTuscaloosaMoore
Precision
(%)
Recall
(%)
F
1
‐
Score
Precision
(%)
Recall
(%)
F
1
‐
Score
Precision
(%)
Recall
(%)
F
1
‐
Score
LocalEntropy94.5093.200.9470.1096.290.8184.0097.450.90
Homogeneity92.1092.280.9263.2096.780.7677.2096.740.86
Energy84.0094.590.8956.2095.740.7169.9097.630.81
LocalStandardDeviation91.1094.500.9362.2093.820.7575.0097.400.85
Gradient79.8091.300.8559.8094.770.7362.3097.040.76
Figure 6. J-M distance of (a) Contrast, (b) Correlation, (c) Energy, and (d) Homogeneity under
different window sizes and gray levels.
3.3. Application Ability of Selected Features
In practical application testing, the selected features (Local Entropy, Homogeneity,
Energy, Local Standard Deviation, and Gradient) can effectively identify collapsed buildings
(Table 3). The recall rates for identifying collapsed buildings from test images are all above
90%, while the precision rates are relatively low, ranging from 55% to 95%. The F
1
scores are
between 0.71 and 0.94. Figure 7shows the result of collapsed buildings identified in Joplin,
and the feature used here is Gradient. It can be seen that almost all collapsed buildings
have been extracted, with almost no missing areas. However, there are some non-collapsed
areas misidentified as collapsed areas.
Table 3. The accuracy of collapsed building recognition when applying selected features to large-scale
remote sensing images.
Features
Joplin Tuscaloosa Moore
Precision
(%)
Recall
(%)
F1-
Score
Precision
(%)
Recall
(%)
F1-
Score
Precision
(%)
Recall
(%)
F1-
Score
Local Entropy 94.50 93.20 0.94 70.10 96.29 0.81 84.00 97.45 0.90
Homogeneity 92.10 92.28 0.92 63.20 96.78 0.76 77.20 96.74 0.86
Energy 84.00 94.59 0.89 56.20 95.74 0.71 69.90 97.63 0.81
Local Standard Deviation 91.10 94.50 0.93 62.20 93.82 0.75 75.00 97.40 0.85
Gradient 79.80 91.30 0.85 59.80 94.77 0.73 62.30 97.04 0.76
Remote Sens. 2023,15, 5747 11 of 18
RemoteSens.2023,15,xFORPEERREVIEW11of18
Figure7.UsingGradienttoIdentifyCollapsedBuildingsinJoplin((a)istheoriginalimage,and(b)
isthecorrespondingidentificationresult).Note:Thestripinthemiddleoftheimagerepresentsthe
areawherebuildingscollapsedduetoahurricane,whiletheupperandlowerredareasarenon-
collapsedbuildingsthathavebeenmisidentifiedascollapsedbuildings.
4.Discussion
4.1.TheBestFeaturesSelectedforRapidRemoteSensingIdentificationofCollapsedBuildings
andTheirInfluencingFactors
Thisstudyfoundthattheabilityofspectralfeaturestodistinguishcollapsedbuild-
ingsfromnon-collapsedbuildingsisgenerallypoor.Becausethespectralcharacteristics
ofbuildingsaremainlyrelatedtomaterialsratherthanstates,acollapsedbuildingmay
haveasimilarspectrumtoitspre-collapsestate[24,25],whilebuildingswithdifferent
materialsmayhavequitedifferentspectralcharacteristics,eventhoughtheyareallcol-
lapsed.
IthasalsobeendiscoveredthatHomogeneity,Energy,LocalEntropy,LocalStandard
Deviation,andGradientcaneffectivelyandstablydistinguishcollapsedbuildingsfrom
non-collapsedbuildings.Thesefeaturesalsohavehighrecognitionaccuracyforcollapsed
buildingsinapplications.Theabilityofthesefeatureshasalsobeenconfirmedinother
studies.Forexample,Lietal.[20]usedtexturefeatureslikeHomogeneityextractedfrom
GLCMtoidentifycollapsedbuildingsinpost-earthquakeimagesofTa x i a n , Xinjiang,with
anoverallaccuracyof90.45%.BasedonvarioustexturefeaturessuchasHomogeneity,
Energy,andContrast,Samadzadeganetal.[21]extractedcollapsedbuildingsfrompost-
earthquakeimagesinBam,withanoverallclassificationaccuracyof74%andaKappa
coefficientof0.63.However,thesestudiesonlyconductedempiricalresearchononeor
severalfeaturesintheirrespectiveresearchareas,whilethisstudycomprehensivelyeval-
uated25featuresbasedonlargeanddiversesamplesdistributedin6regionsworldwide.
Inaddition,thewayweevaluatedthedistinguishingabilityoffeatureswasusingJ-M
distanceandTDratherthananymachinelearningalgorithm,sotheselectedfeaturesare
applicabletoallmethods.Therefore,thescreeningresultsaremoreuniversalandhave
morepracticalguidancevalue.
Thefeaturesselectedinthisstudycanstablyandeffectivelyidentifycollapsedbuild-
ings,butaentionshouldbepaidtotheimpactoftheparametersduringthefeatureex-
tractionprocess.BasedontheresultsinSection3.2,theoptimalparametersandparameter
seingprinciplesforreferencearegiveninTable4.
Figure 7. Using Gradient to Identify Collapsed Buildings in Joplin ((a) is the original image, and (b) is
the corresponding identification result). Note: The strip in the middle of the image represents the area
where buildings collapsed due to a hurricane, while the upper and lower red areas are non-collapsed
buildings that have been misidentified as collapsed buildings.
4. Discussion
4.1. The Best Features Selected for Rapid Remote Sensing Identification of Collapsed Buildings and
Their Influencing Factors
This study found that the ability of spectral features to distinguish collapsed buildings
from non-collapsed buildings is generally poor. Because the spectral characteristics of
buildings are mainly related to materials rather than states, a collapsed building may have
a similar spectrum to its pre-collapse state [
24
,
25
], while buildings with different materials
may have quite different spectral characteristics, even though they are all collapsed.
It has also been discovered that Homogeneity, Energy, Local Entropy, Local Standard
Deviation, and Gradient can effectively and stably distinguish collapsed buildings from
non-collapsed buildings. These features also have high recognition accuracy for collapsed
buildings in applications. The ability of these features has also been confirmed in other
studies. For example, Li et al. [
20
] used texture features like Homogeneity extracted from
GLCM to identify collapsed buildings in post-earthquake images of Taxian, Xinjiang, with
an overall accuracy of 90.45%. Based on various texture features such as Homogeneity,
Energy, and Contrast, Samadzadegan et al. [
21
] extracted collapsed buildings from post-
earthquake images in Bam, with an overall classification accuracy of 74% and a Kappa
coefficient of 0.63. However, these studies only conducted empirical research on one or
several features in their respective research areas, while this study comprehensively evalu-
ated 25 features based on large and diverse samples distributed in 6 regions worldwide.
In addition, the way we evaluated the distinguishing ability of features was using J-M
distance and TD rather than any machine learning algorithm, so the selected features are
applicable to all methods. Therefore, the screening results are more universal and have
more practical guidance value.
The features selected in this study can stably and effectively identify collapsed build-
ings, but attention should be paid to the impact of the parameters during the feature
extraction process. Based on the results in Section 3.2, the optimal parameters and parame-
ter setting principles for reference are given in Table 4.
Remote Sens. 2023,15, 5747 12 of 18
Table 4. Parameters and parameter setting principles that affect the effectiveness of remote sensing
feature extraction.
Feature Parameter Optimum Parameter Parameter Setting Principle
Local Entropy
Window size
The window size is
about 3.5 m The optimal window size is influenced by the size of the roof
area of a non-collapsed building, the size of fragments from
collapsed buildings, and the spatial resolution of the image.
Theoretically, the optimal window should be able to contain
different fragments of collapsed buildings and should not be
larger than the roof area of a non-collapsed building.
Local Standard
Deviation
The window size is
about 2.5 m
Local Coefficient
of Variation
The window size is
about 2.5 m
Contrast
Window size
and gray level
The window size is
about 3.5 m and the gray
level is 8
The optimal window setting principle is the same as above. The
optimal gray level is influenced by the complexity of the image.
In terms of the identification of collapsed buildings, the more
complex the roof structure is, the larger the optimal gray level
should be.
Energy
The window size is
about 2.5 m and the gray
level is 16
Homogeneity
The window size is
about 3.5 m and the gray
level is 64
Note that the window size measured in length can be converted into pixels based on the spatial resolution of the
remote sensing image. For example, for images with a resolution of 0.5 m, the window size of 2.5 m corresponds
to 5 pixels.
4.2. Operational Implementation Process and Application Prospects of Applying Selected Features
to Rapid Identification of Collapsed Buildings
Based on the selected features, we suggested an operational application process to
rapidly extract collapsed buildings. First, indexes such as vegetation index are used
to mask non-constructed areas as much as possible. Then, features selected based on
reliable experiments are used to extract collapsed buildings. Finally, majority analysis
and morphological operations are used to eliminate small fragments. Due to the simple
algorithm, this process can meet the requirement of a “rapid” response after a disaster.
Although the precision rate of this process is low (between 55% and 95%), it can ensure
a high recall rate (all above 90%). Therefore, it can quickly identify potential collapsed
building areas from post-disaster remote sensing images and exclude the majority of
non-collapsed areas, which can save time for post-disaster rescue.
This study demonstrated the application effect of each selected feature on identifying
collapsed buildings on single-temporal images. However, these features can be applied
under more conditions. Each selected feature can not only be used on a single-temporal
image after collapse but can also be used on multi-temporal images before and after
collapse. For example, Pesaresi et al. [
26
] used texture features combined with spectral
and morphological features to detect the collapsed areas in bi-temporal images before and
after the earthquake. In addition, the selected features can be used alone or in combination
with other features; for example, extract spatial statistical features based on the gradient
feature [
27
] or use the ratio of entropy to energy as a new feature for collapsed building
recognition [
28
]. Multiple features have complementary information between each other, so
the combination of multiple features tends to perform better than a single feature, especially
features with a lower correlation. How to combine multiple features, or further develop
new feature extraction techniques, is an interesting topic. It will be further explored in
future research.
5. Conclusions
Both in terms of accuracy and feasibility, the current methods for rapid identification of col-
lapsed buildings in remote sensing are still far from operational. This study extensively tested
25 remote sensing features, screened out features that can well identify collapsed buildings
individually, and suggested an operational application process from a feasibility perspective.
Remote Sens. 2023,15, 5747 13 of 18
Based on 2630 pairs of building samples distributed in 6 regions worldwide, this study
tested the ability of 25 remote sensing features (including spectral and spatial features) to
distinguish collapsed buildings from non-collapsed buildings. Based on large-scale remote
sensing images, the application effect of selected features in identifying collapsed buildings
was also tested. It was found that Homogeneity, Energy, Local Entropy, Local Standard
Deviation, and Gradient can stably and effectively distinguish collapsed buildings from
non-collapsed buildings and have high recognition accuracy when applied to large-scale
images. Contrast, Local Coefficient of Variation, Edge Density, and Global Entropy can
also distinguish collapsed buildings from non-collapsed buildings at a normal level, while
Gradient Orientation Entropy, Fractal Dimension, LBP, Edge, Local Mean, Correlation,
Gradient Orientation Standard Deviation, Global Coefficient of Variation, Gabor feature,
Local Moran’I, and six spectral features have relatively weak abilities. Each of the selected
features is able to identify collapsed buildings individually, and the findings were applicable
to all classification methods.
When applying selected features for extracting collapsed buildings, it is advisable
to set the window size of Local Entropy, Local Standard Deviation, and Local Coefficient
of Variation to 2–4 m. As for the features calculated by GLCM, such as Contrast, the
window size should be set to 2–4 m, and the gray level should be determined according
to the complexity of the image. The features selected in this study can be used alone or in
combination with multiple features. In practical identification processes, the features can
be used based on either single-temporal images or multi-temporal images.
Based on the selected features, we also suggested an operational application process
to rapidly extract collapsed buildings. Firstly, indexes such as vegetation index are used
to mask non-constructed areas as much as possible, then remote sensing features selected
based on reliable experiments are used to extract collapsed buildings, and finally, majority
analysis and morphological operations are used to eliminate small fragments. Due to
the simple algorithm, this process can meet the requirement of a “rapid” response after
a disaster. Although the precision rate of this process is low, it can ensure a high recall
rate. Therefore, it can quickly identify potential collapsed building areas from post-disaster
remote sensing images and exclude the majority of non-collapsed areas, saving time for
post-disaster rescue.
Author Contributions: All authors contributed in a substantial way to the manuscript. W.Z. con-
ceived and designed the research; R.L. analyzed the data and wrote the manuscript; and X.Y. con-
tributed to the writing and reviewing. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by the National Natural Science Foundation of China Major
Program (No. 42192580 and No. 42192581).
Data Availability Statement: Data could be publicly available at https://github.com/Lry99/
Screening-Image-Features-of-Collapsed-Buildings.git.
Acknowledgments: Thanks for the help provided by the whole group of Zhu W.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
In this work, all experiments are conducted with MATLAB 2020b on a desktop com-
puter with an Intel(R) Xeon(R) CPU E5-2650 v3@2.30 GHz. Figure A1 illustrates the
extraction time of 25 features on 2630 pairs of samples. Features extracted based on the
GLCM and Gabor features take the longest time due to the relative complexity of their
algorithms, followed by the features extracted from neighborhoods, and the global features
take the shortest time. Although the complexity of extraction varies from feature to feature,
the features screened in this study are all low-level features that are simple to extract.
With the high computational power of current computers, the extraction complexity of the
features selected in this study will not be a limitation for fast recognition.
Remote Sens. 2023,15, 5747 14 of 18
RemoteSens.2023,15,xFORPEERREVIEW14of18
GLCMandGaborfeaturestakethelongesttimeduetotherelativecomplexityoftheir
algorithms,followedbythefeaturesextractedfromneighborhoods,andtheglobalfea-
turestaketheshortesttime.Althoughthecomplexityofextractionvariesfromfeatureto
feature,thefeaturesscreenedinthisstudyarealllow-levelfeaturesthataresimpleto
extract.Withthehighcomputationalpowerofcurrentcomputers,theextractioncomplex-
ityofthefeaturesselectedinthisstudywillnotbealimitationforfastrecognition.
FigureA1.Extractiontimeof25features.Note:LocalstdisLocalStandardDeviation,LocalCVis
LocalCoefficientofVariat i o n , GOEntropyisGradientOrientationEntropy,LBPisLocalBinary
Paerns,GOStdisGradientOrientationStandardDeviation,andGlobalCVisGlobalCoefficient
ofVar i a t i o n .
AppendixB
TheabilityofLocalMean,LocalEntropy,LocalCoefficientofVar i at io n, andLocal
StandardDeviationtodistinguishcollapsedbuildingsfromnon-collapsedbuildingsvar-
iesunderdifferentwindowsizes(FiguresA2andA3).ThevariationaltrendsofJ-Mdis-
tanceandTDunderdifferentwindowsizesarethesame.TheLocalMeanisnotawell-
performedfeatureandisnotsensitivetochangesinwindowsize;therefore,itwillnotbe
discussedhere.
TheJ-MdistancesandTDsoftheotherthreefeaturesexhibitacharacteristicofini-
tiallyincreasingandthendecreasingasthewindowsizeincreases.Underthepremiseof
asamplespatialresolutionofaround0.5m,theJ-MdistanceandTDofLocalEntropyare
bothhighestwhenthewindowsizeis7(i.e.,around3.5m),andtheJ-MdistanceandTD
ofLocalCoefficientofVariationandLocalStandardDeviationarebothhighestwhenthe
windowsizeis5(i.e.,around2.5m).Therefore,theoptimalwindowsizeforLocalEntropy
is7(i.e.,around3.5m),andtheoptimalwindowsizeforLocalCoefficientofVa ri a ti on
andLocalStandardDeviationis5(i.e.,around2.5m).
Figure A1. Extraction time of 25 features. Note: Local std is Local Standard Deviation, Local CV
is Local Coefficient of Variation, GO Entropy is Gradient Orientation Entropy, LBP is Local Binary
Patterns, GO Std is Gradient Orientation Standard Deviation, and Global CV is Global Coefficient
of Variation.
Appendix B
The ability of Local Mean, Local Entropy, Local Coefficient of Variation, and Lo-
cal Standard Deviation to distinguish collapsed buildings from non-collapsed buildings
varies under different window sizes (Figures A2 and A3). The variational trends of J-M
distance and TD under different window sizes are the same. The Local Mean is not a
well-performed feature and is not sensitive to changes in window size; therefore, it will not
be discussed here.
RemoteSens.2023,15,xFORPEERREVIEW15of18
FigureA2.J-Mdistanceof(a)LocalMean,(b)LocalEntropy,(c)LocalCoefficientofVariation,and
(d)LocalStandardDeviationunderdifferentwindowsizes.
FigureA3.TDof(a)LocalMean,(b)LocalEntropy,(c)LocalCoefficientofVariation, and(d)Local
StandardDeviationunderdifferentwindowsizes.
Figure A2. J-M distance of (a) Local Mean, (b) Local Entropy, (c) Local Coefficient of Variation, and
(d) Local Standard Deviation under different window sizes.
Remote Sens. 2023,15, 5747 15 of 18
RemoteSens.2023,15,xFORPEERREVIEW15of18
FigureA2.J-Mdistanceof(a)LocalMean,(b)LocalEntropy,(c)LocalCoefficientofVariation,and
(d)LocalStandardDeviationunderdifferentwindowsizes.
FigureA3.TDof(a)LocalMean,(b)LocalEntropy,(c)LocalCoefficientofVariation, and(d)Local
StandardDeviationunderdifferentwindowsizes.
Figure A3. TD of (a) Local Mean, (b) Local Entropy, (c) Local Coefficient of Variation, and (d) Local
Standard Deviation under different window sizes.
The J-M distances and TDs of the other three features exhibit a characteristic of initially
increasing and then decreasing as the window size increases. Under the premise of a
sample spatial resolution of around 0.5 m, the J-M distance and TD of Local Entropy are
both highest when the window size is 7 (i.e., around 3.5 m), and the J-M distance and TD
of Local Coefficient of Variation and Local Standard Deviation are both highest when the
window size is 5 (i.e., around 2.5 m). Therefore, the optimal window size for Local Entropy
is 7 (i.e., around 3.5 m), and the optimal window size for Local Coefficient of Variation and
Local Standard Deviation is 5 (i.e., around 2.5 m).
The Contrast, Correlation, Energy, and Homogeneity are calculated based on the
GLCM. Therefore, their ability is easily influenced by window size and gray level
(Figures A4 and A5).
The variation trends of J-M distance and TD are the same under
different window sizes and slightly different under different gray levels, but this does
not affect the optimal parameter results of each feature. Correlation cannot distinguish
collapsed buildings from non-collapsed buildings, so it will not be discussed here.
Remote Sens. 2023,15, 5747 16 of 18
RemoteSens.2023,15,xFORPEERREVIEW16of18
TheContrast,Correlation,Energy,andHomogeneityarecalculatedbasedonthe
GLCM.Therefore,theirabilityiseasilyinfluencedbywindowsizeandgraylevel(Figures
A4andA5).ThevariationtrendsofJ-MdistanceandTDarethesameunderdifferent
windowsizesandslightlydifferentunderdifferentgraylevels,butthisdoesnotaffectthe
optimalparameterresultsofeachfeature.Correlationcannotdistinguishcollapsedbuild-
ingsfromnon-collapsedbuildings,soitwillnotbediscussedhere.
Asthewindowsizeincreases,theJ-MdistancesandTDsofContrast,Energy,and
Homogeneityallshowacharacteristicofincreasingfirstandthendecreasing.Takingthe
feature’sabilityandcomputationalefficiencyintoaccount(thesmallerthewindowis,the
fasterthecalculationis),theoptimalwindowsizeis5or7(i.e.,2–4m)withasample
spatialresolutionofabout0.5m.
Asthegraylevelincreases,theJ-MdistanceofContrastshowsacharacteristicoffirst
increasingandthendecreasing,reachingitspeakwhenthegraylevelis8.WhileTDal-
mostreachessaturationasthegraylevelreaches16,afterwards,thefluctuationisminimal
asthegrayscalelevelincreases.Taki n g intoaccountthefeature’sabilityandcomputa-
tionalefficiency(thesmallerthegraylevelis,thefasterthecalculationis),theoptimal
graylevelis8.TheJ-MdistanceofEnergyalsoexhibitsthecharacteristicoffirstincreasing
andthendecreasing,reachingitspeakwhenthegraylevelreaches16.Similarly,TDal-
mostreachesitspeakwhenthegraylevelincreasesto16.Afterthat,thefluctuationis
minimalasthegraylevelincreases.Therefore,theoptimalgraylevelis16.TheJ-Mdis-
tanceandTDofHomogeneitybothshowacharacteristicoffirstincreasingandthende-
creasing.TheJ-Mdistancereachesitspeakatthegraylevelof32,whileTDreachesits
peakatthegraylevelof64.Takingthefeature’sabilityandcomputationalefficiencyinto
account,theoptimalgrayscaleis32.
FigureA4.J-Mdistanceof(a)Contrast,(b)Correlation,(c)Energy,and(d)Homogeneityunder
differentwindowsizesandgraylevels.
Figure A4. J-M distance of (a) Contrast, (b) Correlation, (c) Energy, and (d) Homogeneity under
different window sizes and gray levels.
As the window size increases, the J-M distances and TDs of Contrast, Energy, and
Homogeneity all show a characteristic of increasing first and then decreasing. Taking the
feature’s ability and computational efficiency into account (the smaller the window is, the
faster the calculation is), the optimal window size is 5 or 7 (i.e., 2–4 m) with a sample spatial
resolution of about 0.5 m.
As the gray level increases, the J-M distance of Contrast shows a characteristic of first
increasing and then decreasing, reaching its peak when the gray level is 8. While TD almost
reaches saturation as the gray level reaches 16, afterwards, the fluctuation is minimal as
the grayscale level increases. Taking into account the feature’s ability and computational
efficiency (the smaller the gray level is, the faster the calculation is), the optimal gray level
is 8. The J-M distance of Energy also exhibits the characteristic of first increasing and then
decreasing, reaching its peak when the gray level reaches 16. Similarly, TD almost reaches
its peak when the gray level increases to 16. After that, the fluctuation is minimal as the
gray level increases. Therefore, the optimal gray level is 16. The J-M distance and TD of
Homogeneity both show a characteristic of first increasing and then decreasing. The J-M
distance reaches its peak at the gray level of 32, while TD reaches its peak at the gray level
of 64. Taking the feature’s ability and computational efficiency into account, the optimal
grayscale is 32.
Remote Sens. 2023,15, 5747 17 of 18
RemoteSens.2023,15,xFORPEERREVIEW17of18
FigureA5.TDof(a)contrast,(b)Correlation,(c)Energy,and(d)Homogeneityunderdifferentwin-
dowsizesandgraylevels.
References
1. Wu,J.;Chen,P.;Liu,Y.;Wang , J.Methodofearthquakecollapsedbuildinginformationextractionbasedonhigh-resolution
remotesensing.Geo.Spat.Inf.Sci.2013,29,35–38+47+2.
2. Lei,L.;Liu,L.;Zhang,L.;Bi,J.;Wu,Y.;Jiao,Q.;Zhang,W.Assessmentandanalysisofcollapsinghousesbyaerialimagesinthe
Wenchuanearthquake.J.RemoteSens.2010,14,333–344.
3. Dong,L.G.;Shan,J.Acomprehensivereviewofearthquake-inducedbuildingdamagedetectionwithremotesensingtechniques.
ISPRSJ.Photogramm.RemoteSens.2013,84,85–99.
4. Deng,L.;Wan g , Y.Post-disasterbuildingdamageassessmentbasedonimprovedU-Net.Sci.Rep.2022,12,15862.
5. Wan g , C.;Ji,L.;Shi,F.;Li,J.;Wang,J.;Enan,I.H.;Wu,T.;Yang,J.Collapsedbuildingdetectioninhigh-resolutionremotesensing
imagesbasedonmutualattentionandcostsensitiveloss.IEEEGeosci.RemoteSens.Lett.2023,20,8000605.
6. Xia,Z.;Li,Z.;Bai,Y.;Yu,J.;Adriano,B.Self-supervisedlearningforbuildingdamageassessmentfromlarge-scalexBDsatellite
imagerybenchmarkdatasets.InProceedingsofDatabaseandExpertSystemsApplications,Cham,Switzerland,22–24Augu st
2022;pp.373–386.
7. Zhou,Y.;Zhang,Y.;Chen,S.;Zhou,Z.;Zhu,Y.;Zhao,R.DisasterdamagedetectioninbuildingareasbasedonDCNNfeatures.
RemoteSens.Nat.Resour.2019,31,44–50.
8. Yang,W.;Zhang,X.;Luo,P.Transferabilityofconvolutionalneuralnetworkmodelsforidentifyingdamagedbuildingsdueto
earthquake.RemoteSens.2021,13,504.
9. Ma,H.;Liu,Y.;Ren,Y.;Yu,J.DetectionofCollapsedBuildingsinPost-EarthquakeRemoteSensingImagesBasedonthe
ImprovedYOLOv3.RemoteSens.2020,12,44.
10. Wu,F.;Wang , C.;Zhang,B.;Zhang,H.;Gong,L.Discriminationofcollapsedbuildingsfromremotesensingimageryusing
deepneuralnetworks.InProceedingsofthe2019IEEEInternationalGeoscienceandRemoteSensingSymposium(IGARSS),
Yok o h a m a , Japan,28July–2Au gust2019;pp.2646–2649.
11. Duan,P.;Kang,X.;Ghamisi,P.;Li,S.HyperspectralRemoteSensingBenchmarkDatabaseforOilSpillDetectionWithan
IsolationForest-GuidedUnsupervisedDetector.IEEETrans.Geosci.Electron.2023,61,5509711.
12. Ye,X.;Wa n g,J.;Qin,Q.DamagedbuildingdetectionbasedonGF-1satelliteremotesensingimage:AcasestudyforNepal
M_S8.1earthquake.ActaSeismol.Sin.2016,38,477–485+509.
13. Zeng,Z.;Li,L.;Wang,Z.;Lei,L.Fastextractionofcollapsedbuildingsinpost-earthquakehigh-resolutionimagesusing
supervisedclassification.RemoteSens.Inf.2011,5,76–80.
Figure A5. TD of (a) contrast, (b) Correlation, (c) Energy, and (d) Homogeneity under different
window sizes and gray levels.
References
1.
Wu, J.; Chen, P.; Liu, Y.; Wang, J. Method of earthquake collapsed building information extraction based on high-resolution
remote sensing. Geo. Spat. Inf. Sci. 2013,29, 35–38+47+2.
2.
Lei, L.; Liu, L.; Zhang, L.; Bi, J.; Wu, Y.; Jiao, Q.; Zhang, W. Assessment and analysis of collapsing houses by aerial images in the
Wenchuan earthquake. J. Remote Sens. 2010,14, 333–344.
3.
Dong, L.G.; Shan, J. A comprehensive review of earthquake-induced building damage detection with remote sensing techniques.
ISPRS J. Photogramm. Remote Sens. 2013,84, 85–99. [CrossRef]
4.
Deng, L.; Wang, Y. Post-disaster building damage assessment based on improved U-Net. Sci. Rep. 2022,12, 15862. [CrossRef]
[PubMed]
5.
Wang, C.; Ji, L.; Shi, F.; Li, J.; Wang, J.; Enan, I.H.; Wu, T.; Yang, J. Collapsed building detection in high-resolution remote sensing
images based on mutual attention and cost sensitive loss. IEEE Geosci. Remote Sens. Lett. 2023,20, 8000605. [CrossRef]
6.
Xia, Z.; Li, Z.; Bai, Y.; Yu, J.; Adriano, B. Self-supervised learning for building damage assessment from large-scale xBD satellite
imagery benchmark datasets. In Proceedings of the Database and Expert Systems Applications, Cham, Switzerland, 22–24 August
2022; pp. 373–386.
7.
Zhou, Y.; Zhang, Y.; Chen, S.; Zhou, Z.; Zhu, Y.; Zhao, R. Disaster damage detection in building areas based on DCNN features.
Remote Sens. Nat. Resour. 2019,31, 44–50.
8.
Yang, W.; Zhang, X.; Luo, P. Transferability of convolutional neural network models for identifying damaged buildings due to
earthquake. Remote Sens. 2021,13, 504. [CrossRef]
9.
Ma, H.; Liu, Y.; Ren, Y.; Yu, J. Detection of Collapsed Buildings in Post-Earthquake Remote Sensing Images Based on the Improved
YOLOv3. Remote Sens. 2020,12, 44. [CrossRef]
10.
Wu, F.; Wang, C.; Zhang, B.; Zhang, H.; Gong, L. Discrimination of collapsed buildings from remote sensing imagery using
deep neural networks. In Proceedings of the 2019 IEEE International Geoscience and Remote Sensing Symposium (IGARSS),
Yokohama, Japan, 28 July–2 August 2019; pp. 2646–2649.
11.
Duan, P.; Kang, X.; Ghamisi, P.; Li, S. Hyperspectral Remote Sensing Benchmark Database for Oil Spill Detection With an Isolation
Forest-Guided Unsupervised Detector. IEEE Trans. Geosci. Electron. 2023,61, 5509711. [CrossRef]
12.
Ye, X.; Wang, J.; Qin, Q. Damaged building detection based on GF-1satellite remote sensing image: A case study for Nepal M_S8.1
earthquake. Acta Seismol. Sin. 2016,38, 477–485+509.
13.
Zeng, Z.; Li, L.; Wang, Z.; Lei, L. Fast extraction of collapsed buildings in post-earthquake high-resolution images using supervised
classification. Remote Sens. Inf. 2011,5, 76–80.
Remote Sens. 2023,15, 5747 18 of 18
14.
Ye, X.; Qin, Q.M.; Liu, M.C.; Wang, J.; Wang, J.H. Building damage detection from post-quake remote sensing image based on
fuzzy reasoning. In Proceedings of the 2014 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Quebec
City, QC, Canada, 13–18 July 2014; pp. 529–532.
15.
Sumer, E.; Turker, M. An integrated earthquake damage detection system. ISPRS Int. Arch. Photogramm. Remote Sens. Spat. Inf.
Sci. 2006,36, 6.
16.
Du, H.; Zhang, F.; Lu, Y.; Lin, X.; Deng, S.; Cao, Y. Recognition of the earthquake damage to buildings in the 2021 Yangbi, Yunnan
MS_6.4 earthquake area based on multi-source remote sensing images. J. Seismol. Res. 2021,44, 490–498.
17.
Li, X.; Yang, W.; Ao, T.; Li, H.; Chen, W. An improved approach of information extraction for earthquake-damaged buildings
using high-resolution imagery. J. Earthq. Tsunami 2011,5, 389–399. [CrossRef]
18.
Ye, X.; Qin, Q.; Wang, J.; Wang, J.; Yang, X.; Qin, X. Detecting damaged buildings caused by earthquake using local gradient
orientation entropy statistics method. In Proceedings of the 2015 IEEE International Geoscience and Remote Sensing Symposium
(IGARSS), Milan, Italy, 26–31 July 2015; pp. 3568–3571.
19.
Rathje, E.M.; Crawford, M.; Woo, K.; Neuenschwander, A. Damage patterns from satellite images of the 2003 Bam, Iran,
Earthquake. Earthq. Spectra 2005,21, S295–S307. [CrossRef]
20.
Li, J.; Zhao, S.; Jin, H.; Li, Y.; Guo, Y. A method of combined texture features and morphology for building seismic damage
information extraction based on GF remote sensing images. Acta Seismol. Sin. 2019,41, 658–670+681.
21.
Samadzadegan, F.; Rastiveis, H.; Viii, W.G. Automatic detection and classification of damaged buildings, using high resolution
satellite imagery and vector data. ISPRS Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2008,37, 415–420.
22.
Gupta, R.; Goodman, B.; Patel, N.; Hosfelt, R.; Sajeev, S.; Heim, E.; Doshi, J.; Lucas, K.; Choset, H.; Gaston, M. Creating xBD: A
dataset for assessing building damage from satellite imagery. In Proceedings of the IEEE/CVF Conference on Computer Vision
and Pattern Recognition Workshops, Long Beach, CA, USA, 15–20 June 2019; pp. 10–17.
23.
Abdullah, W.M.; Yaakob, S.N. Modified excess green vegetation index for uneven illumination. Int. J. Curr. Res. 2017,9,
48656–48661.
24.
Liu, J.; Li, P. Extraction of Earthquake-Induced Collapsed Buildings from Bi-Temporal VHR Images Using Object-Level Homo-
geneity Index and Histogram. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2019,12, 2755–2770. [CrossRef]
25.
Song, J.; Wang, X.; Li, P. Urban building damage detection from VHR imagery by including temporal and spatial texture features.
J. Remote Sens. 2012,16, 1233–1245.
26.
Pesaresi, M.; Gerhardinger, A.; Haag, F. Rapid damage assessment of built-up structures using VHR satellite data in tsunami-
affected areas. Int. J. Remote Sens. 2007,28, 3013–3036. [CrossRef]
27.
Ye, X.; Qin, Q.; Wang, J.; Zheng, X.; Wang, J. Detecting damaged buildings caused by earthquake from remote sensing image
using local spatial statistics method. Geomat. Inf. Sci. Wuhan Univ. 2019,44, 125–131.
28.
Zhang, R.; Zhou, Y.; Wang, S.; Wang, F.; Zhang, T.; He, Y.; You, S. Construction and application of a post-quake house damage
model based on multiscale self-adaptive fusion of spectral textures images. In Proceedings of the 2020 IEEE International
Geoscience and Remote Sensing Symposium (IGARSS), Waikoloa, HI, USA, 26 September–2 October 2020; pp. 6631–6634.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
Available via license: CC BY 4.0
Content may be subject to copyright.
