infrastructures

Review
Image-Based Crack Detection Methods: A Review
Hafiz Suliman Munawar 1 , Ahmed W. A. Hammad 1, *, Assed Haddad 2 , Carlos Alberto Pereira Soares 3
and S. Travis Waller 4

1 School of Built Environment, University of New South Wales, Kensington, Sydney, NSW 2052, Australia;
[email protected]
2 Programa de Engenharia Ambiental, PEA/POLI & EQ, Universidade Federal do Rio de Janeiro,
Rio de Janeiro 21941-909, Brazil; [email protected]
3 Pós-Graduação em Engenharia Civil, Universidade Federal Fluminense, Niterói 24210-240, Brazil;
[email protected]
4 School of Civil and Environmental Engineering, University of New South Wales, Kensington,
Sydney, NSW 2052, Australia; [email protected]
* Correspondence: [email protected]

Abstract: Annually, millions of dollars are spent to carry out defect detection in key infrastructure
including roads, bridges, and buildings. The aftermath of natural disasters like floods and earth-
quakes leads to severe damage to the urban infrastructure. Maintenance operations that follow for
the damaged infrastructure often involve a visual inspection and assessment of their state to ensure
their functional and physical integrity. Such damage may appear in the form of minor or major
cracks, which gradually spread, leading to ultimate collapse or destruction of the structure. Crack
detection is a very laborious task if performed via manual visual inspection. Many infrastructure
elements need to be checked regularly and it is therefore not feasible as it will require significant





 human resources. This may also result in cases where cracks go undetected. A need, therefore, exists
for performing automatic defect detection in infrastructure to ensure its effectiveness and reliability.
Citation: Munawar, H.S.; Hammad,
A.W.A.; Haddad, A.; Soares, C.A.P.;
Using image processing techniques, the captured or scanned images of the infrastructure parts
Waller, S.T. Image-Based Crack can be analyzed to identify any possible defects. Apart from image processing, machine learning
Detection Methods: A Review. methods are being increasingly applied to ensure better performance outcomes and robustness in
Infrastructures 2021, 6, 115. https:// crack detection. This paper provides a review of image-based crack detection techniques which
doi.org/10.3390/infrastructures6080115 implement image processing and/or machine learning. A total of 30 research articles have been
collected for the review which is published in top tier journals and conferences in the past decade.
Academic Editor: A comprehensive analysis and comparison of these methods are performed to highlight the most
Kourosh Khoshelham promising automated approaches for crack detection.

Received: 7 July 2021

Keywords: crack detection; machine learning; artificial intelligence; image processing
Accepted: 3 August 2021
Published: 14 August 2021

Publisher’s Note: MDPI stays neutral

1. Introduction
with regard to jurisdictional claims in
published maps and institutional affil- Annually, millions of dollars are spent to acquire various tools and assets to carry
iations. out defect detection from key infrastructure which includes roads, bridges, buildings, and
water bodies [1]. Civil structures such as roads, bridges, buildings, and pavements are
often exposed to extreme physical stress which may be caused by natural disasters like
earthquakes, catastrophic incidents like blasts or daily usage. Such incidents can either
Copyright: © 2021 by the authors.
cause a complete collapse of the structure or may lead to physical damage that is often
Licensee MDPI, Basel, Switzerland.
represented in the form of cracks. Usually, cracks emerge at a microscopic level on the
This article is an open access article
surface of the infrastructure component [2]. These cracks make the component weak,
distributed under the terms and reduce its loading capacity and lead to discontinuities on the surface [3–5]. If such cracks
conditions of the Creative Commons are detected at an early stage, further damage can be reduced [6]. Undetected cracks
Attribution (CC BY) license (https:// can however spread through the surface and may lead to the complete collapse of the
creativecommons.org/licenses/by/ structure, resulting in fatalities, injuries, and financial loss. Manual methods of crack
4.0/). detection involve experts who examine the component visually and the use of specific

Infrastructures 2021, 6, 115. https://doi.org/10.3390/infrastructures6080115 https://www.mdpi.com/journal/infrastructures

 Infrastructures 2021, 6, x FOR PEER REVIEW 2 of 22
Infrastructures 2021, 6, 115 2 of 20

who examine the component visually and the use of specific tools to identify any deficiency
tools
in thetocomponent
identify any [6].deficiency
However,in themethod
this component [6]. However,
is tedious, this method
labour extensive is tedious,
and prone to hu-
labour
man error.extensive and prone
Automatic crack to human deals
detection error.withAutomatic crack detection
using technologies deals with
to identify using
cracks from
technologies to identify cracks from infrastructures. The level
infrastructures. The level of degradation can be determined by analyzing the length, width, of degradation can be
determined by analyzing the length, width, depth and severity
depth and severity of a crack. These measures can be used to make decisions regarding the of a crack. These measures
can be used toofmake
classification decisions
the crack, regarding
durability of thethe classification
structure and itsof the crack,
usage durability
[7]. Using of the
the traditional
structure and its usage [7]. Using the traditional inspection
inspection procedures which involve manual inspection, it is very time-consuming to de- procedures which involve
manual
termine inspection, it is very
the crack measures time-consuming
which make it difficult to determine the crack
to make inference measures
regarding thewhich
level of
make it difficult to make inference regarding the level of degradation. Hence, for a quick,
degradation. Hence, for a quick, effective, and reliable damage assessment, the crack detec-
effective, and reliable damage assessment, the crack detection process must be automated
tion process must be automated to replace the manual defect inspection methods. Some
to replace the manual defect inspection methods. Some testing methods like laser, infrared,
testing methods like laser, infrared, thermal, radiographic, and thermal testing approaches
thermal, radiographic, and thermal testing approaches have been used in the past to
have been used in the past to automate the process of crack detection [8–10]. However, more
automate the process of crack detection [8–10]. However, more recently, there has been
recently, there has been an increasing trend of using image-based methods for detecting
an increasing trend of using image-based methods for detecting cracks. These methods
cracks. These methods involve capturing images of the target component and analyzing
involve capturing images of the target component and analyzing them programmatically
them programmatically to find and classify cracks. Such methods are fast, less expensive,
to find and classify cracks. Such methods are fast, less expensive, and robust. The methods
and robust. The methods can be categorized into two types namely as image processing and
can be categorized into two types namely as image processing and machine learning. The
machine learning. The image processing methods do not require a model training process
image processing methods do not require a model training process and involve the use
and involve the use of filters, morphological analysis, statistical methods, and percolation
of filters, morphological analysis, statistical methods, and percolation techniques for the
techniques
detection offor the [11,12].
crack detection Onofthecrack
other[11,12].
hand,On thethe other hand,
machine learning the process
machineinvolves
learningthe pro-
cess involves the collection of a dataset of images, which are
collection of a dataset of images, which are supplied to the selected machine learning supplied to the selected ma-
model for training. Such methods may involve image processing steps for preprocessingfor
chine learning model for training. Such methods may involve image processing steps
preprocessing
and noise removal, and noise
but the removal, but the crack
crack detection task isdetection
done by task is donemachine
the trained by the trained
learning ma-
chine learning
model [13]. model [13].
Figure 1 shows
Figure 1 shows the thebasic
basicarchitecture
architectureofofananimage imageprocessing-based
processing-based method
method forfor crack
crack
detection. First using a camera or any other imaging mechanism, high-resolution images of
detection. First using a camera or any other imaging mechanism, high-resolution images
thethe
of target
target component
component areare
collected.
collected.TheThe images
images are are
thenthenpreprocessed
preprocessed which involves
which involvesusing
filters, segmentation and other approaches to remove noise
using filters, segmentation and other approaches to remove noise and shadows from the and shadows from the image.
The image
image. may bemay
The image converted to grayscale
be converted or binary
to grayscale form form
or binary if required by the
if required by specific crack
the specific
detection
crack method
detection beingbeing
method used.used.
The resultant
The resultantimageimageis applied to thetocrack
is applied detection
the crack proce-
detection
dure whichwhich
procedure uses image processing
uses image techniques
processing like edge
techniques like detection, segmentation,
edge detection, or pixel
segmentation,
analysis to highlight or segment the cracked part in the image
or pixel analysis to highlight or segment the cracked part in the image [14]. Parameter [14]. Parameter estimation
involves calculating
estimation the specific the
involves calculating properties
specificof the detected
properties of thecrack such as
detected its length,
crack such as width,
its
depth and
length, width, density.
depthSuch measures
and density. Suchhelp in making
measures helpdecisions
in making regarding
decisionsthe severitythe
regarding of a
crack. of a crack.
severity

Image Crack Feature

Preprocessing Crack Detection
Acquisition Extraction

Figure 1. Image
Figure 1. Image Processing
Processing Methods
Methods for
for Crack
Crack Detection.
Detection.

The
The basic
basic steps
steps to
to build
build aa machine
machine learning model for
learning model for crack
crack detection
detection are
are depicted
depicted in
in Figure 2. In the first step, a dataset must be collected showing surface
Figure 2. In the first step, a dataset must be collected showing surface cracks, which cracks, which
are to
are
be detected using the machine learning model. Previously, a study led by Lin et al.etused
to be detected using the machine learning model. Previously, a study led by Lin al.
used 30,000 low-resolution images for training [15]. The images are preprocessed
30,000 low-resolution images for training [15]. The images are preprocessed using image using
image processing
processing techniques
techniques to reduce
to reduce noise, noise,
remove remove
shadowsshadows and adjust
and adjust other other properties
properties such as
such as size and brightness of images. The cracks in these images then undergo
size and brightness of images. The cracks in these images then undergo pixel-wise annota- pixel-wise
annotation or labelling, where the defected pixels are annotated in the image. This step
tion or labelling, where the defected pixels are annotated in the image. This step can be
can be performed manually or using a labelling tool. One such example of labelling is
performed manually or using a labelling tool. One such example of labelling is to set crack
to set crack pixels as white or “1” in the image while the remaining pixels will be set as
pixels as white or “1” in the image while the remaining pixels will be set as black or “0”.
black or “0”. After this step, a machine learning model needs to be selected, which is to be
After this step, a machine learning model needs to be selected, which is to be used for crack
used for crack detection. In past studies various machine learning models such as support
detection. In past studies various machine learning models such as support vector machines
vector machines (SVM), CNN, and decision trees have been used for crack detection [16]. A
(SVM), CNN, and decision trees have been used for crack detection [16]. A cost/loss
cost/loss optimization function is then formulated to minimize the loss or cost of training
Infrastructures 2021, 6, x FOR PEER REVIEW 3 of 22

Infrastructures 2021, 6, 115 3 of 20

optimization function is then formulated to minimize the loss or cost of training the model.
A weighted cross-entropy loss function can be used for this purpose [17–20]. The designed
model
the will A
model. then be trained
weighted using the set
cross-entropy of function
loss annotatedcanimages collected
be used for thisinpurpose
the dataset. After
[17–20].
training
The the model,
designed a new
model will setbeoftrained
then imagesusing
will the
be applied to the model
set of annotated images tocollected
see if theinmodel
the
successfully
dataset. Afterclassifies crackedaregions
training the model, new setin
ofthe image.
images will be applied to the model to see if
the model successfully classifies the cracked regions in the image.

Dataset Model
Preprocessing Labelling Model Testing
Collection Training

Figure 2. Machine Learning Methods for Crack Detection.

Figure 2. Machine Learning Methods for Crack Detection.
In this paper, we present a review of the image processing and machine learning-
basedInmethods
this paper,forwe
crack detection
present which
a review have
of the beenprocessing
image proposedand overmachine
the last learning-based
decade. For
this purpose,
methods 30 research
for crack articles
detection which from
havejournals and top over
been proposed tier conferences wereFor
the last decade. retrieved
this pur-
and the respective crack detection methods were analyzed concerning the crack
pose, 30 research articles from journals and top tier conferences were retrieved and the re- detection
technique proposed,
spective crack its features,
detection methodsperformance,
were analyzeddataset details and
concerning the specific
the crack component
detection technique
to which the method is applicable. The results of each method are documented
proposed, its features, performance, dataset details and the specific component to which alongthe
with the corresponding limitations. A comparative analysis of these methods
method is applicable. The results of each method are documented along with the corre- is conducted
to highlight
sponding the most promising
limitations. A comparativemethods for automatic
analysis crack detection.
of these methods is conductedThetopaper is
highlight
organized as follows: Section 2 presents the methodology followed to collect
the most promising methods for automatic crack detection. The paper is organized as fol- data for the
study. The article retrieval and screening process are explained in detail in this section.
lows: Section 2 presents the methodology followed to collect data for the study. The article
Section 3 presents the result of the study, discussing the crack detection techniques proposed
retrieval and screening process are explained in detail in this section. Section 3 presents the
in the articles and highlighting the strengths and limitations of each method. Section 4
result of the study, discussing the crack detection techniques proposed in the articles and
discusses the results, analyzing the features of the methods and presenting a general picture
highlighting the strengths and limitations of each method. Section 4 discusses the results,
of the current advancement in this domain. Section 5 summarizes the goals, outcomes, and
analyzing the features of the methods and presenting a general picture of the current ad-
achievements of this research along with the prospects.
vancement in this domain. Section 5 summarizes the goals, outcomes, and achievements of
this
2. researchand
Materials along with the prospects.
Methods
The aim was to assess the development in this field and how these advanced tools are
2. Materials and Methods
facilitating post-disaster scenarios. To achieve the desired goals, top journals were searched
The aim
for recent and was to assess
significant workthe carried
development
out in thein this field and
domain. The how
reviewthese advanced
process tools are
was carried
facilitating post-disaster scenarios. To achieve
out in two phases i.e., retrieving articles and screening them. the desired goals, top journals were searched
for recent and significant
To retrieve the research work carried
articles out study,
for this in the the
domain.
chosen The review
search process
engines werewas carried
Scopus,
out in two
Google phases
Scholar, i.e., retrieving
Science articlesand
Direct, Elsevier, andSpringer
screeningfor them.
finding the latest developments
To retrieve the research
and interdisciplinary researcharticles for this
in the field. Thestudy,
next the
stepchosen
was tosearch
formulateengines
a setwere Scopus,
of queries
Google Scholar, Science Direct, Elsevier, and Springer for finding
to be used in each of the search engines to retrieve the articles. The major aim was to the latest developments
and interdisciplinary
fully exhaust the search research
database in the
andfield. The next
retrieve step wasnumber
a maximum to formulate a set of
of articles queries to
matching
be used
the domain in each of the search
of interest. We used engines to retrieveofthe
three categories articles.
terms The major
representing theaim was to fully
subdomains,
exhaust
to extractthe search of
a variety database
research and retrieve
articles. a maximum
After entering the number
search ofqueries,
articles amatching the do-
set of articles
main ofbased
ranked interest.
on We
theirused three categories
relevance of terms
were retrieved. Therepresenting
first category theofsubdomains,
phrases was to extract
formu-
a variety
lated of research
to retrieve articles.
articles that After
proposedentering
flood the search queries,
prediction models a set
usingof articles
image ranked
processing based
on their relevance
technologies were retrieved.
that utilised multispectralThe first category
sensors. The of phrases
phrases werewasformed
formulated to retrieve
by using key-
articlesrelated
words that proposed flood prediction
to flood prediction models “flood
which include using image processing
prediction”, “flood technologies
risk analysis” that
and “flood
utilised hazard mapping”
multispectral sensors.along with phrases
The phrases like “image
were formed by usingprocessing”
keywordsand “artificial
related to flood
intelligence”.
prediction which Theinclude
second “flood
category of terms was
prediction”, formulated
“flood to retrieve
risk analysis” articles
and “flood that pro-
hazard map-
posed
ping” flood
alongprediction
with phrases methods using these
like “image technologies.
processing” For this purpose,
and “artificial we used
intelligence”. The flood
second
prediction
category ofkeywords
terms wasalong with the
formulated to keyword “edge that
retrieve articles detection”,
proposed “mining patterns from
flood prediction meth-
images, Synthetic Aperture Radar, and “Image-based flood Alarm
ods using these technologies. For this purpose, we used flood prediction keywords along model”.
withAfter the first“edge
the keyword phasedetection”,
based on article
“mining retrieval,
patternsthe fromarticles
images,were passed Aperture
Synthetic through aRa-
screening phase to further
dar, and “Image-based floodnarrow
Alarmdown model”.the selection criteria. Four assessment criteria
were After
defined theto first
evaluate
phase thebased
articles:
on article retrieval, the articles were passed through a
screening
(1) phase to further narrow down the selection criteria. Four assessment criteria were
No duplicates
defined
(2) Timeto interval:
evaluate 2010–2021
the articles:
(3) Document type: research article, abstract, book chapter
(4) English language only
Infrastructures 2021, 6, 115 4 of 20

Thus, by filtering the articles based on these metrics, the most recent, applicable,
and unique research articles written in the English language were extracted. From the
1250 articles retrieved in the first phase, 94 articles passed all four selection criteria. Hence,
this review is based on these screened articles. The number of articles from each term
category i.e., image processing, artificial intelligence and integrated approach that passed
the screening phase is shown in Figure 1. The articles were screened for duplicates, non-
English articles, and review papers. Around 520 papers were removed for duplicates, 240
for non-English articles and 396 for review papers. Hence, overall, 94 papers were finally
collected as an output of the screening phase.
In this section, the method adopted to retrieve the data relevant to the study is
presented. The article retrieval and screening process are discussed in detail. Figure 3
shows the overall methodology followed for the study. To retrieve the articles most relevant
to the research questions proposed in the study, we define two categories of research articles.
These are:
• Cat-1
Infrastructures 2021, 6, x FOR PEER REVIEW Crack Detection using Machine Learning 5 of 22
• Cat-2: Crack Detection using Image Processing

Figure 3. Methodology
Figure 3. Methodology of
of article
article screening.
screening.

After defining the categories, we defined keywords related to each category. Figure 4
More precisely, the following assessment criteria for screening were outlined:
shows the keywords are search phrases that are to be entered in the search engines of a
1) Published
research between 2010
article repository. Thetoaim
2020
was to generate a maximum number of keywords for
2) category
each English Language
to retrieveonly
as many relevant research articles as possible. For this purpose, we
3) Websites
defined a basicmust
set ofbe: MDPI, Elsevier,
keywords IEEE
which are: Xplore,
{crack Arxiv,using
detection Science Directlearning, crack
machine
detection using image processing, crack detection, crack measurement,(letters,
4) Article type must be research article, review or book chapter abstracts and
crack classification}.
comments are excluded)
A keyword search process was conducted to find the most frequent keywords reported in
5) literature,
the No duplicates
relevant to the previous set of keywords. The retrieved keywords include:
“segmentation”,
The articles“support
meeting vector machine”,”
the specified technology”,
criteria ”computer
for screening vision” andA“classi-
were downloaded. total of
107 articles were retrieved at this stage. After this step, each article was carefully studied so
that only relevant articles were kept for the research. This involved reading the abstract,
methodology and results of each article. After the detailed content analysis, the articles not
found relevant to the defined categories were discarded. At the end of this process, a total
Infrastructures 2021, 6, 115 5 of 20

fier” as shown in Figure 4. The retrieved keywords illustrate that data that is to be gathered
Infrastructures 2021, 6, x FOR PEER REVIEW
for the
research must revolve around “crack detection technologies”. Keywords from 6 ofthis
22

set were used in combination with the base set of keywords, to completely exhaust the
database and retrieve a maximum number of articles relevant to the study.

Figure
Figure4.4.Keywords
Keywordsfor
forCrack
CrackDetection.
Detection.

The next step was to specify the resource from where the research articles are to be
retrieved. For this purpose, recent research articles from a wide range of journals and
conferences were retrieved. Search results were obtained for each specific search phrase
entered on the search engine. Screening criteria were selected based on the time frame
relevance to the topic of study, and authenticity of websites for article retrieval.
More precisely, the following assessment criteria for screening were outlined:
(1) Published between 2010 to 2020
(2) English Language only
(3) Websites must be: MDPI, Elsevier, IEEE Xplore, Arxiv, Science Direct
(4) Article type must be research article, review or book chapter (letters, abstracts and
comments are excluded)
(5) No duplicates
The articles meeting the specified criteria for screening were downloaded. A total of
107 articles were retrieved at this stage. After this step, each article was carefully studied
so that only relevant articles were kept for the research. This involved reading the abstract,
methodology and results of each article. After the detailed content analysis, the articles
Figure 5. Year-wise Distribution of Articles.
not found relevant to the defined categories were discarded. At the end of this process, a
total of 30 research articles were finally selected for the study. These articles met all the
3. Results
assessment criteria and were found relevant to the research questions posed in this study.
Figure 5 shows the year-wise distribution of the research articles which were selected
A total of 30 research papers were collected because of the article retrieval and
for the study. The graph shows a growing trend among studies that are moving towards
screening method. Figure
machine learning methods 6 shows the overall
for crack distribution
detection, especiallyofduring
these articles in the
the most image
recent pro-
years
cessing and machine learning domains. The pie chart shows that 67% of articles
(2016–2020). The use of image processing methods was common at the start of the decade proposed
crack detection
(2010), however,methods
with thethat
rapidemployed
advancementmachine learning
in the field ofmodels
artificialwhile 33% used
intelligence image
(AI), the
processing methods.
machine learning In this section,
methods took overa comprehensive
and gained rapid review of these
attention papers is
of various presented.
researchers
Comparative
who aimed toanalysis
introduceof automation
these techniques
in theiscrack
donedetection
in tabular forms. The performance out-
process.
comes, technique, dataset, imaging method and limitations of each method is presented to
make inference regarding the feasibility of the method and its applicability in real-time
crack detection tasks.
 Infrastructures 2021, 6, 115 Figure 4. Keywords for Crack Detection. 6 of 20

Figure 5. Year-wise Distribution of

Year-wise Distribution of Articles.
Articles.

3. Results
3. Results
A total of 30 research papers were collected because of the article retrieval and screen-
ing method. Figure
A total 6 shows
of 30 the papers
research overall distribution of these
were collected articles
because of in thearticle
the imageretrieval
processingand
and machine learning domains. The pie chart shows that 67% of articles proposed
screening method. Figure 6 shows the overall distribution of these articles in the image crack
pro-
detection
cessing andmethods
machine that employed
learning machine
domains. Thelearning
pie chartmodels
shows while 33%ofused
that 67% image
articles pro-
proposed
cessing methods. In this section, a comprehensive review of these papers
crack detection methods that employed machine learning models while 33% used image is presented.
Comparative analysis of these techniques is done in tabular forms. The performance
processing methods. In this section, a comprehensive review of these papers is presented.
outcomes, technique, dataset, imaging method and limitations of each method is 7presented
Infrastructures 2021, 6, x FOR PEER REVIEW of 22 out-
Comparative analysis of these techniques is done in tabular forms. The performance
to make inference regarding the feasibility of the method and its applicability in real-time
comes, technique, dataset, imaging method and limitations of each method is presented to
crack detection tasks.
make inference regarding the feasibility of the method and its applicability in real-time
crack detection tasks.

Figure
Figure 6. Distribution
6. Distribution of articles
of articles among
among machine
machine learning
learning andand image
image processing
processing domains
domains forfor crack
crack
detection.
detection.
3.1. Image Processing Based Crack Detection
3.1. Image Processing based Crack Detection
Table 1 summarizes the image processing-based methods reviewed in this paper for
Table 1 summarizes the image processing-based methods reviewed in this paper for
crack detection. Each of the methods has been discussed in the subsequent sections.
crack detection. Each of the methods has been discussed in the subsequent sections.

Table 1. Image Processing Methods for Crack Detection.

Imaging
Method Features Domain Image Details Results Limitations Ref.
Device/Source
Precision =
206
Recursive Tree 0.79&&&Recall = Increased runtime
Crack Detection Pavement images&&&800 - [4]
edge pruning 0.92&&&F- (up to 30 s)
× 600
Measure = 0.85
GP and Image 17&&&(varying
Crack Detection Concrete Digital Camera Accuracy = 80% - [10]
Infrastructures 2021, 6, 115 7 of 20

Table 1. Image Processing Methods for Crack Detection.

Imaging
Method Features Domain Image Details Device/Source Results Limitations Ref.

206 images Precision = 0.79

Recursive Tree edge Crack Detection Pavement - Recall = 0.92 Increased runtime [4]
pruning 800 × 600 F-Measure = 0.85 (up to 30 s)
17
GP and Image (varying
Crack Detection Concrete Digital Camera Accuracy = 80% - [10]
Filtering resolution)
5 Canon IXUS Results presented
Gabor Filter Crack Detection Pavement 336 × 339 pixels 80 IS Precision up to 95% [13]
on 5 images only
Crack Detection 14 IP Camera Error Range = -
Particle Filter & Measurement Civil Structures 12 MP 7.51–8.59% [14]
Cant calculate crack
Crack width; manual
Beamlet Transform Detection, Pavement 256 × 256 pixels - A method is fast & setting of thresholds [18]
measurement & robust to noise
Classification prevents full
automation
Median filter,
Hessian Matrix, Noisy Concrete 60 images SONY
probabilistic Crack detection Cyber-shot AUC = 0.9903 - [21]
Surfaces 640 × 480 pixels DSC-F828
relaxation
500 + 1969 + Crack500,
206 + 118 + 38 AIU = 0.081
FPHBN Crack Detection Pavement GAPs384, Time = 0.241 Method is not real [22]
(varying CrackTree200, time
s/image
resolution) CFD, Aigle-RN
Canny edge
72 images Detection rate =
detector, dilate Crack Detection Bridges 4288 × 2848 UAV - [23]
operators, Frangi resolution 98.7%
filter
UAS not stable in
Crack Detection DJI Mavic Pro most the absence of GPS
UAS Operator and Bridges Real-time crack DJI Mavic Pro suitable camera to [24]
detection and windy
measurement visualize cracks
atmosphere
The system is robust
consumer- to varying Accuracy affected
Shi-Tomasi feature Crack Detection Bridges Real-time crack grade digital by noise-limited [25]
point detection detection illumination
camera conditions and camera resolution
complex textures

3.1.1. Tree Structures

One of the many challenges faced by researchers in crack detection is the presence
of noise in the images, which makes it difficult to identify crack pixels [4,9,22,23]. To
tackle this problem, Zou et al. presented an approach to reduce shadows from the input
pavement images to make the cracks more prominent and easier to identify. A tensor
voting scheme was proposed that builds probability maps for cracks which involve using
visual clues of vicinity and connectivity [4]. Minimum spanning trees (MSTs) were built
which show all the possible links of the determined crack sources. The unwanted edges
were eliminated to get the output crack curves. The dataset for validation of the model
consisted of 206 pavement images. The system takes up to 16 s to determine cracks from
an input image. The precision-recall and F-measure values have been recorded as 0.79, 0.92
and 0.85 respectively indicating high accuracy and significance of the methods.

3.1.2. Genetic Programming

Nishikawa et al. employed image processing methods for automatic crack detection
from concrete images. Major cracks are detected through genetic programming (GP) [26].
Genetic programming has been used previously for space optimisation by implementing
augmented Lagrangian genetic algorithm, optimize largescale structures by application
of the fuzzy genetic algorithm, controlled system problems by introducing floating-point
genetic algorithm, applied for management purposes for senor allocation and forecasting
accident durations [11–13,27]. The remaining noise from the images is reduced by applying
filters. Minor cracks are then detected by repeatedly applying an image filter to the areas
around major cracks. Spatial derivatives of the intensity patterns in the crack segments are
calculated to determine crack widths. An accuracy of 80% has been achieved on a dataset
of 18 test images.
Infrastructures 2021, 6, 115 8 of 20

3.1.3. Image Filters

Some concrete materials have more textured surfaces, which makes the crack detection
task even more challenging. To deal with this issue, Salman et al., proposed an image
processing-based approach to detect cracks from pavement images. A Gabor filter was
employed for multidirectional crack detection [13]. A Gabor filter is a linear filter that
analyzes the texture in a region to determine the presence of content having a particular
frequency in a specific direction. Hence, this method is found to be very effective to detect
cracks from pavements having rich textures. The experimental results showed a precision
of 95%.
Statistical methods have been applied to images to detect cracks. Lins and Givigi
applied a statistical filtering approach to detect cracks and then used image-processing
based methods to determine the lengths and widths of the detected cracks. A particle
filter is used for crack detection on civil structures [14]. This filter is originally designed to
monitor objects in clutter. A vector is used to represent the position of each object at time
t. The error range recorded by this method is between 7.51% to 8.59%. To determine the
width and length of each crack, the total number of crack pixels determined by the crack
detection method is multiplied by the pixel resolution. The angle of a crack is calculated by
drawing a line between any two points on the crack and then using trigonometric rules.
Fujita and Hamamoto worked on a crack detection system that is robust to the presence
of noise in concrete images. The test image first passes through a pre-processing stage where
a median filter is applied to reduce noise and shade present in the image [28]. To make
the cracks more prominent in images, a line filter is applied which is a multi-scale filter
that uses a Hessian matrix. Multiscale Hessian filtering is useful for the enhancement and
segmentation of narrow fractures in 3D image data. Finally, to detect cracks, a probabilistic
relaxation method is applied to the resultant image. Adaptive thresholds are applied to
further improve the precision of crack detection. The method is tested on 60 concrete
images containing noise. The area under the curve (AUC) recorded by the system is 0.9903,
indicating the accuracy and precision of the system.
Yeum and Dyke worked on detecting defects from bridge images. They particularly
focused on detecting cracks that are present close to the bolts on steel surfaces. Images are
captured from varying angles and positions [29]. From each image, the region of interest
(ROI) is the area near to bolt but not the bolt itself. Median filter, canny edge detector
and dilate operators are used to extracting bolts from each image. After removing bolts
from the images, a Hessian matrix-based method called Frangi filter is applied to detect
crack like edges. The detection rate of the system is 98.7%, depicting the efficiency of crack
detection using this method.

3.1.4. Beamlet Transform

Ying and Salari proposed a beamlet transform method for the detection and classifi-
cation of cracks on pavement images. Beamlets are an organization of line segments at
varying angles, scales, and locations [30]. This method is used to retrieve linear features
from images such as edges and lines. Hence, it is deemed as effective in detecting cracks,
which are curvilinear features, from the noisy and textured surface images of pavements.
The extracted crack parts are connected, and each crack is then categorized into one of the
four categories, which are: (1) horizontal, (2) vertical, (3) transversal and (4) block.

3.1.5. Unmanned Aerial System (UAS)-Based Approach

Dorafshan et al. worked on crack detection from bridge images. A UAS was utilized
and the impact of lighting and distance of a crack from the camera on crack detection results
was investigated. Hence a measure called “achievable crack to platform (ACP) distance”
was evaluated [31]. This measure represents the maximum distance of the camera from
the platform from where a crack can be accurately detected. The UAS operator monitored
the image and adjusted lighting and distance until the crack was visible. This operator
also measures the length and width of each crack. Among all the cameras tested by the
Infrastructures 2021, 6, 115 9 of 20

system, the DJI Mavic Pro was the best camera used in the UAS for enhancing crack
visibility [32–35].

3.1.6. Shi-Tomasi Algorithm

Kong and Li worked on detecting cracks from steel bridges. A short video is captured
from the target bridge structure. Differential features resulting from the opening and
closing of cracks are tracked at each frame of the video [36]. Feature detection was done
using the Shi-Tomasi algorithm. The gaps in the surface movement of the bridge part in
the video stream were identified and tracked to detect cracks. Experimental results showed
the robustness and efficiency of this method even in the presence of varying illumination
conditions. One limitation is the increased dependency on the camera resolution, as the
accuracy was reduced on low-resolution videos [37–39].

3.2. Machine Learning-Based Crack Detection

Table 2 summarizes the machine learning methods reviewed. Each of the methods
has been discussed in the subsequent sections.

Table 2. Machine Learning-based Crack Detection.

Method Features Domain Dataset Device/Source Results Limitations Ref.

Need 16 s to find
64000 crack & Precision = 80.13% cracks on an
GoogleNet CNN, Crack Civil 64000 non-crack Canon Camera Recall = 86.09% [40]
FPN delineation Structures image of 6000 ×
images F-Measure = 81.55%
4000 pixels
Accuracy = 91.5%
Calf 584 images
CNN Defect detection Robotic Arm (training), 70.35% - [41]
Leather 400 × 400 pixels
(testing)
Precision = 0.8696
CNN Pavement 500 images Smartphone Recall = 0.9251 -
Crack detection 3264 × 2448 sensor [42]
F-Measure = 0.8965
Reduced
Crack Detection 20,000 crack & performance for
and density AP = 89.3%
FCN Concrete 20,000 non crack Public Dataset F-Measure = 89.3% crack density [43]
evaluation 227 × 227 evaluation in the
presence of noise
Crack detection,
measurement Less accuracy in
K-means clustering, and Characteri- 84 images detection of
Road Digital Camera F-Measure = 97% [44]
Gaussian Models zation and 1536 × 2048 pixels narrow cracks (<2
severity mm)
assessment
STRUM, SVM, Crack Detection 100 images
Adaboost, Random and density Bridge Robotic Accuracy = 95% - [45]
Forest 1920 × 1280 pixels Scanning
evaluation
7250 images Line scan Precision = 98.29%
SVM, MDNMS Crack Detection Road 4000 × 1000 cameras, laser Recall = 93.86% - [46]
pixels and HW-SW
260 training CrackTree, Does not work
CNN Crack Detection Pavement images CRKWH100, F-Measure = 0.87 well for cracks on [47]
CrackLS315,
512 × 512 pixels Stone331 stone images
500 Results subject to
CNN Crack Detection Pavement 3264 × 2448 Smartphones Accuracy = 91.3% [48]
pixels location variance
800
FCN Crack Detection Pavement (varying Accuracy = 97.96% -
& Measurement & Walls Digital Camera [49]
resolution)
Random Structured Crack Detection 38 + 118 images CDN, AigleRN Crack width not
& Characteriza- Road Precision = 96.73% measured; Not [50]
Forests, SVM tion 480 × 320 pixels Datasets tested on videos
Crack Detection 100
MorphLink C, ANN & Characteriza- Road 0.99 mm per pixel LRIS MSE = 0.0094–0.0105 - [51]
tion
To avoid
Nuclear overfitting a large
147344 crack, 20 captured
NB-CNN Crack Detection Power 149460 non-crack Average AUC= 96.8% number of [52]
Plant Com- videos training images
120 × 120 pixels 720 × 540 pixels
ponents required; Reliance
on GPU
644 + 323 images
Morphologic Image 0.1 × 0.1 × 0.0053
Processing, Logistic Crack Detection Steel Slabs mm 3D Profile Data Accuracy above 80% - [53]
Regression (width length
depth) resolution
Infrastructures 2021, 6, 115 10 of 20

Table 2. Cont.

Method Features Domain Dataset Device/Source Results Limitations Ref.

Transfer Learning Crack & Sealed 800 images 2000 ImageNet recall= 0.951; precision=
Pavement Dataset - [54]
(CNN) Crack Detection × 4000 pixels 0.847
Canny Algorithm, Success rate= 88% for
Crack Detection 400 images crack detection Not tested in
decision tree Pavement Digital Camera [55]
& classification 320 × 320 pixels 80% for crack real-time
heuristic classification
Morphological Parameters
analysis, setting need to be
segmentation, Crack Detection Subway 38000 images CMOS line scan
6144 × 1024 Accuracy > 90% done for images [56]
& Classification Tunnels cameras
extreme learning of different
machine classifier resolutions
2000 Image Library Precision (90.13%), Reduced accuracy
CNN Crack Detection Pavement 3D images of 5000 3D Recall (87.63%) and in finding hairline [57]
1 mm resolution images F-measure (88.86%) cracks
Morphological Crack 1910 non- crack,
Detection, Civil 3961 crack images Canon EOS 7D NN: Accuracy = 79.5% -
operations, NN, depth Structures 5184 × 3456 SVM: Accuracy = 78.3% [58]
SVM perception pixels.
Deep convolutional Recall = 71.98%
Crack Detection Road 527 images Black-box Precision = 77.68% - [59]
encoder-decoder Camera Intersection of Union =
network 59.65%

3.2.1. Convolutional Neural Network (CNN)

The CNN model is commonly found in the literature for crack detection. This model
has three layers of neurons: convolutional layer, pooling layer and fully connected layer.
The convolutional layer extract features from images which enables it to learn to distinguish
between crack and non-crack image. The pooling layer is used for down-sampling the
image and reducing its size by adjusting its dimensions. A fully connected layer is used in
the final stage of a CNN model, as it takes the output from the previous layer as input and
maps it to an output label.
Crack delineation involves both identification and segmentation of cracks from images.
Ni et al., automated these tasks using feature map fusion and pixel classification [40].
A CNN architecture named GoogleNet CNN was applied for crack classification. The
output was processed using feature pyramid network (FPN), which contains fusion layers
and consecutive convolutional layers which collectively perform crack delineation. The
results showed that the network was able to do crack delineation accurately achieving
a precision of 80.13%. One limitation is the long processing time of the network as it
takes 16 s approximately to detect all the cracks from an image having a resolution of
6000 × 4000 pixels. Deep learning with CNN architecture for surface crack detection with
enhanced computation. The proposed model was compared with other methods such as
VGG16, Inception, and ResNet. The proposed shallow CNN architecture model achieved
maximum accuracy of 99.8% in the minimum computation.
To assure driving safety, detecting defects from pavements and roads have also been
a subject of interest for many researchers. A defect detection system has been proposed
that detects cracks from pavement images using a CNN model [41]. A total of 500 images
for the dataset have been collected using a smartphone sensor. The CNN model is trained
using these images along with manually annotated ground truth data. To speed up the
training process a graphics processing unit (GPU) is utilized. In CNN architecture, rectified
linear units (ReLU) is used as an activation function, to further accelerate the process. The
model recorded a precision, recall and F-measure of 0.8696, 0.9251 and 0.8965 respectively.
Zou et al. [4] proposed DeepCrack, which is a trainable deep CNN model that detects
crack from input pavement images, by utilizing high-level features such as ridges to
represent cracks. The DeepCrack network is constructed on the architecture of SegNet
which is based on encoder-decoder design. The features produced in the convolutional
stages of the encoder network and the decoder network are fused in a pairwise manner.
Four standard crack datasets are used for evaluations which are: CrackTree, CRKWH100,
CrackLS315, Stone331. From these datasets, three are used for testing and one for training.
Infrastructures 2021, 6, 115 11 of 20

Experimental results show an F-measure above 0.87. The method works well for pavement
images and did not show good results on input stone images.
Regular examination of the components in nuclear power plants is deemed necessary
to ensure the safety of the workers. A total of 20 videos were captured from components in
a nuclear power plant and integrated CNN and a naïve Bayesian (NB) classifier to fuse and
analyze the data acquired from each video frame. In this framework, CNN detects cracks
from each input frame while NB eliminates the false positives in the output [42]. The data
fusion approach preserves each crack’s spatiotemporal consistency. This system achieved a
hit rate of 98.3%.
CrackNet is a CNN-based model presented by Zhang et al. [3] for crack detection on
pavements. There are no pooling layers in the CrackNet architecture, unlike the traditional
CNN model. This architecture ensures accuracy up to the pixel level as the image length
and width remains unchanged in all layers. Convolutional and fully connected layers
are the hidden layers of this network [26]. Training of the model was done using 1800
three-dimensional (3D) images. A total of 200 3D pavement images were captured for
testing the model. The system achieved a precision, recall and F–measure of 90.13%, 87.63%
and 88.86%.
Yang et al. [15] used a variation of CNN called fully convolutional network (FCN)
for pixel-level segmentation of cracks on the images of walls and pavements. The FCN
model is trained using multiple kinds of crack images [15]. Pixel skeletons having a width
of only one pixel are used to represent the crack segments predicted by the FCN model.
The morphological features of these cracks, like length, width and topology are then
measured using these skeletons. For validation, the crack segments acquired by the model
are compared to ground truth and the results produced by FCN model crack detection
systems. The overall accuracy achieved by the crack segmentation system is 97.96%. This
method outperforms the CrackNet model in that it provides pixel-level segmentation and
reduces the training time. However, its performance is less than CrackNet in terms of
accuracy.
Bang et al. [31] worked on the detection of road cracks at the pixel level. A deep
convolutional encoder-decoder network was used for this purpose [31]. The encoder part
of this network consists of convolutional layers which extract the crack features while
the decoder part consists of de-convolutional layers which find the crack locations in
images. The model was trained using 427 black box images which were extracted from
black-box videos and was tested on 100 images. The system recorded a precision, recall
and intersection of a union of 77.68%, 71.98% and 59.65% respectively.
Pauly et al. [14] used CNN to detect cracks from pavement images . A total of
500 images were collected using smartphones from pavements across the United States
of America (USA). The resolution of each image was 3264 × 2448. Each image was then
partitioned into patches of 99 × 99 pixels. Next, these patches were labelled as either
cracked or non-cracked. The experimental results showed that by increasing the depth of
the neural network, better performance outcomes are achieved. A limitation of this method
is the location variance problem according to which, when the model is tested on pavement
images collected from a different location, the performance is reduced. An accuracy of
91.3% was recorded by this crack detection system.
Researchers have proposed several methods to calculate the dimensions and mea-
surements of the detected cracks such as length, width and depth along with the densities.
Dung and Anh have proposed a system that detects cracks from concrete images and
calculates their densities. They performed semantic segmentation of cracks on concrete
images [5]. The system uses an encoder-decoder FCN model. A total of 500 annotated
images were used to train the encoder based on VCG16. The average precision achieved by
the network is 90%. The system detects cracks and also calculates the densities of these
cracks with reasonable performance. To measure the density of a crack, the total number of
segmented crack pixels is divided by the total number of pixels in the image. This ratio is
referred to as the pixel density [40–43].
Infrastructures 2021, 6, 115 12 of 20

Differentiating between cracks and sealed cracks is another challenge that exists in the
literature. Zhang et al. [23] worked on crack detection on pavement images having complex
textures while focusing on the problem of distinguishing between cracks and sealed cracks
having identical width and brightness. A CNN model is trained for the classification of
a pavement image into crack/sealed crack and background sections [43]. A section-wise
thresholding process is applied to the output image for pixel-based segmentation of cracks
and sealed cracks. A curve identification method based on tensor voting is employed to
extract the crack or sealed crack. A total of 800 images are used to test the system. The
system showed a recall of 0.951 and a precision of 0.847.

3.2.2. K-Means Clustering

Crack characterization methods deal with classifying the detected cracks based on
their types [44–46]. Oliveira and Correia presented such a crack detection and classification
approach, that does not require manual labelling of dataset images [6]. A total of 84 road
images were captured for the training of the system using a digital camera. Unsupervised
training of the system is done using images from the training dataset. A K-means clustering
method and a mixture of two Gaussian models were tested to detect cracks from the input
images. The results showed that the Gaussian models’ mixture showed the best F-Measure
i.e., 93.5% and the least error rate that is 0.6%. In the case of recall, this method achieved
the second-best performance which is 95.5%. The detected cracks are classified either as
longitudinal cracks, transversal cracks or miscellaneous. This is done by analyzing the
connected components of each crack and computing a crack skeleton. Using the crack
skeleton, a crack width is determined. The width is further analyzed to determine the
crack severity level. One limitation is the lack of accuracy of the system in the detection of
narrow cracks (width < 2 mm).

3.2.3. Logistic Regression

In steel products, consistent casting causes lasting cracks on the surfaces of steel
slabs [47–49]. Hence, to avoid wasting time and costs in dealing with defective products, it
is crucial to detect these cracks at an early stage. Landstrom and Thurley presented a crack
detection and measurement system that uses morphological image processing. Initially,
80% length of a crack in an image is extracted by applying segmentation and minor defects
or cracks are discarded [22]. After that, statistical classification is performed using logistic
regression on these segmented images which detects all the major cracks. Overall accuracy
recorded by the system is above 80%.

3.2.4. Feature Pyramid and Hierarchical Boosting Network (FPHBN)

Yang et al. [15] proposed FPHBN to detect cracks on pavement surfaces. To assess the
performance of crack detection on images, the researchers presented a new measurement
method called “average intersection over union” (AIU) [15]. The method was evaluated
on five standard crack datasets. The AIU achieved by the method was 0.081 and the time
taken to show output on a single image is 0.241 s. Other studies have used this technique
for pavement and concrete crack detection [50–53].

3.2.5. Support Vector Machines (SVM)

Gavilan et al. [8] presented a road distress detection method. Road images were
captured by a vehicle having line scan cameras, laser beams and the required Hardware
and Software (HW-SW) for scanning and storage [8]. After preprocessing of images, a
multiple directional non-minimum suppression (MDNMS) method is applied for crack
detection. A linear SVM classifier is used to differentiate between various pavements
across Spain to determine optimal parameters for crack detection. The performance of the
crack detection method is improved by adapting parameters specific to the pavement. The
method achieved a precision of 98.29% and recall of 93.86%. Several studies have applied
SVM for crack detection [54–56].
Infrastructures 2021, 6, 115 13 of 20

3.2.6. SVM and Random Forest

Defect detection from bridges play a vital role in the maintenance of their structure
and ensuring the safety of people who use them. Prasanna et al. performed crack detection
on bridge images captured using robotic imaging. They proposed a crack detection method
called spatially tuned robust multi-feature (STRUM). Machine learning methods: SVM,
Adaboost and random forest are used to finally classify the crack and non-crack pixels [7].
Robust curve fitting is utilized to locate the possible crack regions, despite the existence
of noise in the images. An accuracy of 95% is achieved using the STRUM classification
approach as opposed to a 69% accuracy which has been achieved using a traditional image
processing approach. A continuous bridge mosaic is created using series of images from
the robot. This mosaic is further used to develop crack density maps [57–59].

3.2.7. SVM and Artificial Neural Network (NN)

A crack detection system for civil structures that performs segmentation by consider-
ing the depth parameters has been proposed. Crack depths are determined by performing
reconstruction of 3D scenes [27]. A distinguishing factor of this method is that it retrieves
the whole crack from the input image while the other methods that use edge detection just
segment the cracked part. Morphological methods are adopted for crack segmentation and
extraction. NN and SVM classifiers were used for final crack classification. For the NN
model, the accuracy was 79.5% while for SVM, 78.3% accuracy was recorded. Similar work
has been reported using SVM and NN for enhanced precision for crack detection [60–62].

3.2.8. Artificial Neural Network (ANN)

The fragmentation of cracks in the images is a problem frequently faced by researchers.
To tackle this issue, Wu et al. [19] proposed a crack defragmentation method called
“MorphLink-C”. This technique connects different crack parts present in an image [16]. A
dilation transform is applied to group crack fragments. These fragments are then connected
by applying a thinning transform. Crack width is also determined using this approach.
The system is developed specifically for crack detection in road and pavement images.
An artificial neural network (ANN) is used to perform the final classification to label the
images as “crack” or “no-crack”. The MorphLink-C method has been demonstrated to
improve the classification accuracy while reducing the training time of the classifier [63–65].

3.2.9. Random Structured Forests

CrackForest is a crack detection system for roads [66–68], such as Shi et al. [15] which
uses random structured forests. This model targets the issue of inhomogeneity of the
intensity of cracks present in the road images. To detect cracks in such images, integral
channel features are applied to achieve an improved depiction of cracks [15]. After this
step, a random structured forests method is employed to detect cracks. This method can
accurately detect arbitrary and complex cracks from images. An SVM model is employed
to classify the cracks concerning their type. Instead of having fixed labels for the crack type,
this method extends the crack types to thousands of dimensions. This enables the system
to characterize the arbitrary cracks as having complicated forms. The overall precision
achieved for crack classification is 96.73%.

3.2.10. Decision Tree

Fernandez et al. used various image processing and machine learning methods to
detect cracks from 400 pavement images. The operations include morphology-based filter,
canny edge detector, bilateral filter and transformation using logarithms [23]. Finally, a
decision tree classifier was used to classify the crack according to its type. The crack types
are transverse cracks, longitudinal cracks and alligator cracks. The success rate of crack
detection was 88% and for crack classification, the success rate was 80%. Application of a
decision tree approach has been useful for crack detection [69].
 tect cracks from 400 pavement images. The operations include morphology-based filter,
canny edge detector, bilateral filter and transformation using logarithms [23]. Finally, a de-
cision tree classifier was used to classify the crack according to its type. The crack types are
transverse cracks, longitudinal cracks and alligator cracks. The success rate of crack detec-
Infrastructures 2021, 6, 115 14 of 20
tion was 88% and for crack classification, the success rate was 80%. Application of a decision
tree approach has been useful for crack detection [69].

4. Analysis and Discussion

4. Analysis and Discussion
In this section, we present an analysis of the crack detection studies reviewed in this
In this section, we present an analysis of the crack detection studies reviewed in this
paper based on the functions they have covered, the source of the dataset, the domain to
paper based on the functions they have covered, the source of the dataset, the domain to
which the method is applicable and the performance of the method.
which the method is applicable and the performance of the method.
4.1.
4.1. Functionality
Functionality BasedBased Analysis
Analysis
All the reviewed articles proposed crack
All the reviewed articles proposed crackdetection
detectionmethods.
methods.However,
However,aafew fewofofthem
them
also
also focused on crack classification, which deals with categorizing the crack accordingtotoits
focused on crack classification, which deals with categorizing the crack according
type e.g.,
its type transversal
e.g., transversalororlongitudinal
longitudinalcracks,
cracks,and
and sealed cracks. Many
sealed cracks. Manyarticles
articlesperformed
performed
measurements of the detected crack. For example, various methods
measurements of the detected crack. For example, various methods were proposed were proposed to meas-
to
ure the crack
measure length,
the crack width,
length, depth
width, and and
depth density [34,40–42].
density Hence,
[34,40–42]. crack
Hence, classification
crack and
classification
measurement
and measurement are some important
are some problems
important that have
problems been been
that have tackled in several
tackled studies
in several giving
studies
them
giving anthem
edgean over studies
edge over that just that
studies focused
just on the crack
focused detection
on the task. Figure
crack detection 7 shows
task. Figurethat
7
most
showsstudies
that most(16)studies
focused(16)
on focused
crack detection
on crackonly, whileonly,
detection 10 ofwhile
them10dealt withdealt
of them bothwith
crack
detection
both crackand measurement.
detection Just four studies
and measurement. proposed
Just four studiesmethods
proposedfor classifying
methods cracks after
for classifying
detecting
cracks after them. Hence, them.
detecting this analysis
Hence, points
this towards
analysis the needtowards
points for further
thedevelopment in the
need for further
classification
developmentmethods for cracks. methods
in the classification Figure 7 presents
for cracks.theFigure
distribution of the
7 presents thearticles basedofon
distribution
their features
the articles or the
based onfunctionality
their featuresthey have
or the provided. they have provided.
functionality

Figure 7.
Figure Functionality based
7. Functionality based Analysis
Analysis of
of Crack
Crack Detection
Detection Methods.
Methods.

4.2. Crack Classification Analysis

4.2. Crack Classification Analysis
Different articles were reviewed in the study focused on crack classification. Table 3
Different articles were reviewed in the study focused on crack classification. Table 3
shows the crack types identified by researchers while categorization of the detected crack.
shows the crack types identified by researchers while categorization of the detected crack.
This table illustrates that all the articles proposed some unique categories to classify cracks.
This table illustrates that all the articles proposed some unique categories to classify cracks.
Longitudinal (caused due to late or shallow saw cutting, inadequate base support, and
Longitudinal (caused due to late or shallow saw cutting, inadequate base support, and built-
built-in warping and curling) and transversal cracks (extends along the cross-section of
in warping and curling) and transversal cracks (extends along the cross-section of structure,
structure, usually perpendicular to its centerline) have been tackled in two studies [6,23]
usually perpendicular to its centerline) have been tackled in two studies [6,23] while a new
while a new crack type called “sealed crack” was proposed in [20]. The authors of [15]
crack type called “sealed crack” was proposed in [20]. The authors of [15] classified the
classified the cracks based on their dimensions, hence concluded that thousands of types
exist for the cracks detected in their study.

Table 3. Studies focusing on crack classification.

Crack Types Ref.

Longitudinal cracks, transversal cracks or miscellaneous. [6]
Types based on dimensions [15]
Crack, sealed crack [20]
Transverse cracks, longitudinal cracks and alligator cracks. [23]
Infrastructures 2021, 6, 115 15 of 20

4.3. Crack Measurement Analysis

Ten articles reviewed in this study proposed mechanisms to measure various proper-
ties of the detected cracks. These properties include the length, depth, width and density,
etc. of the detected crack. Such measures give important inference about the crack severity
and the condition of the structure of the component from where it is detected. Table 4
shows that the majority of research articles focused on finding the length and width of
the crack. Other measures such as density, topology and angle severity have also been
calculated in some of the articles. Among these methods, the method proposed in the
article [5] calculates the highest number of measures, followed by [12] and [14] which
introduce two new measures which are topology and angle, respectively.

Table 4. Studies focusing on Crack Measurement.

Crack Measurement Ref.

Length, width, depth, density [5]
Density [7]
Widths [10]
Length, width & topology [12]
Angle, width & length [14]
Depth [27]
Width, severity [6]
Width [16]
Length, width [24]
Length, width [29]

4.4. Image Source

Before developing any image-based crack detection system, collecting images for the
dataset is another problem that needs to be resolved. From the analysis conducted on
the reviewed articles, we can infer that 70% of researchers build their dataset (Figure 8).
This involved using various types of cameras such as line cameras, robotic scanning, and
smartphones. Developing custom datasets is preferred by researchers as this enables them
to capture the specific requirements of the crack detection algorithm being used. About 30%
of researchers used pre-built crack datasets, which contained crack images. These datasets
are being used as standards for testing and training crack detection models. Examples of
these datasets include Crack500, GAPs384, CrackTree200 and CFD, Aigle-RN. Such datasets
Infrastructures 2021, 6, x FOR PEER REVIEW 17 of 22
are widely used in machine learning models, as they require a large number of images or
training, hence, it is deemed more feasible to use standard domain-specific datasets.

Figure 8. Dataset Source of the Reviewed Studies.

4.5. Domain of Crack Detection

Each method reviewed in this study is designed to detect cracks from a specific element
or component. In this review, it is observed that a large majority of papers focused on de-
tecting cracks from pavements (Figure 9). A large portion of articles focused on civil struc-
tures in general. Hence, these methods can detect cracks from any infrastructure element
Infrastructures 2021, 6, 115 16 of 20
Figure 8. Dataset Source of the Reviewed Studies.

4.5. Domain of Crack Detection

4.5. Domain of Crack Detection
Each method reviewed in this study is designed to detect cracks from a specific element
Each method reviewed in this study is designed to detect cracks from a specific
or component. In this review, it is observed that a large majority of papers focused on de-
element or component. In this review, it is observed that a large majority of papers focused
tecting crackscracks
on detecting from pavements (Figure
from pavements 9). A large
(Figure portion
9). A large of articles
portion focused
of articles on civil
focused on struc-
civil
tures in general. Hence, these methods can detect cracks from any infrastructure
structures in general. Hence, these methods can detect cracks from any infrastructure element
like roads,
element bridges,
like walls and
roads, bridges, pavements.
walls Crack detection
and pavements. from other
Crack detection frommaterials such as
other materials
leather,
such as leather, steel and nuclear power plant components are also considered instudies
steel and nuclear power plant components are also considered in three of the three
[2,17,19]. However, the rest of the articles only focused on infrastructure components
of the studies [2,17,19]. However, the rest of the articles only focused on infrastructure like
subway tunnels, roads, bridges, pavements and walls. Hence, this points towards
components like subway tunnels, roads, bridges, pavements and walls. Hence, this points the need
to focus on
towards thedeveloping methods
need to focus for crackmethods
on developing detectionforwhich
crackcan be applied
detection which tocan
other
be elements
applied
as well such as leather, steel, textiles and other industrial products.
to other elements as well such as leather, steel, textiles and other industrial products.

Thedomain
Figure9.9.The
Figure domainof
ofthe
theCrack
CrackDetection
DetectionTechniques.
Techniques.

4.6. Precision Level

4.6. Precision Level
The applicability and reliability of any crack detection method can be assessed by
The applicability and reliability of any crack detection method can be assessed by re-
reviewing its performance outcomes. In the reviewed articles, researchers have used
viewing its performance outcomes. In the reviewed articles, researchers have used varying
varying performance measures to validate their systems such as accuracy, AUC, precision,
performance measuresHowever,
recall and F-measure. to validate their systems
precision was usedsuch
mostas commonly
accuracy, AUC,
among precision, recall
these articles
and F-measure.
for the evaluationHowever, precision
of the crack was model.
detection used most
Thecommonly among
Table 5 shows the these articles
precision for of
values the
evaluation of the crack detection model. The table 5 shows the precision values
these methods. This shows the methods presented in [8,15] recorded the highest precision of these
methods.
values whichThis shows
rangedthe methods
between presented
95% to 100%.in All
[8,15]
of recorded the highest
these methods precision
showed values
reasonable
which ranged between 95% to 100%. All of these methods showed reasonable
performance as shown by the precision values which ranged between 75% to 100%. performance
as shown by the precision values which ranged between 75% to 100%.
Table 5. Performance (precision) analysis of the various crack detection methods.

Precision (%) Ref.

95–100 [8,15]
90–95 [13,26]
85–90 [3,5]
80–85 [1,20]
75–80 [4,28]

5. Gaps and Challenges

Crack detection is a challenging task as cracks have an irregular form with no specific
shape or size. Hence, no pre-built method can be used to recognize a crack or distinguish it
from the background texture and noise. In this paper, we reviewed crack detection methods
belonging to two domains, namely image processing and machine learning. By examining
the methodology presented in the articles and the limitations, we can infer that both these
domains pose specific challenges when it comes to accurate crack detection from any
component. Crack classification has been rarely addressed in the articles. The type of the
detected crack facilitates the inspection process by giving important indications regarding
the nature of the crack, its cause and its severity. More research needs to be focused on
Infrastructures 2021, 6, 115 17 of 20

developing methods to classify the detected crack so that the system can recognize the
type of the crack and the maintenance operations can be focused on dealing with the
specific type of crack. Image processing methods have given satisfactory performance on
the custom datasets built by the researchers. However, these methods are dependent on
the lighting conditions, resolution of the images and the level of the noise present in the
images [12,34,41]. Also, the surfaces of the concrete structures have varying textures as
they are exposed to external disturbances and may not have the same texture even if they
are built from the same material. Hence, they may not give as good results when a new
image having a different texture, brightness, resolution or noise level is given as input.
Furthermore, crack detection in transversal direction is not enough accurate as compared
to measurement carried out longitudinally. this difference in directional measurement
could be an issue when establishing a relationship between the width and longitude of the
crack. Hence, the practical applicability of using image processing-based methods is still
obscure [10,24,35].
On the other hand, machine learning methods also pose several limitations to the
researchers. Increased processing time has been observed in many methods [70]. Many
methods require manual parameter setting of the model which limits the full automation
of the crack detection method [24]. Reliance on GPU is another limitation as some methods
do not work as efficiently without it due to a large number of images in the dataset [17]. To
avoid overfitting of the model it becomes necessary to train the model using a large dataset.
These methods require extensive labelling of data images. In practical scenarios, a limited
option for labelling is also available, so the acquisition of labels can be a difficult task [26].
Due to the difference in surface conditions, a different algorithm may be needed to accu-
rately detect the cracks. In addition, crack detection is carried out offline so performance
in real-time detection is poor. Therefore, there is a need to improve the performance of
algorithms and detection accuracy in real-time. The method selected should be robust as
factors such as climatic conditions may impact crack detection. Limitations in terms of the
need for large datasets in order to train neural network can be overcome by empirically
decomposing the fitted networks into ensembles of low-bias sub-networks, thus making
use of small data sets. In addition, deep learning methods can be applied to unsupervised
tasks, using a small dataset that does not require extensive labelling of data, thus reducing
time and cost [48,50].
The presence of noise, shadows, blemishes and other disturbances in the images is
a problem commonly faced by researchers in using both image processing and machine
learning methods [5,30]. Hence, more research needs to be conducted to develop methods
that can remove noise and other irregularities from images [23,71].

6. Conclusions
The paper focuses on the domains of image processing and machine learning for crack
detection. It reviewed state-of-the-art crack detection methods that have been developed
in the past decade with results are published in top-tier journals and conferences. A
total of 30 research articles were reviewed that were screened after applying criteria and
performing a detailed examination of their content. These articles have been assessed
based on the method they have used, the dataset details, imaging method, performance
outcomes, features and limitations. From the analysis, it can be inferred that a wide range
of articles focuses on crack detection only. However, calculating the dimensions of the
crack was not performed in the majority of the studies. Measurements of the detected
crack such as its length, width, density and depth give important indications regarding the
state of the component and its durability and thus helps in making decisions regarding the
structure’s further usage. More research needs to be focused on crack measurement and the
system must be able to provide a final verdict regarding the severity of the detected crack
using different techniques and algorithms. Most methods showed excellent performance
outcomes as the precision values for crack detection ranged between 75% to 100%. Another
observation is that the researchers prefer using their custom-built dataset, which caters
Infrastructures 2021, 6, 115 18 of 20

to the specific requirements of the method used in the system. Most of the reviewed
crack detection techniques apply to civil infrastructure and concrete components. Overall
and specifically in recent years (2016 to 2020), most of the studies have focused on using
machine learning methods instead of image processing for crack detection. Among these
methods, CNN has been most frequently used for this problem. In the future, the research
can be extended to include assessment criteria to assess the performance such as the
runtime of the algorithm, its resource consumption and applicability in real-time scenarios.

Author Contributions: Conceptualization, H.S.M., A.W.A.H., and S.T.W.; methodology H.S.M.,

A.W.A.H., A.H.; software, H.S.M., A.H.; validation, A.W.A.H., C.A.P.S. and S.T.W.; formal analysis,
H.S.M.; investigation, H.S.M., A.W.A.H., and S.T.W. resources, A.W.A.H., C.A.P.S. and A.H. data
curation, H.S.M.; writing—original draft preparation, H.S.M., A.W.A.H. and A.H. funding acquisition,
A.H. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ni, F.; Zhang, J.; Chen, Z. Pixel level crack delineation in images with convolutional feature fusion. Struct. Control. Health Monit.
2019, 26, e2286. [CrossRef]
2. Liong, S.T.; Gan, Y.S.; Huang, Y.C.; Yuan, C.A.; Chang, H.C. Automatic defect segmentation on leather with deep learning. arXiv
2019, arXiv:1903.12139.
3. Zhang, L.; Yang, F.; Zhang, Y.D.; Zhu, Y.J. Road crack detection using a deep convolutional neural network. In Proceedings of the
International Conference on Image Processing (ICIP), Phoenix, AZ, USA, 25–28 September 2016; pp. 3708–3712.
4. Zou, Q.; Cao, Y.; Li, Q.; Mao, Q.; Wang, S. CrackTree: Automatic crack detection from pavement images. Pattern Recognit. Lett.
2012, 33, 227–238. [CrossRef]
5. Dung, C.V.; Anh, L.D. Autonomous concrete crack detection using deep fully convolutional neural network. Autom. Constr. 2019,
99, 52–58. [CrossRef]
6. Oliveira, H.; Correia, P.L. Automatic road cracks detection and characterization. IEEE Trans. Intell. Transp. Syst. 2012, 14, 155–168.
[CrossRef]
7. Prasanna, P.; Dana, K.J.; Gucunski, N.; Basily, B.B.; La, H.M.; Lim, R.S.; Parvardeh, H. Automated crack detection on concrete
bridges. IEEE Trans. Autom. Sci. Eng. 2014, 13, 591–599. [CrossRef]
8. Gavilán, M.; Balcones, D.; Marcos, O.; Llorca, D.F.; Sotelo, M.A.; Parra, I.; Ocaña, M.; Aliseda, P.; Yarza, P.; Amírola, A. Adaptive
road crack detection system by pavement classification. Sensors 2011, 11, 9628–9657. [CrossRef] [PubMed]
9. Zou, Q.; Zhang, Z.; Li, Q.; Qi, X.; Wang, Q.; Wang, S. Deepcrack: Learning hierarchical convolutional features for crack detection.
IEEE Trans. Image Process. 2018, 28, 1498–1512. [CrossRef] [PubMed]
10. Nishikawa, T.; Yoshida, J.; Sugiyama, T.; Fujino, Y. Concrete crack detection by multiple sequential image filtering. Comput. Civ.
Infrastruct. Eng. 2012, 27, 29–47. [CrossRef]
11. Adeli, H.; Cheng, N. Augmented Lagrangian genetic algorithm for structural optimization. J. Aerosp. Eng. 1994, 7, 104–118.
[CrossRef]
12. Lee, Y.; Wei, C.-H. A computerized feature selection using genetic algorithms to forecast freeway accident duration times. Comput.
Civ. Infrastruct. Eng. 2010, 25, 132–148. [CrossRef]
13. Liu, X.H.; Danczyk, A. Optimal sensor locations for freeway bottleneck identification. Comput. Aided Civ. Infrastruct. Eng. 2009,
24, 535–550. [CrossRef]
14. Pauly, L.; Peel, H.; Luo, S.; Hogg, D.; Fuentes, D.H.A.R. Deeper networks for pavement crack detection. In Proceedings of the
34th ISARC, Taipei, Taiwan, 28 June–1 July 2017; pp. 479–485.
15. Yang, X.; Li, H.; Yu, Y.; Luo, X.; Huang, T.; Yang, X. Automatic pixel-level crack detection and measurement using a fully
convolutional network. Comput. Aided Civ. Infrastruct. Eng. 2018, 33, 1090–1109. [CrossRef]
16. Salman, M.; Mathavan, S.; Kamal, K.; Rahman, M. Pavement crack detection using the Gabor filter. In Proceedings of the 16th
International IEEE Conference on Intelligent Transportation Systems (ITSC 2013), The Hague, The Netherlands, 6–9 October 2013;
pp. 2039–2044.
17. Lins, R.G.; Givigi, S. Automatic crack detection and measurement based on image analysis. IEEE Trans. Instrum. Meas. 2016, 65,
583–590. [CrossRef]
Infrastructures 2021, 6, 115 19 of 20

18. Shi, Y.; Cui, L.; Qi, Z.; Meng, F.; Chen, Z. Automatic road crack detection using random structured forests. IEEE Trans. Intell.
Transp. Syst. 2016, 17, 3434–3445. [CrossRef]
19. Wu, L.; Mokhtari, S.; Nazef, A.; Nam, B.; Yun, H.B. Improvement of crack detection accuracy using a novel crack defragmentation
technique in image-based road assessment. J. Comput. Civ. Eng. 2016, 30, 04014118. [CrossRef]
20. Chen, F.-C.; Jahanshahi, M.R. NB-CNN: Deep learning-based crack detection using convolutional neural network and Naïve
Bayes data fusion. IEEE Trans. Ind. Electron. 2017, 65, 4392–4400. [CrossRef]
21. Ying, L.; Salari, E. Beamlet transform-based technique for pavement crack detection and classification. Comput. Aided Civ.
Infrastruct. Eng. 2010, 25, 572–580. [CrossRef]
22. Landstrom, A.; Thurley, M. Morphology-based crack detection for steel slabs. IEEE J. Sel. Top. Signal Process. 2012, 6, 866–875.
[CrossRef]
23. Zhang, K.; Cheng, H.D.; Zhang, B. A unified approach to pavement crack and sealed crack detection using preclassification based
on transfer learning. J. Comput. Civ. Eng. 2018, 32, 04018001. [CrossRef]
24. Fujita, Y.; Hamamoto, Y. A robust automatic crack detection method from noisy concrete surfaces. Mach. Vis. Appl. 2011, 22,
245–254. [CrossRef]
25. Yang, F.; Zhang, L.; Yu, S.; Prokhorov, D.; Mei, X.; Ling, H. Feature pyramid and hierarchical boosting network for pavement
crack detection. IEEE Trans. Intell. Transp. Syst. 2019, 21, 1525–1535. [CrossRef]
26. Fernández, A.C.; Rodríguez-Lozano, F.J.; Villatoro, R.; Olivares, J.; Palomares, J.M. Efficient pavement crack detection and
classification. EURASIP J. Image Video Process. 2017, 2017, 1–11.
27. Zhang, W.; Zhang, Z.; Qi, D.; Liu, Y. Automatic crack detection and classification method for subway tunnel safety monitoring.
Sensors 2014, 14, 19307–19328. [CrossRef] [PubMed]
28. Yeum, C.M.; Dyke, S.J. Vision—Based automated crack detection for bridge inspection. Comput. Aided Civ. Infrastruct. Eng. 2015,
30, 759–770. [CrossRef]
29. Zhang, A.; Wang, K.C.P.; Li, B.; Yang, E.; Dai, X.; Peng, Y.; Fei, Y.; Liu, Y.; Li, J.Q.; Chen, C. Automated pixel-level pavement crack
detection on 3D asphalt surfaces using a deep-learning network. Comput. Aided Civ. Infrastruct. Eng. 2017, 32, 805–819. [CrossRef]
30. Jahanshahi, M.R.; Masri, S.F. Adaptive vision-based crack detection using 3D scene reconstruction for condition assessment of
structures. Autom. Constr. 2012, 22, 567–576. [CrossRef]
31. Bang, S.; Park, S.; Kim, H.; Kim, H. Encoder-decoder network for pixel-level road crack detection in black-box images. Comput.
Aided Civ. Infrastruct. Eng. 2019, 34, 713–727. [CrossRef]
32. Dorafshan, S.; Thomas, R.J.; Maguire, M. Fatigue crack detection using unmanned aerial systems in fracture critical inspection of
steel bridges. J. Bridge Eng. 2018, 23, 04018078. [CrossRef]
33. Dorafshan, S.; Maguire, M.; Hoffer, N.V.; Coopmans, C. Fatigue crack detection using unmanned aerial systems in under-bridge
inspection. Ida. Transp. Dep. 2017, 1, 1–120.
34. Dorafshan, S.; Thomas, R.J.; Maguire, M. Benchmarking image processing algorithms for unmanned aerial system-assisted crack
detection in concrete structures. Infrastructures 2019, 4, 19. [CrossRef]
35. Kim, I.H.; Jeon, H.; Baek, S.C.; Hong, W.H.; Jung, H.J. Application of crack identification techniques for an aging concrete bridge
inspection using an unmanned aerial vehicle. Sensors 2018, 18, 1881. [CrossRef] [PubMed]
36. Kong, X.; Li, J. Vision—Based fatigue crack detection of steel structures using video feature tracking. Comput. Aided Civ. Infrastruct.
Eng. 2018, 33, 783–799. [CrossRef]
37. Mstafa, R.J.; Younis, Y.M.; Hussein, H.I.; Atto, M. A new video steganography scheme based on Shi-Tomasi corner detector. IEEE
Access 2020, 8, 161825–161837. [CrossRef]
38. Zhu, J.; Ren, M.W.; Yang, Z.J.; Zhao, W. Fast matching algorithm based on corner detection. J. Nanjing Univ. Sci. Technol. 2011, 35,
755–758.
39. Bansal, M.; Kumar, M.; Kumar, M.; Kumar, K. An efficient technique for object recognition using Shi-Tomasi corner detection
algorithm. Soft Comput. 2021, 25, 4423–4432. [CrossRef]
40. Mohan, A.; Poobal, S. Crack detection using image processing: A critical review and analysis. Alex. Eng. J. 2018, 57, 787–798.
[CrossRef]
41. Budiansky, B.; O’connell, R.J. Elastic moduli of a cracked solid. Int. J. Solids Struct. 1976, 12, 81–87. [CrossRef]
42. Aboudi, J. Stiffness reduction of cracked solids. Eng. Fract. Mech. 1987, 26, 637–650. [CrossRef]
43. Dhital, D.; Lee, J.R. A fully non-contact ultrasonic propagation imaging system for closed surface crack evaluation. Exp. Mech.
2012, 52, 1111–1122. [CrossRef]
44. Shan, B.; Zheng, S.; Ou, J. A stereovision-based crack width detection approach for concrete surface assessment. KSCE J. Civ. Eng.
2016, 20, 803–812. [CrossRef]
45. Shan, Q.; Dewhurst, R.J. Surface—Breaking fatigue crack detection using laser ultrasound. Appl. Phys. Lett. 1993, 62, 2649–2651.
[CrossRef]
46. Wang, P.; Huang, H. Comparison analysis on present image-based crack detection methods in concrete structures. In Proceedings
of the 2010 3rd International Congress on Image and Signal Processing (CISP2010), Yantai, China, 16–18 October 2010; Volume 5,
pp. 2530–2533.
47. Koch, C.; Georgieva, K.; Kasireddy, V.; Akinci, B.; Fieguth, P. A review on computer vision based defect detection and condition
assessment of concrete and asphalt civil infrastructure. Adv. Eng. Inform. 2015, 29, 196–210. [CrossRef]
Infrastructures 2021, 6, 115 20 of 20

48. Feng, C.; Liu, M.Y.; Kao, C.C.; Lee, T.Y. Deep active learning for civil infrastructure defect detection and classification. In
Computing in Civil Engineering 2017; American Society of Civil Engineers (ASCE): Reston, VA, USA, 2017; pp. 298–306.
49. Lin, H.; Li, B.; Wang, X.; Shu, Y.; Niu, S. Automated defect inspection of LED chip using deep convolutional neural network. J.
Intell. Manuf. 2019, 30, 2525–2534. [CrossRef]
50. Olson, M.; Wyner, A.J.; Berk, R. Modern neural networks generalize on small data sets. In Proceedings of the 32nd International
Conference on Neural Information Processing Systems 2018; Curran Associates Inc.: Red Hook, NY, USA, 2018; pp. 3623–3632.
51. Huyan, J.; Li, W.; Tighe, S.; Deng, R.; Yan, S. Illumination compensation model with k-means algorithm for detection of pavement
surface cracks with shadow. J. Comput. Civ. Eng. 2020, 34, 04019049. [CrossRef]
52. Lei, B.; Wang, N.; Xu, P.; Song, G. New crack detection method for bridge inspection using UAV incorporating image processing.
J. Aerosp. Eng. 2018, 31, 04018058. [CrossRef]
53. Yoo, H.S.; Kim, Y.S. Development of a crack recognition algorithm from non-routed pavement images using artificial neural
network and binary logistic regression. KSCE J. Civ. Eng. 2016, 20, 1151–1162. [CrossRef]
54. Hoang, N.D.; Nguyen, Q.L.; Tran, X.L. Automatic detection of concrete spalling using piecewise linear stochastic gradient descent
logistic regression and image texture analysis. Complexity 2019, 2019, 5910625. [CrossRef]
55. Zhou, Q.; Qu, Z.; Cao, C. Mixed pooling and richer attention feature fusion for crack detection. Pattern Recognit. Lett. 2021, 145,
96–102. [CrossRef]
56. Wu, X.; Ma, J.; Sun, Y.; Zhao, C.; Basu, A. Multi-scale deep pixel distribution learning for concrete crack detection. In Proceedings
of the 2020 25th International Conference on Pattern Recognition (ICPR) 2021, Milan, Italy, 10–15 January 2021; pp. 6577–6583.
57. Sari, Y.; Prakoso, P.B.; Baskara, A.R. Road crack detection using support vector machine (SVM) and OTSU algorithm. In
Proceedings of the 2019 6th International Conference on Electric Vehicular Technology (ICEVT) 2019, Ungasan, Indonesia, 18–21
November 2019; pp. 349–354.
58. Hasni, H.; Alavi, A.H.; Jiao, P.; Lajnef, N. Detection of fatigue cracking in steel bridge girders: A support vector machine approach.
Arch. Civ. Mech. Eng. 2017, 17, 609–622. [CrossRef]
59. Han, T.; Jiang, D.; Zhao, Q.; Wang, L.; Yin, K. Comparison of random forest, artificial neural networks and support vector machine
for intelligent diagnosis of rotating machinery. Trans. Inst. Meas. Control. 2018, 40, 2681–2693. [CrossRef]
60. Sharma, M.; Anotaipaiboon, W.; Chaiyasarn, K. Concrete crack detection using the integration of convolutional neural network
and support vector machine. Sci. Technol. Asia 2018, 23, 19–28.
61. Noshad, Z.; Javaid, N.; Saba, T.; Wadud, Z.; Saleem, M.Q.; Alzahrani, M.E.; Sheta, O.E. Fault detection in wireless sensor networks
through the random forest classifier. Sensors 2019, 19, 1568. [CrossRef] [PubMed]
62. Saravanan, N.; Siddabattuni, V.K.; Ramachandran, K.I. Fault diagnosis of spur bevel gear box using artificial neural network
(ANN), and proximal support vector machine (PSVM). Appl. Soft Comput. 2010, 10, 344–360. [CrossRef]
63. Kankar, P.K.; Sharma, S.C.; Harsha, S.P. Vibration-based fault diagnosis of a rotor bearing system using artificial neural network
and support vector machine. Int. J. Model. Identif. Control. 2012, 15, 185–198. [CrossRef]
64. Hoang, N.D. An artificial intelligence method for asphalt pavement pothole detection using least squares support vector machine
and neural network with steerable filter-based feature extraction. Adv. Civ. Eng. 2018, 2018, 7419058. [CrossRef]
65. Wang, B.S.; He, Z.C. Crack detection of arch dam using statistical neural network based on the reductions of natural frequencies.
J. Sound Vib. 2007, 302, 1037–1047. [CrossRef]
66. Fan, R.; Bocus, M.J.; Zhu, Y.; Jiao, J.; Wang, L.; Ma, F.; Liu, M. Road crack detection using deep convolutional neural network and
adaptive thresholding. In Proceedings of the 2019 IEEE Intelligent Vehicles Symposium (IV) 2019, Paris, France, 9–12 June 2019;
pp. 474–479.
67. Scholar, P.G. Review and analysis of crack detection and classification techniques based on crack types. Int. J. Appl. Eng. Res.
2018, 13, 6056–6062.
68. Palermo, F.; Konstantinova, J.; Althoefer, K.; Poslad, S.; Farkhatdinov, I. Implementing tactile and proximity sensing for crack
detection. In Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA) 2020, Paris, France,
31 May–31 August 2020; pp. 632–637.
69. Luo, Q.; Ge, B.; Tian, Q. A fast adaptive crack detection algorithm based on a double-edge extraction operator of FSM. Constr.
Build. Mater. 2019, 204, 244–254. [CrossRef]
70. Sheng, P.; Chen, L.; Tian, J. Learning-based road crack detection using gradient boost decision tree. In Proceedings of the 2018 13th
IEEE Conference on Industrial Electronics and Applications (ICIEA) 2018, Wuhan, China, 31 May–2 June 2018; pp. 1228–1232.
71. Bhat, S.; Naik, S.; Gaonkar, M.; Sawant, P.; Aswale, S.; Shetgaonkar, P. A survey on road crack detection techniques. In Proceedings
of the 2020 International Conference on Emerging Trends in Information Technology and Engineering (Ic-ETITE) 2020, Vellore,
India, 24–25 February 2020; pp. 1–6.