The Journal of Supercomputing (2022) 78:17721–17735

[Link]

Assessing severity of road cracks using deep

learning‑based segmentation and detection

Jongwoo Ha1 · Dongsoo Kim2 · Minsoo Kim1

Accepted: 23 April 2022 / Published online: 22 May 2022

© The Author(s) 2022

Abstract
For the successful operation of pavement management system, it is necessary to
automate the detection, classification, and severity assessment of road cracks, which
are bottlenecks in the entire process. Although good results for the detection and
classification of road cracks have been published in many related studies, the num-
ber of crack types detected is still insufficient for actual field use, and studies on
crack severity assessment are difficult to find. In this study, the number of crack
types are expanded to five types (alligator crack, longitudinal crack, transverse crack,
pothole, and patching) to meet the needs of fieldwork, and the assessment of crack
severity is also included in the proposed model. In this system constructed using
SqueezeNet, U-Net, and Mobilenet-SSD models together, an accuracy of 91.2% has
been achieved for both crack type and severity assessment. The authors have per-
formed segmentation of the input images using separately trained U-Nets for linear
cracking and area cracking to improve object detection performance and automate
crack severity assessment. With reference to the system presented in this study, it is
expected that an automated pavement management system that better reflects each
country’s requirements for various crack types and severity standards is possible.

Keywords Crack segmentation · Crack severity · Object detection · Pavement

management · Road crack detection

* Minsoo Kim
minsky@[Link]
1
Major in Industrial Data Science and Engineering, Department of Industrial and Data
Engineering, Pukyong National University, Yongso‑ro 45, Nam‑Gu, Busan 48513, South Korea
2
Department of Industrial and Information Systems Engineering, Soongsil University, 369
Sangdo‑ro, Dongjak‑gu, Seoul 06978, South Korea

13
Vol.:(0123456789)
17722 J. Ha et al.

1 Introduction

Pavement management is the process of planning maintenance and repair of roads to

retain the road networks in an optimal state in terms of life cycle costs. Since pavement
management requires complex decisions involving resource allocation and work sched-
uling, a software tool called a pavement management system (PMS) [33] is used to
support these decisions. The key to successful pavement management is to detect and
repair defects on roads in a timely manner so that pavement deterioration does not cause
inconveniences to safe driving. In large cities with high traffic, fast crack detection and
maintenance becomes even more important because the number of vehicles exposed to
cracks is higher and the severity of defects rapidly aggravates with heavy traffic. When
there are too many roads to manage with limited resources, quantified decision-making
based on risk assessment is required to handle these situations effectively. Determining
the severity of road cracks is very important for a proper risk assessment. To achieve
an effective risk assessment in PMS, the severity assessment should be accompanied by
the crack type detection because different types of road cracks have different severity
assessment methods. As summarized in Fig. 1, pavement management basically con-
sists of 4 phases. During the data collection phase, massive images of road surfaces
are continuously captured from patrol vehicles via road scanner or camera equipment,
along with location information (GPS data) for that section of the road. In the case of
Seoul city, the image capacity required to scan and store 8,323.7 km of paved roads at
1-m interval in 2020 is about 370 TB. This data capacity easily doubles as the number
of lanes and the number of scans per year increases. Data collection is usually done
automatically in most countries using dedicated hardware. During the crack detection
phase, the crack type is determined along with the presence or absence of cracks in the
images. Although many recent studies have reported achievements related to automated
crack detection [2, 32], to the best of our knowledge, these results have not yet been
applied to PMS and appear to be still in the research phase. The crack detection phase
of Korean PMS also still relies heavily on visual inspections performed by many opera-
tors, making it a bottleneck process that causes significant delays in the overall process.
Lots of operators are classifying crack types in road images and manually mark crack

Fig. 1  The main process of road pavement management

13
Assessing severity of road cracks using deep learning‑based… 17723

pixels. However, recent advances in crack detection research are moving this phase
from the conventional manual process to a semi-automatic process by country. In the
severity assessment phase, manually marked pixel data is used to evaluate the indi-
vidual severity for cracks and the overall severity of that road segment is determined.
Finally, in the maintenance planning phase, repair schedule for road sections with high-
severity cracks is generated by reflecting available resources [8].
To accelerate the entire pavement management process, it is necessary to expe-
dite the crack detection and severity assessment, which are bottlenecks that are
highly dependent on human resources. Recently, AI-related research to assist or
replace human tasks has been actively conducted in various industries. In particu-
lar, research related to deep neural networks (DNNs) using GPU’s high-performance
computing power has shown good results and is expanding the field of application.
In relation to the detection and classification of road cracks, studies using DNNs
to determine the presence of cracks or to classify crack types have been actively
conducted, and results with significantly improved accuracy are being published [1,
6, 7, 13]. However, since many studies have been limited to classifying a relatively
small number of crack types, it is necessary to expand the number of classifiable
crack types for actual field application. In addition, research on automatic estima-
tion of crack severity is still in its early stages, making it difficult to fully automate
the process within PMS. To automate the estimation of crack severity, it is neces-
sary to accurately confine the crack region from the road image and calculate the
degree of deformation according to the crack type. In this paper, the authors focus
on implementing practical applications for automating crack detection and sever-
ity assessment in PMS while adopting and extending existing research. The authors
have expanded the types of cracks that can be classified into five categories: alligator
crack (AC), longitudinal crack (LC), transverse crack (TC), pothole, and patching
so that it can be used in actual fields. They also have built a pilot system that can
determine the severity of cracks by identifying crack regions through object detec-
tion. In this pilot system, DNN-based image segmentation is performed to clearly
enclose crack regions from the input image for severity assessment, and crack clas-
sification is performed before segmentation to improve segmentation accuracy by
reflecting the characteristics of each crack type. The overall accuracy of pilot system
has reached to 91.2% for 1330 test images, which appears to be applicable to actual
fieldwork. The content of the paper is organized as follows: Sect. 2 introduces back-
ground knowledge and prior research related to this study. Section 3 describes the
model details and experimental results of the proposed approach to assess the sever-
ity of road cracks. Overall structure of the pilot system for analyzing road cracks is
presented in Sect. 4. Finally, the conclusion in Sect. 5 briefly reviews the contribu-
tions of this study and presents suggestions for further research.

2 Related works

The configuration and relative frequency of cracks in asphalt roads can vary depend-
ing on several factors, such as the root cause, the surrounding climate, and the road
usage pattern. For this reason, many countries want to customize the types and

13
17724 J. Ha et al.

severity measure of road cracks to better reflect their own situation, and to manage
them systematically. In Korea, as shown in Fig. 2, the top five crack types that are
most frequent and require strict management with importance are identified [13, 32].
Alligator crack or crocodile crack is a common type of distress in asphalt pavement,
which is characterized by interconnecting or interlaced cracking in the asphalt layer
like the patterns in crocodile hide. Longitudinal crack is a form of distress whose
direction is typically parallel to the edge of the pavement shoulder. Transverse
crack occurs roughly perpendicular to the centerline of the pavement, mainly due to
shrinkage of the asphalt layer or reflection from an existing crack. A pothole refers
to a hole of various sizes and shapes that occurs when the weak spot in the asphalt
layer collapses or is displaced by the weight of a passing vehicle. The term patching
refers to the process of filling potholes or overlaying excavated areas in the asphalt
pavement. Though patching is not an actual asphalt distress, it is managed as an
important crack type in Korea because there is a high possibility that the underlying
crack is easily exposed due to wear and tear in case of thin surface patch.
Road crack analysis initially identifies the type and location of cracks from the
input image, and the severity is determined by measuring the maximum width of the
crack or by calculating the area of the distress according to the type of crack. Many
existing studies on road crack detection using DNN still focus only on the classifica-
tion of crack types and do not cover all five types mentioned above [8, 12, 16]. In
particular, research on patch detection is difficult to find, so additional research is
needed in Korea, where patching is to be managed as a major crack type. Unlike
previous studies that use a separate analysis algorithm to identify crack segments

Fig. 2  Five types of road cracks [32]

13
Assessing severity of road cracks using deep learning‑based… 17725

while evaluating the severity of each crack type [1, 2, 10], this study uses an object
detection technique to simultaneously handle the classification of crack types and
the confinement of crack regions.

2.1 Crack detection using deep learning

There have been many studies trying to detect cracks in objects such as concrete
walls, bridges, pipelines, glass and asphalt pavements by combining image process-
ing techniques and deep learning [2–4, 6, 8, 10–13, 24–27, 30, 34]. Basically, crack
detection is performed according to three main steps: pre-processing, detection and
classification. In these studies, images were pre-processed by applying conventional
image processing techniques such as smoothing, normalization and filtering meth-
ods. In detection phase, the existence of crack is determined by applying analytical
or logical methods such as Otsu [28] methods, statistical approaches, and threshold
methods. Many of road crack detection researches can be assigned into this category
that actually determines the existence of cracks by incorporating various segmenta-
tion techniques to improve image quality or generates crack regions from the input
image in the form of bounding boxes [35–37, 39], but they does not actually deter-
mine the individual type of cracks. There are researches that utilize additional types
of input data such as acoustic-sensor data [38] or 3D scanned data [40] to better
detect cracks hidden below the surface. Finally, the actual multi-type classification
is done using deep learning methods like convolutional neural networks (CNNs)
or using various mathematical techniques. Approaches using analytical or logical
methodologies are generally fast in processing, but have an accuracy of about 80 to
90%, which is somewhat insufficient for practical use [20].
In the case of CNN approaches, detection is usually performed together with clas-
sification rather than a separate step, or a segmentation step to find crack regions or
contours in the input image is often included instead of the detection step. Although
CNN approaches require more computational resources than analytical or logical
methods, they show improved accuracy of more than 90% through the development
of new network models and continuous learning on the accumulated data. Studies
on the detection of cracks on asphalt pavements have also shown a similar trend
to the other fields of research on cracks [3, 18]. Zou et al. proposed a deep convo-
lutional neural network (DCNN) called DeepCrack based on the encoder-decoder
architecture employing hierarchical multi-scale features for automatic crack detec-
tion in pavement and stone surface images. It is reported that DeepCrack achieves
F-measure over 0.87 on the test dataset [21]. Feng, X et al. proposed a method based
on a DCNN fusion model, which combines the advantages of the multitarget single-
shot multibox detector (SSD) CNN model and the U-Net model.
Segmentation and crack type classification is carried out sequentially in this
model. Test results for this fusion model show that the recognition accuracy of the
pavement crack for TC, LC, and AC is 86.8%, 87.6%, and 85.5%, respectively [22].
Although this model performs crack detection and classification for three types with
relatively high accuracy, it has the disadvantage requiring substantial computational
capacity due to a large model using many parameters. Hu G.X. et al. conducted

13
17726 J. Ha et al.

several experiments applying a set of YOLOv5 object detection models for pave-
ment crack detection. In their experiments, the YOLOv5l model recorded the high-
est detection accuracy of 88.1% and the YOLOv5s model recorded the shortest
detection time of 11.1 ms for each image [23].
In terms of road pavement management, if the existing studies are simply grouped
according to the functions they are dealing with, most of them can be classified into
three types: ‘detection—classification’, ‘segmentation—classification’, and ‘detec-
tion—severity assessment’. As mentioned earlier, there are many studies focusing on
classification, and few studies on severity assessment [29, 31]. Implementing PMS,
however, requires handling both classification and severity assessment for all major
crack types, including pothole and patching.

2.2 Road crack classification and severity assessment

To quickly implement classification and severity assessment together in PMS, it

is practical to use a combination of methods proposed in previous studies. This
approach, however, has a drawback in that the size of model increases due to the
combined use of existing networks. Because on-site reprocessing is required to
validate the PMS results, the size of the model must be considered so that it can
operate even with limited computing resources of the mobile terminals. Jo, H.
et al. published an experiment comparing the performance of two self-designed
CNNs and SqueezeNet in the process of classifying crack severity into high,
medium, and low for 7 crack types. As is well known, SqueezeNet is useful for
applications with memory or computational limitations and shows good perfor-
mance while achieving AlexNet-level accuracy on ImageNet dataset. Although
SqueezeNet is a CNN designed to have a small number of parameters, it showed
relatively high accuracy in classifying the severity of road cracks in their experi-
ment [6, 19]. Ha J. et al. conducted various experiments on segmentation and
classification to detect road cracks and proposed a network using U-Net and
Mobilenet-SSD together [32]. U-Net is a CNN developed for biomedical image
processing to work with fewer training images and to yield segmentations that are
more precise. Because U-Net adopts a patch method that splits the entire image
into grid tiles and processes them separately, it shows faster processing speed
than the conventional sliding window method that frequently recalculates the
overlapped window area. In particular, the authors believed that U-Net’s ability
to segment cell structures well would be suitable for segmenting lattice-shaped
cracks such as alligator cracks [14, 32]. The authors conducted object detection
using an SSD model to confine the crack region and assess the severity. SSD is
one of the representative object detection model that uses a single-stage approach
to detect multiple objects in an input image, as opposed to a two-stage model that
uses a regional proposal mechanism such as R-CNN and Fast(er)-R-CNN. SSD
is known for its speed and simplicity, which uses feature maps of different sizes
to provide high detection performance while minimizing the impact of changes
in the size of objects in the image. In their experiment, the SSD300 model
employing Mobilenet v1 as the backbone is used to achieve fast operation while

13
Assessing severity of road cracks using deep learning‑based… 17727

Fig. 3  Two networks trained on different image datasets

Table 1  Performance ` Original image Mask image

comparison of two networks
trained on different image Class AP mAP AP mAP
datasets
Alligator crack 0.4822 0.6818 0.8202 0.9382
Longitudinal crack 0.6858 0.9188
Transverse crack 0.6545 0.9719
Patching 0.7123 0.9895
Pothole 0.8742 0.9906

maintaining good performance even with limited computing resources [5, 9, 15].
It is necessary to briefly review the experiments they conducted in that this study
is an extension of their study so that a severity assessment can be done. Follow-
ing Fig. 3 outlines two experiments conducted by Ha J. et al. They compared the
performance of two Mobilenet-SSD networks trained separately on the original
and the mask images to detect five types of cracks. The results of the network
trained on both images are summarized in Table 1 briefly. The mAP (mean Aver-
age Precision) values obtained from the two networks were 0.6818 and 0.9382,
respectively. The network trained on the original image did not properly identify
the crack types, but the network trained on the mask image was able to identify
them with very high accuracy. This result indicates that masking of images is
very important for crack identification using object detection.
Thus, a good segmentation method is needed to obtain an mask image automati-
cally from the original image without manual intervention. For this reason, Ha J.
et al. added U-Net, FPHBN [17], and FPN networks for crack segmentation before
object detection network and compared their performance. However, even in U-Net,
which showed the best result among the three, the value of mIoU (mean Intersection
over Union) was only about 0.4256, so the crack could not be properly segmented.
To improve crack segmentation performance, another experiment was performed
with the configuration shown in Fig. 4 below, using different segmentation networks
for linear cracking (AC, LC, and TC) and area cracking (patching and porthole).

13
17728 J. Ha et al.

Fig. 4  Separated segmentation networks for object-detection

Table 2  Test result of separated Class mIoU mAP

segmentation approach
AC, LC, and TC 0.8650 0.9315
Pothole, patching 0.7920

Significant performance improvements were achieved in this new configuration, and

the results are summarized in Table 2 below.
The study of Ha et al., however, still needs to add severity assessment for the
implementation of automated PMS. In addition, to classify the input images fed to
the two U-Nets that perform segmentation by crack type, we need another classifier
in front of the U-Nets. The remainder of this paper deals with the extension of the
above network and its experimental results.

3 Extensions for assessing crack severity

The severity criteria for road cracks can vary from country to country, but in gen-
eral, depending on the type of crack, the maximum crack width or the relative ratio
of the damaged area is calculated and classified into three levels: low, medium, and
high. High severity cracks can lead to dangerous situations, so it’s very important for
PMS to detect and repair them quickly. Table 3 summarizes the severity assessment
criteria used in this study, which was made with the help of construction expert from
South Korea by referring previous studies [13]. The authors applied the criteria in
Table 3 to determine crack severity for the input image representing an area 0.6 m
wide and 1.06 m long.

3.1 Severity of linear cracking

The severity of linear cracking such as AC, TC, and LC is assessed based on the
maximum width or thickness of the crack line. The actual size of a pixel in the
input image is calculated from the size of the captured area easily. In this study,
since an area of 0.6 m in width and 1.06 m in height is captured as a 224 × 224

13
Assessing severity of road cracks using deep learning‑based… 17729

Table 3  Criteria for assessment of road crack severity

Class Measurement (X) Severity Limits

Linear cracking Max-width (mm) Low X < 10 mm

Medium 10 mm ≤ X < 75 mm
High 75 mm ≤ X
Area cracking Area (%) Low X < 10%
Medium 10% ≤ X < 25%
High 25% ≤ X

image, 1 pixel is about 2.68 mm in width and 4.73 mm in height. Crack severity
is estimated as high, medium, and low based on the maximum crack width that is
calculated from the segmented image. The accuracy of the model is measured by
comparing this estimated severity with the severity obtained from the manually-
created mask image. Figure 5 is a brief summary of the severity estimation pro-
cess for linear cracking used in this study.
The accuracy of the severity estimation for the 4383 linear cracking images—
AC: 1755, LC: 1123, TC: 1505—achieved 94.39%. This high-accuracy result
shows that automatically segmented images can be practically utilized for crack
severity estimation instead of manually-created mask images. Table 4 summa-
rizes the severity estimation results for linear cracking.

Fig. 5  Severity estimation for linear cracking

13
17730 J. Ha et al.

Table 4  Result of severity estimation for linear cracks

Class Severity # total images # Correct images Accuracy (%)

Linear cracking Low 1423 1297 94.39

Medium 2029 1942
High 931 898

Fig. 6  Severity estimation for area cracking

3.2 Severity of area cracking

The severity of area cracking such as pothole and patching is assessed by calcu-
lating the proportion of distress region to the whole image. The crack severity of
area cracking is also estimated as high, medium, and low for segmented images,
just like linear cracking. The accuracy is again measured by comparing this esti-
mated severity for the segmented image with the severity obtained from the man-
ually-created mask image. Figure 6 outlines the severity estimation process for
area cracking used in this study.
In this study, the severity is calculated as the ratio of the crack area to the total
image area, but it can also be defined as the ratio of the crack area to the bound-
ing box that is found through object detection. In such cases, the metric values
that define the severity level should be adjusted accordingly. The accuracy of the
severity estimation for the 2267 area cracking images—pothole: 1562, patching:
705—recorded 89.68%. Compared to linear cracking, the accuracy is relatively

13
Assessing severity of road cracks using deep learning‑based… 17731

Table 5  Result of severity estimation for area cracking

Class Severity # Total images # Correct images Accuracy (%)

Area cracking Low 483 452 89.68

Medium 1009 862
High 775 719

Fig. 7  Automatic crack analysis system

low, but it is still sufficient for practical use. Table 5 summarizes the severity esti-
mation results for area cracking.

4 Automatic crack analysis for PMS

Through our experiments, it was confirmed that the severity estimation could be
performed with high accuracy even with the use of automatically segmented images.
For the crack image to be properly segmented, however, the input image must be
pre-classified into linear and area cracking and fed to each segmentation network.
That is, a classifier that determines the type of input image as a linear or area crack-
ing from the start is required for severity assessment in PMS. In this study, the
SqueezeNet-based classifier trained in the previous study was used [6, 19]. Figure 7
shows the overall architecture of proposed system for automatic crack analysis.
For the training of SqueezeNet and U-Net, the following sets of parameters sum-
marized in Table 6 have been used. These values are selected from the best training
result over the random sampled values within the search range.
The SqueezeNet used for classification was trained with 5320 images that
account for 80% of dataset from the total 6650 images. It discriminates whether
the input image is a linear or area cracking with an accuracy of 99.6%. The
image classified by the SqueezeNet is fed to dedicated U-Net that is responsible
for segmentation of linear cracking or area cracking to generate a black & white
segmented image. This segmented image is fed again to Mobilenet-SSD, which
performs object detection, and the detailed crack types and crack regions are
determined. Finally, the detailed crack types and crack regions are used together
with the segmented image to evaluate the crack severity. Table 7 shows the final

13
17732 J. Ha et al.

Table 6  Parameter values for experiment

Model Epoch Batch size Weight decay Learning rate Learning Momentum
rate decay

SqueezeNet 100 64 0.0002 0.01 0.1 0.9

U-Net 200 16 0.0001 0.001 – 0.9

accuracy of our automatic crack analysis system. For this result, fivefold cross
validation has been carried out, and the average accuracy is given in Table 7.
The final accuracy represents the percentage of accurate predictions on both
the crack type and severity of cracks tested on 1330 test images that account for
20% of total images left for testing. For linear cracking, testing with 858 images
showed that both crack type and crack severity could be determined with 93.27%
accuracy. For area cracking, the accuracy of 87.43% is recorded for 472 images.
In total, suggested crack analysis system achieved 91.2% accuracy for both crack
type classification and crack severity assessment. Since the accuracy of classifica-
tion done by the SqueezeNet is about 99.6%, the remaining 0.4% of test images
are misclassified in types, and fed to wrong U-Net for segmentation. Out of those
5 misclassified test images, actually 4 images which account for 80% of misclas-
sified test images are correctly reclassified into their original crack types at the
object detection stage done by Mobilenet-SSD. If an image is initially misclassi-
fied, it is possible that the segmentation result will be of poor quality, leading to
the wrong severity level. It seems likely that the more test data will be needed to
better evaluate the detailed performance of our proposed system, especially with
respect to the impact of misclassification on severity assessment.
In this system, a classifier using SqueezeNet was used to classify the input
into linear cracks and area cracks before segmentation, but after segmentation,
detailed classification of crack types using Mobilenet-SSD is being performed
again. This somewhat overlapping function of the current system needs to be fur-
ther optimized through continuous research and the introduction of new networks
that segment and classify cracks altogether. Until then, it seems practical to use
existing networks in combination to satisfy field needs.

Table 7  Overall performance of Class Accuracy (%) Overall accuracy (%)

crack analysis system
Linear cracking 93.27 91.2
(858 test images)
Area cracking (472 87.43
test images)

13
Assessing severity of road cracks using deep learning‑based… 17733

5 Conclusions

In this study, the existing research has been advanced to enable the realization of
PMS by expanding the crack types into five types and performing crack severity
assessment with high accuracy through automated segmentation. The contribu-
tion of this study is significant in that it can perform crack detection, classifica-
tion, and severity assessment in one system with high accuracy by synthesizing
studies related to road cracks that have not been effectively combined so far. To
configure the whole system, SqueezeNet, U-Net, and Mobilenet-SSD have been
combined. By using two U-Nets to separate segmentations for linear cracking
and area cracking, the accuracy of crack severity assessment has been improved
to 94.39% for linear cracking and 89.68% for area cracking. The final system
achieved an accuracy of 91.2% for both the assessment of crack severity and the
classification of crack type.
The data used for training and testing in this study are 2D images without any
depth information. If we can augment the data with depth information, then we
can further enhance our result to better assess the severity of cracks like patching
and pothole. Although the authors vaguely consider using pixel darkness to infer
crack depth from the 2D images, this can be a challenging task because the light-
ing conditions of captured image greatly affects to the brightness of the whole
image. With the introduction of 3D road scanner, it is expected that a higher level
of accuracy can be achieved by utilizing the depth information from the captured
3D model of the road pavement in the near future. The authors put a lot of effort
to use small and efficient networks, but the use of many network modules inevi-
tably has led to a large system. In particular, since the operation of the classifica-
tion network added to increase the segmentation performance overlaps with the
identification of crack types in the object detection network, an in-depth study
should be conducted to reduce the size of the overall system.

Acknowledgements This work was supported by the Pukyong National University Research Fund in
2019.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as
you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com-
mons licence, and indicate if changes were made. The images or other third party material in this article
are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is
not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission
directly from the copyright holder. To view a copy of this licence, visit [Link]
ses/​by/4.​0/.

13
17734 J. Ha et al.

References
1. Ashraf S, Hegazy I, Elarif TL (2019) Algorithm for automatic crack analysis and severity iden-
tification. In: 2019 IEEE Ninth International Conference on Intelligent Computing and Informa-
tion Systems, pp. 74–79
2. Cubero-Fernandez A, Rodriguez-Lozano FJ, Villatoro R, Olivares J, Palomares JM (2017) Effi-
cient pavement crack detection and classification. EURASIP J Image Video Process 39:1–11
3. Feng C, Liu MY, Kao CC, Lee TY (2017) Deep active learning for civil infrastructure defect
detection and classification. Comput Civ Eng 298–306
4. Gopalakrishnan K (2018) Deep learning in data-driven pavement image analysis and automated
distress detection: a review. Data 3(3):1–19
5. Howard AG, Zhu M, Chen B, Kalenichenko D, Wang W, Weyand T, Adam H (2017) Mobilen-
ets: efficient convolutional neural networks for mobile vision applications. arXiv preprint
arXiv:1704. 04861
6. Jo H, Kim D, Pak KW, Kim M (2020) Road damage detection over road scanner images using
deep convolutional neural network. ICIC Express Lett 14(10):1001–1008
7. Kirillov A, Girshick R, He K, Dollar P (2019) Panoptic feature pyramid networks. In: Proceed-
ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp
6399–6408
8. Koch C, Brilakis I (2011) Pothole detection in asphalt pavement images. Adv Eng Inf
25(3):507–515
9. Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu CY, Berg AC (2016) Ssd: single shot
multibox detector. In: European Conference on Computer Vision, Springer, Cham, pp 21–37
10. Naddaf-Sh M, Hosseini S, Zhang J, Brake NA, Zargarzadeh H (2019) Realtime road crack mapping
using an optimized convolutional neural network. Complexity, pp 1–17
11. Pauly L, Hogg D, Fuentes R, Peel H (2017) Deeper networks for pavement crack detec-
tion. In: Proceedings of the 34th International Symposium on Automation and Robotics in
Construction(ISARC), IAARC, pp 479–485
12. Rababaah H, Vrajitoru D, Wolfer J (2005) Asphalt pavement crack classification: a comparison of
GA, MLP, and SOM. In: Proceedings of Genetic and Evolutionary Computation Conference, Late-
Breaking Paper
13. Ragnoli A, De Blasiis MR, Benedetto AD (2018) Pavement distress detection methods: a review.
MDPI Infrastruct 3(58):1–19
14. Ronneberger O, Fischer P, Brox T (2015) U-net: CONVOLUTIONAL networks for biomedical
image segmentation. In: International Conference on Medical Image Computing and Computer-
Assisted Intervention. Springer, Cham, pp 234–241
15. Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC (2018) Mobilenetv2: inverted residuals and
linear bottlenecks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recog-
nition, pp 4510–4520
16. Sorncharean S, Phiphobmongkol S (2008) Crack detection on asphalt surface image using enhanced
grid cell analysis. In: 4th IEEE International Symposium on Electronic Design, Test and Applica-
tions, pp 49–54
17. Yang F, Zhang L, Yu S, Prokhorov D, Mei X, Ling H (2020) Feature pyramid and hierarchical
boosting network for pavement crack detection. IEEE Trans Intell Transp Syst 21(4):1525–1535
18. Zhang L, Yang F, Zhang YD, Zhu YJ (2016) Road crack detection using deep convolutional neural
network. In: Proceedings of the 2016 IEEE International Conference on Image Processing(ICIP),
Phoenix, AZ, USA, 25–28 September 2016, pp 3708–3712
19. Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K (2016) SqueezeNet: AlexNet-
level accuracy with 50x fewer parameters and< 0.5 MB model size. arXiv preprint arXiv:​1602.​
07360
20. Geethalakshmi SN (2018) A survey on crack detection using image processing techniques and deep
learning algorithms. Int J Pure Appl Math 118(8):215–220
21. Zou Q, Zhang Z, Li Q, Qi X, Wang Q, Wang S (2018) Deepcrack: learning hierarchical convolu-
tional features for crack detection. IEEE Trans Image Process 28(3):1498–1512
22. Feng X, Xiao L, Li W, Pei L, Sun Z, Ma Z, Ju H (2020) Pavement crack detection and segmentation
method based on improved deep learning fusion model. Math Proble Eng

13
Assessing severity of road cracks using deep learning‑based… 17735

23. Hu GX, Hu BL, Yang Z, Huang L, Li P (2021) Pavement crack detection method based on deep
learning models. Wirel Commun Mobile Comput
24. Baohua S, ShijeZheng JO (2016) A stereovision-based crack width detection approach for concrete
surface assessment. KSCE J Civ Eng 20(2):803–812
25. Shivprakashlyer SK (2005) Sinha, A robust approach for automatic detection and segmentation of
cracks in underground pipeline images. Image Vis Comput 23 (1):931–933
26. Yang Y-S, Yang C-M, Huang C-W (2015) Thin crack observation in a reinforced concrete bridge
pier test using image processing and analysis. Adv Eng Softw 83:99–108
27. Sinha SK, Fieguth PW (2006) Automated detection of cracks in buried concrete pipe images.
Autom Constr 15(1):58–72
28. Talab AM, Huang Z, Xi F, HaiMing L (2016) Detection crack in image using Otsu method and mul-
tiple filtering in image processing techniques. Optik 127(3):1030–1033
29. Tran TS, Tran VP, Lee HJ, Flores JM, Le VP (2020) A two-step sequential automated crack detec-
tion and severity classification process for asphalt pavements. Int J Pav Eng. [Link]
10298​436.​2020.​18365​61
30. Yusof NAM, Osman MK, Noor MHM, Ibrahim A, Tahir NM, Yusof NM (2018) Crack detection
and classification in asphalt pavement images using deep convolution neural network. In: 2018
9th IEEE International Conference on Control System, Computing and Engineering, 23–25, 2018
November
31. Zumrawi MME (2015), Survey and evaluation of flexible pavement failures. Int J Sci Res
32. Ha, J., Park, K., Kim, M, A Development of Road Crack Detection System Using Deep Learning-
based Segmentation and Object Detection, The Journal of Society for e-Business Studies, Vol. 26,
No. 1, pp.93–106, 202.
33. Marcelino P, Antunes ML, Fortunato E (2019) Current international practices on pavement condi-
tion assessment. In: Pavement and Asset Management (pp. 359–363). CRC Press
34. Liang S (2021) Feature extraction of broken glass cracks in road traffic accident site based on deep
learning. Complexity, pp12 (2021)
35. Nguyen NHT, Perry S, Bone D, Le HT, Nguyen TT (2021) Two-stage convolutional neural network
for road crack detection and segmentation. Expert Syst Appl, 186
36. Aravindkumar S, Varalakshmi P (2021) Automatic road crack detection and classification using
multi-tasking faster RCNN. J Intell Fuzzy Syst 41(6):6615–6628
37. Kyslytsyna A, Xia K, Kislitsyn A, Kader IAE, Wu Y (2021) Road surface crack detection method
based on conditional generative adversarial networks. Sensors 21(21):7405. [Link]
s2121​7405
38. Pratico FG, Fedele R, Naumov V, Sauer T (2020) Detection and monitoring of bottom-up cracks in
road pavement using a machine-learning approach. Algorithms. [Link]
39. Wu S, Fang J, Zheng X, Li X (2019) Sample and structure-guided network for road crack detection.
IEEE Access 7:130032–130043
40. Li Q, Yao M, Yao X, Xu B (2010) A real-time 3D scanning system for pavement distortion inspec-
tion. Meas Sci Technol 21(1)

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.

13