Article

Feasibility of Advanced Reflective Cracking Prediction and

Detection for Pavement Management Systems Using Machine
Learning and Image Detection
Sung-Pil Shin 1, Kyungnam Kim 2,* and Tri Ho Minh Le 3,*

1 Department of Highway & Transportation Research, Korea Institute of Civil Engineering and Building
Technology, 283 Goyandae-ro, Ilsanseo-gu, Goyang-si 10233, Republic of Korea; [email protected]
2 Pavement Research Division, Korea Expressway Corporation Research Institute, Dong-tansunhwan-daero

17-gil, Hwaseong-si 18489, Republic of Korea

3 Faculty of Civil Engineering, Nguyen Tat Thanh University, 300A Nguyen Tat Thanh Street, District 4,

Ho Chi Minh City 70000, Vietnam

* Correspondence: [email protected] (K.K.); lhmtri@n .edu.vn (T.H.M.L.)

Abstract: This research manuscript presents a comprehensive investigation into the prediction and
detection of reflective cracking in pavement infrastructure through a combination of machine learn-
ing approaches and advanced image detection techniques. Leveraging machine learning algorithms,
reflective cracking prediction models were developed and optimized for accuracy and efficiency.
Additionally, the efficacy of image detection methods, particularly utilizing Mask R-CNN, was ex-
plored for robust and precise identification of reflective cracking on pavement surfaces. The study
not only aims to enhance the predictive capabilities of pavement management systems (PMSs)
through machine learning-based models but also seeks to integrate advanced image detection tech-
nologies to support real-time monitoring and assessment of pavement conditions. By providing ac-
curate and timely detection of reflective cracking, these methodologies contribute to the optimiza-
tion of pavement maintenance strategies and the overall improvement of pavement infrastructure
Citation: Shin, S.-P.; Kim, K.; Le, management practices. Results indicate that the developed machine learning models achieve an
T.H.M. Feasibility of Advanced average predictive accuracy of over 85%, with some models achieving accuracies exceeding 90%.
Reflective Cracking Prediction and Moreover, the utilization of a mask region-based convolutional neural network (Mask R-CNN) for
Detection for Pavement image detection demonstrates exceptional precision, with a detection accuracy of over 95% on av-
Management Systems Using erage across different pavement types and weather conditions. The results demonstrate the prom-
Machine Learning and Image ising performance of the developed machine learning models in predicting reflective cracking,
Detection. Buildings 2024, 14, 1808.
while the utilization of Mask R-CNN showcases exceptional accuracy in the detection of reflective
h ps://doi.org/10.3390/
cracking from images. This research underscores the importance of leveraging cu ing-edge tech-
buildings14061808
nologies to address challenges in pavement infrastructure management, ultimately supporting the
Academic Editor: Pengfei Liu sustainability and longevity of transportation networks.
Received: 10 May 2024
Revised: 2 June 2024
Keywords: reflective cracking prediction; machine learning algorithms; image detection techniques;
Accepted: 13 June 2024 mask R-CNN; pavement management systems
Published: 14 June 2024

Copyright: © 2024 by the authors.

1. Introduction
Licensee MDPI, Basel, Swi erland. The construction and maintenance of road infrastructure are vital for facilitating eco-
This article is an open access article nomic development and ensuring safe transportation systems [1]. However, roads are
distributed under the terms and susceptible to various environmental factors and climatic conditions that can lead to pave-
conditions of the Creative Commons ment distress [2], including reflective cracking. Reflective cracking, a prevalent form of
A ribution (CC BY) license
pavement distress, poses significant challenges to the durability and performance of road
(h ps://creativecommons.org/license
infrastructure [3]. These cracks typically originate from underlying cracks or joints in the
s/by/4.0/).
pavement structure and propagate upwards [4], often exacerbated by factors such as

Buildings 2024, 14, 1808. https://doi.org/10.3390/buildings14061808 www.mdpi.com/journal/buildings

Buildings 2024, 14, 1808 2 of 24

temperature fluctuations, traffic loading, and moisture infiltration [5]. The formation of
reflective cracks is a complex phenomenon influenced by various factors, including pave-
ment design, material properties, construction practices, and environmental conditions
[6]. Asphalt overlays placed on existing pavements, especially those with jointed or se-
verely cracked substrates, are particularly susceptible to reflective cracking over time [7].
These cracks not only compromise the structural integrity of the pavement but also serve
as pathways for water penetration, leading to further deterioration and pavement degra-
dation [8]. Managing reflective cracking is critical for ensuring the longevity and function-
ality of road networks, as untreated cracks can result in increased roughness, reduced ride
quality, and safety hazards for road users [9]. Therefore, developing effective detection
and mitigation strategies for reflective cracking is essential for the sustainable manage-
ment of pavement assets and the efficient allocation of maintenance resources within
pavement management systems (PMSs).
Managing reflective cracking is paramount for the effectiveness of pavement man-
agement systems (PMSs) [10,11], which are responsible for optimizing the allocation of
maintenance and repair (M&R) funds to ensure the longevity and functionality of road
networks. Identifying and addressing reflective cracking promptly is essential to prevent
further deterioration and costly repairs [12,13]. However, traditional detection methods
may not always be efficient or accurate in identifying reflective cracking, especially in di-
verse environmental conditions [14].
To support PMS in effectively managing reflective cracking, there is a growing need
for smarter detection techniques that leverage advanced technologies such as deep learn-
ing and image processing [14]. By developing automated and intelligent systems capable
of detecting reflective cracking with high accuracy and efficiency [15,16], government
agencies can streamline maintenance and rehabilitation (M&R) efforts and allocate re-
sources more effectively [17,18]. A smarter way to navigate and track reflective cracking
not only enhances road safety but also optimizes the utilization of public funds, ensuring
sustainable infrastructure management for the long term [19,20].
Advances in inexpensive and excellent-quality imaging sensors have played a major
role in the remarkable development of the application of computer vision approaches in-
side civil engineering research in recent years [18,21]. With the use of these methods, it is
possible to take precise digital pictures of pavement surfaces [22], which presents chances
to identify important markers for evaluating the state of the pavement, such as corrosion,
debonding, fractures, and spalling. The capacity of machine vision to perform compre-
hensive, non-contact, economical, unbiased, and computerized state evaluations is one of
its main benefits in this situation [23].
Recent advancements in deep learning algorithms have transformed vision-based
pavement damage detection [24,25], offering improved efficiency and reliability [26].
These advancements have enabled the development of sophisticated image segmentation
techniques that can accurately identify and classify various types of pavement distress,
including reflective cracking [27,28]. Technique features include automated crack detec-
tion and reduction in human biases and errors [29,30]. Object detection, crucial for civil
engineering infrastructure maintenance and safety, has seen significant progress with al-
gorithms like YOLOv6, v7, and v8, each introducing new features and improved perfor-
mance [31–34]. Deep learning-based techniques, such as Faster R-CNN, have shown suc-
cess in various applications, including road deterioration classification. Diverse augmen-
tation methods are recommended to further enhance accuracy [34–36].
This research manuscript presents a comprehensive investigation into the prediction
and detection of reflective cracking in pavement infrastructure through a combination of
machine learning approaches and advanced image detection techniques. Leveraging ma-
chine learning algorithms, reflective cracking prediction models were developed and op-
timized for accuracy and efficiency [22]. Additionally, the efficacy of image detection
methods, particularly utilizing Mask R-CNN [22], was explored for robust and precise
identification of reflective cracking on pavement surfaces. The study not only aims to
Buildings 2024, 14, 1808 3 of 24

enhance the predictive capabilities of PMS through machine learning-based models but
also seeks to integrate advanced image detection technologies to support real-time moni-
toring and assessment of pavement conditions [37–39]. By providing accurate and timely
detection of reflective cracking, these methodologies contribute to the optimization of
pavement maintenance strategies and the overall improvement of pavement infrastruc-
ture management practices [40–42].
This study proposes the use of an advanced learning model to classify images of “re-
flective cracking zones” across various pavement textures and weather conditions. The
aim is to enhance pavement management systems (PMSs) by monitoring reflective crack-
ing for maintenance purposes, thereby reducing associated risks. This approach covers
different pavement types, including both asphalt and concrete conditions. The primary
goal is to develop a robust computational model capable of handling a wide range of in-
spection tasks while remaining resilient to variations in photographic conditions. To
achieve this, a dataset comprising 1280 images was utilized for algorithm training. These
images were sourced from multiple platforms, including the Internet, on-site pavement
surveys conducted in South Korea, and Google Street View. Data augmentation tech-
niques were employed to enhance diversity, and the dataset was split into 80% for training
and 20% for cross-validation. The research progressed through two pivotal stages: first
through the development of a Convolutional Neural Network (CNN) architecture for the
classification of reflective cracking, and second through the implementation of an image
segmentation technique for reflective cracking detection, followed by an analysis of the
training dataset.

2. Methodology
2.1. Development of Deep Learning Prediction Models for Reflective Cracking in Highways
2.1.1. Overview
This section explores the development of deep learning prediction models for reflec-
tive cracking in highways, focusing on both empirical and analytical approaches. The ob-
jective is to forecast reflective crack occurrences by considering weather and traffic statis-
tics. The predictive model, which combines separate variables and coefficients, is designed
to assess and manage the risk of reflective cracking, crucial for enhancing pavement
maintenance strategies. Supervised learning methods for machine learning forecasting,
including decision trees, multiple regression, support vector machines (SVMs), and
Gaussian process analysis, are examined to provide a comprehensive analysis of the data
and improve forecasting accuracy.
Previous research efforts have focused on enhancing forecasts of pavement defor-
mation and fractures through weather-specific adjustments. However, in this study, situ-
ated within the uniform meteorological framework of the Seoul Metropolitan Govern-
ment, we chose not to implement such modifications, capitalizing on the consistent envi-
ronmental conditions present across the examined sites. Using both empirical and analyt-
ical approaches, the prior research aimed to develop a reliable technique for forecasting
reflective cracking on specific expressways in Seoul. The empirical model concentrated on
variables such as average temperature, precipitation, and traffic volume, while the analyt-
ical model included variables like maximum temperature, precipitation, and minimum
relative humidity. Despite having a similar computational structure, these models exam-
ined different sets of independent factors [22].
The simulations were designed to help authorities plan maintenance and repair ac-
tivities more efficiently by offering insights into the likelihood of reflective cracking de-
velopment through regression modeling and variable normalization.

2.1.2. Predictive Model and Enhancements for Improved Accuracy

In order to anticipate the development of reflective cracking, this study presents a
predictive model that combines separate variables and coefficients. The reflective cracking
Buildings 2024, 14, 1808 4 of 24

prediction amount (E) depends on a combination of independent factors and unknown

coefficients (β0, …, β45). The empirical forecasting approach employs factors like mean tem-
perature, rainfall, maximum snowfall, maximum consecutive days with precipitation, and
traffic volume to predict reflective cracking, whereas the analytical framework incorpo-
rates the highest temperature, precipitation, and the amount of traffic. These variables are
inputs and are standardized between −1 and 1, where a value of −1 denotes a low proba-
bility of reflective cracking development and a value of 1 indicates a higher probability.
This model provides a methodical way to assess and control reflective cracking risk, which
is crucial for improving pavement management and upkeep techniques. Equation (1) il-
lustrates how the design of experiments (DOEs) method was used to create the regression
model [22].
= + + + + + + +
+ + + + + + + (1)
+ + + + + +

Here is a detailed explanation of the factors and coefficients:

 : This represents the reflective cracking prediction amount, which is the dependent
variable we a empt to estimate.
 to : These are the coefficients of the independent variables in the predictive
model. Each coefficient represents the contribution of the corresponding independ-
ent variable to the prediction of reflective cracking.
 to : These are the independent variables used in the predictive model. Each var-
iable represents a different environmental or exploitation factor that may influence
the occurrence of reflective cracking. Here is a breakdown of what each variable rep-
resents:
o : Mean temperature;
o : Relative humidity;
o : Largest amount of fresh snowfall;
o : Precipitation days;
o : Traffic volume.
In the empirical forecasting approach, to are standardized among −1 and 1,
where a value of −1 denotes a low probability of reflective cracking development and a
value of 1 indicates a higher probability. Each method provides unique insights into the
prediction model, and these factors are selected based on their possible impact on the oc-
currence of reflective cracking [22].
Although helpful for broad forecasts in Seoul, the prior predictive model has many
drawbacks. Its forecasts may not be accurate outside of its limited range because it mostly
uses local data. Its dependability is also in doubt because it is purely based on past data
and has not been verified by actual forecasts. The model was tested for accuracy by com-
paring it with real data for regions such as Buk-bu Expressway. The comparison showed
that the model’s success rate declined over time, suggesting that it might only be trust-
worthy for information up until 2022.
This shows that additional elements that might influence the occurrence of reflective
cracking, such as changes in policies, were not taken into consideration by the model. The
current study intends to improve the model’s accuracy by taking into account further el-
ements and upgrading its database to solve these problems.
In general, the focus is on forecasting reflective crack occurrences, considering
weather and traffic statistics. Data from the National Meteorological Service and the Seoul
Traffic Information Center revealed that high temperatures, rainfall, snowfall, humidity,
and traffic volume significantly affect reflective crack predictions. Using these factors, a
model was developed and tested, with data normalized to minimize errors. The analysis
also accounted for traffic volume’s impact on road deterioration, despite regulations on
Buildings 2024, 14, 1808 5 of 24

large vehicles. Multiple regression analysis was then used to assess the relationship be-
tween these variables and monthly reflective crack.

2.1.3. Supervised Learning Methods for Machine Learning Forecasting

This investigation employed machine learning methods using supervised learning
techniques, including decision and regression trees, multiple regression, support vector
machines (SVMs) [30], and Gaussian process analysis [43]. These methods, which segment
data according to optimization levels, were used to analyze data and forecast results. For
example, multiple regression explains the link between independent (X) and dependent
(Y) variables, while SVMs use hyperplanes to group variables into clusters. Regression
and decision trees make decisions based on reducing uncertainty and maximizing purity
[44]. Gaussian process regression uses kernel functions to estimate continuous dependent
variables, representing the relationship between input and output variables.
Each method offers a unique perspective to forecasting, contributing to a comprehen-
sive examination of the data. The connection between the variables in the independent
group (X) and the dependent variable (Y) is described by the analysis technique known
as multiple regression [30]. This relationship can be determined by calculating the matrix’s
determinant for the estimation of parameters, as indicated in Equation (2):

= − = [ ]− [ ]
(2)

=( )
where ̂ stands for the residual sum of squares and N is the number of independent var-
iables. Regression analysis was used to obtain the value, which led to the prediction
model.

2.2. Reflection Cracking Detection Model

The second goal of the project was to develop effective techniques for creating a deep
learning framework that is adaptable to different environments. Compiling the image col-
lection required taking pictures of various pavement surfaces and ambient lighting. Pre-
liminary data analysis highlighted these variables’ potentially significant effects on reflec-
tive cracking detection performance. Moreover, anchor boxes are frequently produced by
traditional bounding box methodologies [45], which are often used for reflective cracking
surveillance but may not be precise in measuring and evaluating the amount of reflective
cracking. To address this, our work used state-of-the-art Image Segmentation techniques
to accurately delineate the area of reflective cracking areas [37].

2.2.1. Data Collecting and Pre-Processing

For this study, a training set of data was collected from various sources, including
websites, in-person surveys, and Google Street View, to capture a wide range of pavement
and air quality situations. For quality control reasons, the gathered images showed a range
of resolutions, usually from 360p to 720p, with careful a ention paid to obtaining a well-
balanced depiction across various meteorological situations, lighting circumstances, and
traffic se ings [46]. After that, these pictures were manually labeled using the MATLAB
Image Labeler application [47]. Eight hundred pictures were carefully classified as “Re-
flective cracking” based on professional advice and earlier studies. Comprehensive and
exhaustive coverage of the data was ensured by the dataset’s painstaking organization in
COCO format, which included full annotations and captions for every image [47].
Buildings 2024, 14, 1808 6 of 24

2.2.2. Data Labeling

Under the direction of a Korean pavement assessment specialist, reflection crack lay-
outs were sketched and transcribed into JSON archives. The 1280 photos in the dataset
were labeled using MATLAB 2023b Image Labeler, with the categories of “Reflective
cracking” separated evenly [47]. The COCO-forma ed dataset contains bounding poly-
gon labels and comprehensive picture specs.
Table 1 presents a breakdown of the dataset used for training, validation, and testing
in the context of pavement damage analysis. It categorizes the data by pavement types
and weather conditions, indicating the number of samples allocated for each combination.
For asphalt pavement, both clear and multiple weather scenarios were considered, with
256 samples each for training, 32 for validation, and 32 for testing, totaling 320 samples
per condition. Similarly, concrete pavement data follows the same distribution, resulting
in a total dataset size of 1280 samples [48]. This structured dataset ensures comprehensive
coverage across various conditions, facilitating robust model development and evaluation
for pavement damage detection.

Table 1. Overview of the composition of the dataset (number of images).

Validation
Pavement Types Weather Types Train Data Test Data Total
Data
Clear 256 32 32 320
Asphalt pavement
Multiple weather 256 32 32 320
Clear 256 32 32 320
Concrete pavement
Multiple weather 256 32 32 320
Total 1024 128 128 1280

2.2.3. Data Augmentation

Image Data Generator enhanced photos to increase the diversity of the collection and
make up for the lack of high-quality reflective cracking examples. Techniques included
color upgrades (contrast, saturation) as well as positioning changes (flipping, scaling, and
rotating) [16,49,50]. Processing was expedited by standardizing photos to 300 × 300 pixels
and enlarging them from 200 to 600 pixels. To avoid over-augmentation for efficient model
convergence, the dataset was divided into two halves: 80% for training and 20% for test-
ing. This guided the creation of the model with the suggested architecture.

2.2.4. Deep Learning-Driven Object Identification

As shown in Figure 1, the automated reflection cracking segment was made possible
by the modified Mask R-CNN design. This version of Mask R-CNN was specifically de-
signed to enhance the recognition of reflective cracking, and it was trained using labeled
photos [47]. Modeling creation was made simpler by using Detectron2, a modern object
identification technique, which removed the need for a new Mask R-CNN network. Quick
image segmentation was made possible by utilizing the Detectron2’s pre-trained model,
which facilitated transfer learning for object recognition in a variety of domains. Further-
more, three convolutional blocks and max pooling layers from the Keras deep learning
framework were smoothly included in the sequential network.
Buildings 2024, 14, 1808 7 of 24

Figure 1. Framework for reflection cracking prediction and detection.

2.2.5. Mask R-CNN

By adding object masks to bounding boxes, Mask R-CNN—a Faster R-CNN exten-
sion—improves the identification and segmentation of objects efficiency as shown in Fig-
ure 2. Pixel-level segmentation is made possible by RoIAlign, which guarantees accurate
spatial features in region-of-interest pooling [51]. The Mask head comes next. The already
trained models from Detectron2 were mainly employed in their default setups, with some
hyperparameter value alterations [36]. In Mask R-CNN, models such as R101-FPN per-
form be er than others, despite longer training times and the occasional excessive fi ing
issue.

Figure 2. Mask R-CNN structure.

2.2.6. Applications
The main programming language used in this study was Python, and Google Colab
provided the GPU environment—a Nvidia Tesla P100/K80/T4—for the development of all
deep learning systems [52]. The popular data-flow computing and neural network crea-
tion tools TensorFlow and Keras were used to create the Mask R-CNN approach to reflec-
tive cracking identification. A model that was trained from the Common Objects in
Buildings 2024, 14, 1808 8 of 24

Context (COCO) database modified to fit the reflective cracking database was used for
training.

2.2.7. Hyperparameters
After network topologies and datasets are finalized, it is imperative to set network
hyperparameters before training. These parameters were found via heuristics as opposed
to direct dataset estimates because they are not dependent on any particular dataset [48].
Although Mask R-CNN provides a wide range of hyperparameters to be adjusted during
training, it can take a lot of time and resources to explore every possible configuration.
Because of this, class numbers in the current study were designated as reflective cracking
zones, and default se ings were mostly employed. The default se ings from Detectron2
were kept, while the parameters of the model, such as the learning algorithm, were ad-
justed using Optuna, a hyperparameter tuning tool [48].
Table 2 outlines the key hyperparameters for Mask R-CNN, crucial for its perfor-
mance. These include a base learning rate of 0.00027, five images per batch, a gamma value
of 0.06 for learning rate adjustment, and a maximum of 2000 iterations [47]. It also specifies
18 regions of interest per image and handles three distinct classes. Additionally, parame-
ter cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST is set at 0.6 to balance recall and
precision during evaluation. These se ings collectively ensure the model’s robustness and
effectiveness across various computer vision tasks [47].

Table 2. Mask R-CNN hyperparameter se ing.

Model Parameter Value

cfg.SOLVER.BASE_LR 0.00027
cfg.SOLVER.IMS_PER_BATCH 5
cfg.SOLVER.GAMMA 0.06
cfg.SOLVER.MAX_ITER 2000
cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE 18
cfg.MODEL.ROI_HEADS.NUM_CLASSES 3
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST 0.6

2.2.8. Comparable Architectures

The most recent version of Ultralytics’ YOLO V8 significantly increases object detec-
tion tasks’ speed and accuracy [53]. Its integrated framework supports tasks like picture
categorization and instance segmentation, while its rebuilt backbone network and an-
chor-free head refine detecting powers. It provides models that have been trained with
the recognition of object stages and is flexible in terms of export forms and CPU/GPU
compatibility. YOLO V8, which comes in detection, segmentation, and classification vari-
ations, is a reliable choice that has gained recognition for its contributions to artificial
recognition [53].
The authors of this study used a Kaggle dataset of 1280 reflective cracking photos,
and they used the Pascal VOC annotation format to accurately pinpoint the reflective
cracking spots in the photos. The dataset’s separation into training and testing sets and
intricacy were overcome by converting it into COCO format to satisfy the YOLOV8
model’s needs. After preprocessing and spli ing, the efficacy of the YOLOV8 framework
for identifying reflective cracking forecasts was assessed during training. Improving the
efficiency of models and guaranteeing correct object detection was made possible by data
annotation with the help of tools such as RoboFlow. A crucial method that also became
apparent was data augmentation, which improved the model’s robustness versus changes
in source data and enhanced the training dataset, increasing its precision and dependabil-
ity in real-world projections.
Table 3 presents a comparative analysis of hyperparameter configurations across sev-
eral notable architectures in the realm of object detection. The Yolov4 model [31], with an
Buildings 2024, 14, 1808 9 of 24

input size of 300 × 300, utilizes a momentum of 0.9 alongside a decay rate of 0.00005 and
a learning rate of 0.0013, employing Leaky ReLU activation. In contrast, Yolov5 adopts a
larger input dimension of 416 × 416, maintaining a momentum of 0.9 [38], a slightly higher
decay of 0.0005, and a learning rate of 0.001, employing the ReLU activation function.
Lastly, Yolov8 [53], also with a size of 416 × 416, employs a momentum of 0.85, a decay
rate of 0.0003, and a learning rate of 0.002, utilizing the Mish activation function. These
distinct configurations highlight the nuanced choices made in adjusting hyperparameters,
catering to the specific requirements and architectural intricacies of each model, ulti-
mately impacting their performance in object detection tasks.

Table 3. Hyperparameter configuration of comparable architectures.

Learning
Model Width × Height Momentum Decay Activation
Rate
Yolov4 300 × 300 0.9 0.00005 0.0013 Leaky ReLU
Yolov5 416 × 416 0.9 0.0005 0.001 ReLU
Yolov8 416 × 416 0.85 0.0003 0.002 Mish

2.2.9. Comparative Analysis and Evaluation

Using a method that takes into account multiple parameters, an in-depth comparison
was carried out to assess segmentation algorithms based on pixel recognition precision.
Global accuracy (GA) is computed using an assessment method that includes true posi-
tives (TPs), true negatives (TNs), false positives (FPs), and false negatives (FNs). In order
to additionally improve the segmentation framework, accuracy and Intersection-over-Un-
ion (IoU) scores were combined. IoU quantifies the overlap across predicted and reality
segments [47]. Accuracy was ensured by establishing standards: a class was considered
existent only if the final result possibility was more than 0.5, and true detection necessi-
tated comparing the indicated bounding box with real data and requiring an IoU score of
at least 50% [35]. This all-encompassing strategy made it possible to evaluate segmenta-
tion methods in depth, which helped with accurate choices on model improvement.
The following Equations (3) and (4) describe a comparison analysis conducted using
the proportion of properly recognized pixels:
TP+TN
GA= (3)
TP+TN+FP+FN

TP
IoU= (4)
TP+FP+FN

2.2.10. Cross-Entropy Loss Function

The initial use of the traditional cross-entropy loss function (Equation (5)) in this
study treats background and reflective cracking pixels equitably. However, this approach
may not be the best choice for datasets including instances of uneven reflective cracking.
To address this problem, a weighted binary cross-entropy loss function was devised,
which is represented by Equation (6), where various weights are allocated to different
types of pixels [54,55].
The traditional binary cross-entropy loss function is defined as follows:
L(y, ) = −[ylog( ) + (1 − y)log(1 − )], (5)
where
 is the true label (0 for background, 1 for reflective cracking);
 is the predicted probability;
 log denotes the natural logarithm.
Buildings 2024, 14, 1808 10 of 24

To account for the uneven distribution of background and reflective cracking pixels,
a weighted binary cross-entropy loss function is introduced. The weighted loss function
is defined as
L(y, ) = −[ log( ) + (1 − )log(1 − )], (6)
where
 is the true label (0 for background, 1 for reflective cracking);
 is the predicted probability;
 is the weight assigned to the reflective cracking pixels;
 is the weight assigned to the background pixels;
 log denotes the natural logarithm.
By assigning different weights and to the different types of pixels, this
weighted loss function be er accommodates the uneven distribution in the dataset, im-
proving the model’s performance in identifying reflective cracking.

3. Results and Discussions

3.1. Reflection Cracking Prediction Results
3.1.1. Initial Multilinear Regression Analysis
Table 4 outlines the findings of multilinear regression analysis. It shows a strong cor-
relation (R = 0.858) and a reasonable coefficient of determination (R2 = 0.731). The Root
Mean Squared Error (RMSE) is 1795.00, indicating the average deviation between ob-
served and predicted values. The Mean Percentage Error (MPE) is 30.60%, suggesting
overall accuracy.

Table 4. Result of multilinear regression analysis.

Model R R2 RMSE MPE (%) Std. Error of the Estimate

Regression Equation 0.858 0.731 1795.00 30.60 845.00

Table 5 uses multiple linear regression and correlation analysis to illustrate how dif-
ferent factors affect the prediction of reflective cracking. Both analyses highlight the strong
positive relationships between reflective cracking frequency and traffic volume and pre-
cipitation, as well as the significant positive influence of the lowest temperature. The cor-
relation and regression studies indicate that the maximum continuous precipitation day
and average temperature are positively correlated with the occurrence of reflective crack-
ing.

Table 5. Correlation Analysis Results for Predictive Factors in Pavement Damage.

Rank Correlation Analysis Multi-Linear Regression Analysis

Correlation Factor p-Value
1 Precipitation 0.612
2 Traffic Volume 0.550
3 Min. Temperature 0.510
4 Max. Continuous Precipitation Day 0.505
5 Avg. Temperature 0.502
6 Precipitation Day 0.446
7 Max. Snowfall 0.420
8 Max. New Snowfall 0.382
9 Relative Humidity 0.369
10 Min. Relative Humidity 0.317
11 Max. Temperature 0.192
12 Evaporation Loss 0.081
Buildings 2024, 14, 1808 11 of 24

Additionally, there are further positive relationships between reflective cracking oc-
currence and variables such as maximum snowfall, maximum day of precipitation, and
maximum amount of fresh snowfall. Regression analysis also supports the moderate pos-
itive associations between reflective cracking occurrence and minimum relative humidity,
as well as the weaker positive relationships between evaporation loss and maximum tem-
perature and reflective cracking occurrence. Overall, temperature-related factors, traffic
volume, and precipitation appear to be the main determinants of the likelihood of reflec-
tive cracking.
According to the multilinear regression analysis, the key predictors for reflective
cracking are monthly minimum temperature, traffic volume, total precipitation, highest
consecutive days with precipitation, and average temperature. Each of these variables has
a significant correlation of 50% or more. The monthly minimum and average tempera-
tures, relative humidity, and precipitation appear to be independent in terms of collinear-
ity among the independent variables. Additionally, a distinct group is formed by the low-
est relative humidity, the greatest number of consecutive days with precipitation, and the
maximum depth of newly fallen snow.
The lowest temperature has the highest correlation with the occurrence of reflective
cracking, followed by relative humidity, precipitation, and traffic volume, as indicated by
the standardization coefficient ranking. Despite its strong link with reflective cracking,
minimum relative humidity is not selected because it overlaps with other variables. In-
stead, the ultimate independent factors chosen are the greatest number of consecutive
days with rainfall and the maximum depth of newly fallen snow.

3.1.2. Comparison of Machine Learning Techniques for Model Optimization

Machine learning techniques were applied to optimize both empirical and analytical
models, aiming to minimize errors using RMSE, MSE, and MAE (Table 6). Training veloc-
ity and time were also evaluated. The multilinear regression model emerged as the opti-
mal choice due to its minimal error, while stepwise regression, despite its fast training
speed, exhibited lower prediction efficiency and larger margins of error.

Table 6. Machine learning yielded optimization results.

Model Metric NH19 Empirical NH19 Analytical NH23 Empirical NH23 Analytical
RMSE 0.278 0.295 0.238 0.248
MSE 0.077 0.087 0.057 0.062
Linear
MAE 0.195 0.198 0.164 0.184
Regression
T.V (n/s) 2000 2100 2100 1800
T.T (s) 1.467 1.563 1.476 1.524
RMSE 4.737 0.948 2.815 0.496
MSE 22.439 0.898 7.923 0.246
Stepwise
MAE 1.115 0.583 0.926 0.383
Linear Reg.
T.V (n/s) 3100 2900 2800 3000
T.T (s) 101.97 110.48 113.97 114.96
RMSE 0.413 0.429 0.432 0.353
MSE 0.171 0.184 0.186 0.125
Decision Tree MAE 0.298 0.323 0.318 0.244
T.V (n/s) 4300 3600 3200 3800
T.T (s) 0.81 1.016 0.876 0.938
RMSE 0.368 0.359 0.334 0.331
Support
MSE 0.136 0.129 0.112 0.109
Vector
MAE 0.248 0.241 0.229 0.231
Machine
T.V (n/s) 4400 3600 4700 3700
Buildings 2024, 14, 1808 12 of 24

T.T (s) 0.414 0.609 0.476 0.568

RMSE 0.379 0.405 0.365 0.325
MSE 0.144 0.164 0.133 0.106
Ensemble MAE 0.257 0.297 0.268 0.227
T.V (n/s) 1400 1300 1400 1600
T.T (s) 1.532 1.8 1.45 1.34
RMSE 0.406 0.383 0.343 0.317
MSE 0.165 0.146 0.118 0.1
Gaussian Pro-
MAE 0.28 0.263 0.252 0.214
cess Reg.
T.V (n/s) 3500 2100 3500 1500
T.T (s) 0.797 1.676 0.441 0.465
Root Mean Squared Error (RMSE); Mean Squared Error (MSE); Mean Absolute Error (MAE); Traffic
volume (T.V); Travel time (T.T).

The optimized multilinear regression model underwent coefficient adjustments

based on previous studies, and the updated variables of Equation (1) are presented in
Table 7. Using these refined coefficients, occurrences of reflective cracking were predicted.
Figure 3 illustrates the comparison between actual and predicted values, showcasing the
improved performance of the new predictive model, which integrates weather, traffic vol-
ume, and reflective cracking survey data from 2022. Standardization of variables was cru-
cial to enhance accuracy, especially given the diverse ranges in traffic and temperature.

Table 7. Determination of regression coefficients.

Empirical Approach Analytical Approach

NH19 Roadway NH23 Expressway NH19 Roadway NH23 Expressway
R2 0.686 0.803 0.635 0.783
MSE 0.077 0.057 0.087 0.062
MAE 0.195 0.164 0.198 0.184
β0 −1.363 −2.571 −1.043 1.365
β10 2.328 2.102 0.744 −0.427
β20 −3.154 −0.711 −0.813 0.608
β30 0.001 −3.793 −0.491 −0.437
β40 1.128 −0.271 −0.420 2.439
β50 0.473 −1.635 −1.000 3.224
β11 0.311 0.197 −0.212 0.079
β22 −1.855 1.258 0.306 −2.471
β33 0.862 −0.728 −0.247 0.327
β44 −0.537 0.336 0.151 −0.667
β55 0.285 0.546 0.162 −0.363
β12 1.271 −0.815 0.389 −0.633
β13 1.696 0.929 0.995 0.050
β14 −0.499 −0.465 −0.426 0.235
β15 −0.100 1.542 −0.007 0.143
β23 −1.857 −0.627 −1.380 −0.696
β24 1.878 0.856 0.128 3.515
β25 0.571 −0.962 0.010 4.274
β34 0.155 0.053 0.130 0.412
β35 −0.015 −3.320 −1.132 −0.182
β45 0.133 −0.521 −0.391 −0.218
Buildings 2024, 14, 1808 13 of 24

(a) (b)
Figure 3. The numbers of predicted reflective cracking: (a) NH19 Roadway and (b) NH23 Express-
way.

Further evaluation involved a comparative analysis between analytical and empirical

approaches, with the predictive model’s efficacy confirmed by forecasting reflective crack-
ing numbers for 2022, a period not included in the initial training or verification phase.
Figure 4 illustrates the performance of the prediction model based on actual reflective
cracking occurrences, utilizing standardized values ranging from −1 to 1.

(a)
Buildings 2024, 14, 1808 14 of 24

(b)
Figure 4. Assessment of predictive model performance for NH19 and NH23. (a) Prediction results
from NH19 roadway zone. (b) Prediction results from NH23 roadway zone.

Figure 4a displays the prediction outcomes from the NH19 roadway area, whereas
Figure 4b showcases the prediction outcomes from the NH23 roadway area. This finding
presents the performance evaluation of the developed prediction models, with the empir-
ical method showing proper performance, particularly on NH19 and NH23 expressways.
Given its efficiency across locations, the empirical approach is recommended for stand-
ardized reflective cracking prediction in pavement management systems. Additionally,
the recalibrated model can aid in cost analysis by accurately predicting reflective cracking
occurrences, facilitating optimal budget allocation, and preventing budgetary discrepan-
cies.
Furthermore, Table 8 provides a comprehensive assessment of the developed predic-
tion models, considering both the standardization method and the methodologies used
(empirical versus analytical). The performance metrics evaluated include Root Mean
Squared Error (RMSE), Mean Absolute Error (MAE), Mean Squared Error (MSE), coeffi-
cient of determination (R2), and the p-value.
For the NH19 district, both empirical and analytical approaches demonstrate com-
petitive results, with slight variations observed in RMSE, MAE, and MSE values. How-
ever, the R2 values indicate be er explanatory power for the analytical approach, suggest-
ing a stronger fit to the data. The corresponding p-values underscore the statistical signif-
icance of the models, further validating their reliability.
Similarly, for the NH23 district, both approaches yield comparable results in terms
of RMSE, MAE, and MSE. However, the analytical approach exhibits higher R2 values,
indicating superior predictive performance. Additionally, the significantly low p-values
emphasize the statistical significance of the analytical model in capturing the underlying
pa erns of reflective cracking occurrence.
In summary, both empirical and analytical approaches demonstrate promising per-
formance in predicting reflective cracking in the NH19 and NH23 districts. However, the
analytical approach appears to offer slightly be er predictive accuracy and statistical sig-
nificance, particularly evident in the R2 values and p-values. These findings highlight the
importance of considering both methodological approaches and standardization tech-
niques in developing robust predictive models for reflective cracking.
Buildings 2024, 14, 1808 15 of 24

Table 8. Assessment of the effectiveness of developed prediction models.

Standardization
Empirical Approach Analytical Approach
RMSE MAE MSE R2 p-Value RMSE MAE MSE R2 p-Value
NH19 0.5355 0.4217 0.2867 0.5612 5.81 × 10−6 0.5743 0.4182 0.3299 0.7151 0.0003
NH23 0.6138 0.4549 0.3767 0.4314 0.4548 0.5691 0.4619 0.3239 0.4302 9.68 × 10−10
Reflective cracking Prediction
Empirical Approach Analytical Approach
RMSE MAE MSE R2 p-Value RMSE MAE MSE R2 p-Value
NH19 344 272 118,598 0.5613 0.2495 369 269 136,390 0.7148 0.0869
NH23 490 363 239,995 0.4314 0.0231 454 369 206,142 0.4303 0.0040

3.1.3. Limitations
Although our finding has provided valuable insights into predictive modeling for
pavement reflective cracking using multilinear regression, it is important to acknowledge
its limitations. One notable limitation is the exclusive reliance on polynomial functions
within the regression model, neglecting the potential benefits of non-polynomial func-
tions. Non-polynomial functions, particularly in the context of mechanics of cracked me-
dia, offer a more nuanced representation of complex phenomena, including possible sin-
gularities that may arise in pavement degradation processes. Therefore, in the next stage
of the research, the aim is to address this limitation by considering non-polynomial func-
tions and leveraging insights from this method [56]. This expansion of modeling tech-
niques will enable a more comprehensive understanding of reflective cracking mecha-
nisms and pave the way for enhanced predictive accuracy and robustness in pavement
management practices.

3.2. Reflective Cracking Categorization

Table 9 presents the average precision scores obtained from different types of datasets
for reflective cracking categorization. Two machine learning methods, including image
classification, were employed across various pavement types to assess their precision rates
under distinct conditions. For concrete pavement, image classification achieved an aver-
age precision of 95.9% under clear weather conditions and 91.4% under multiple weather
scenarios. Similarly, for asphalt pavement, the average precision was recorded at 92.7%
under clear weather conditions and 82.6% under multiple weather conditions. These pre-
cision scores provide insights into the effectiveness of image classification techniques in
accurately categorizing reflective cracking across different pavement types and weather
conditions.
In concrete pavement, the detection effectiveness of reflective cracking is typically
be er due to several factors. One key reason is the stark contrast between the cracks and
the concrete background. Cracks in concrete pavement often appear as dark lines against
a light or white background, making them easier to detect using image processing tech-
niques. The high contrast between the cracks and the background facilitates the segmen-
tation of cracks from the surrounding pavement surface, leading to more accurate detec-
tion.
Additionally, concrete pavement tends to have a smoother surface texture compared
to asphalt pavement, which can further aid in the detection process. The relatively uni-
form texture of concrete makes it easier to distinguish cracks from other surface irregular-
ities, reducing the likelihood of false positives in the detection results. Moreover, concrete
pavement is often characterized by its durability and resistance to deformation, which can
result in more distinct and well-defined cracks compared to asphalt pavement. These well-
defined cracks provide clearer visual cues for automated detection algorithms, improving
the overall effectiveness of crack detection systems.
Buildings 2024, 14, 1808 16 of 24

Table 9. The accuracy results of reflective cracking categorization.

No. Machine Learning Methods Pavement Types Precision Clear (%) Multiple Weather (%)
1 Image classification Concrete pavement 95.9 91.4
2 Image classification Asphalt pavement 92.7 82.6

3.3. Reflective Cracking Segmentations

3.3.1. Overall Results
The results validate the effectiveness of detecting reflective cracking on roadways us-
ing the Image Segmentation technique, as illustrated in Figure 5. The method developed
for detecting reflective cracking with Mask R-CNN performs satisfactorily, effectively dif-
ferentiating between good pavement and reflective cracking pavement.
In the examples provided, the detection performance for reflective cracking is im-
pressive on both types of pavements, with accuracy exceeding 90% in most cases. How-
ever, there is a notable distinction in detection effectiveness between concrete and asphalt
pavements. On concrete pavement, the detection score approaches nearly 99%, showcas-
ing a superior capability to accurately identify reflective cracks. Conversely, on asphalt
pavement, both in rural and urban se ings, the detection rates hover around 96%, indi-
cating a slightly lower but still commendable performance. This discrepancy suggests that
while detection remains highly effective on both pavement types, the inherent character-
istics of concrete pavement, such as its smoother texture and the high contrast between
cracks and the surface, contribute to its exceptional detection capabilities.

(a)
Buildings 2024, 14, 1808 17 of 24

(b)

(c)
Buildings 2024, 14, 1808 18 of 24

(d)

(e)
Figure 5. Results from the trained model for detecting reflective cracking. (a) Original photo of re-
flective cracking on urban road; (b) Classified photo of reflective cracking on urban road; (c) Original
vs. classified photo of reflective cracking on a rural road; (d) Original photo of reflective cracking on
concrete road; (e) Classified photo of reflective cracking on concrete road.

3.3.2. Weather Impact on Training Efficiency

Training scenarios involving loss and precision in relation to various climate varia-
bles are illustrated in Figure 6. Overall, the proposed model for reflective cracking detec-
tion demonstrates satisfactory efficiency indicators. As a detection approach, Mask R-
CNN performs well, consistently achieving a total loss of less than 0.3 and a precision
greater than 0.9. Convergence typically occurs after 2000 iterations, except for databases
Buildings 2024, 14, 1808 19 of 24

containing diverse meteorological conditions, which require additional training up to

4000 iterations.

(a) (b)
Figure 6. Bounding box-based object identification method training outcomes are shown as (a) Total
loss versus Iterations and (b) Mask-RCNN accuracy versus Iterations.

The impact of climate information on training effectiveness is significant; introducing

multiple meteorological photographs is particularly impactful, with larger databases cor-
relating with lower results. This outcome highlights the negative effect of varied weather
conditions on accuracy, aligning with other findings in image classification. The perfor-
mance variance between datasets containing only clear weather photographs and those
with all meteorological types underscores the crucial role that climate parameters play in
successful training. This can be a ributed to the similar characteristics of reflective crack-
ing and other surface irregularities, especially in cloudy and rainy weather.
The model’s efficacy is validated by the test dataset evaluation, with all testing
groups achieving an accuracy of more than 80% using Mask R-CNN. Despite a slight dis-
crepancy in performance between the training and test datasets—possibly due to fluctua-
tions in surrounding conditions—the model performs consistently in all circumstances.
Even with a few small errors, the results support the applicability of this method for
computerized reflective cracking assessment on roads. There were times when overesti-
mations of reflective cracking severity occurred, especially in the identification of climatic
conditions. Expanding the quantity of the training dataset may improve the classifier’s
accuracy and flexibility for future uses. In addition, a comprehensive analysis of the
model’s parameters was carried out to maximize performance. Using a mini-batch size of
four, the loss function was minimized iteratively throughout learning in order to guaran-
tee thorough convergence and avoid excessive fi ing. The learning rate of 0.00025, the
momentum of 0.8, the regularization of 0.0001, and the mini-batch size of 4 are the opti-
mum model parameters.

3.3.3. Results of Average Precision at 50%

Table 10 provides an overview of the segmentation accuracy used to identify reflec-
tive cracking. The findings show that, out of the two categories, the second circumstance
has the least accuracy, while the “Clear se ings” have the most reliability. For example,
the mean accuracy for the first and second circumstances is 92.5% and 83.7%, respectively,
Buildings 2024, 14, 1808 20 of 24

for IOU = 0.5 (AP50). These results imply that different contexts may have an effect on
reflective cracking identification by picture segmentation. Furthermore, the incorporation
of concrete pavement brings uncertainty to the accuracy of the model.
Since the segmentation may mistakenly see reflected cracking as a pa ern on the
pavement surface, the “black” color of asphalt pavement could lead to errors throughout
training. Once reflective cracking occurs throughout the training period, it could be mis-
taken for the black pa ern on the road surface. These findings demonstrate how difficult
it can be to recognize reflective cracking in practical environments.

Table 10. AP50 results.

AP50: the Average Precision at IOU = 0.5

Combined Weather Data
No. Pavement Types 1st Cond. 2nd Cond.
Clear Multiple Weathers
1 Concrete pavement 92.5% 83.7%
2 Asphalt pavement and Concrete pavement 89.1% 80.3%

3.4. Architecture Comparison

Table 11 provides a comprehensive comparison of performance metrics across mul-
tiple architectures utilized for the pavement damage detection task. The evaluated archi-
tectures include Mask R-CNN, Yolov8, Yolov5, and Yolov4. Notably, the table showcases
the time required per iteration for each architecture, with Mask R-CNN having a time of
0.92 s, while Yolov8 demonstrates the most efficient performance with a reduced time re-
quirement of 0.75 s. Furthermore, the table presents the average precision at 50% Intersec-
tion over Union (AP50) on clear weather conditions, specifically on asphalt pavement.
Mask R-CNN achieves an impressive AP50 score of 92.5%, closely followed by Yolov8
with 91.3%. These findings underscore the efficiency and accuracy of different architec-
tures in the context of pavement damage detection, offering valuable insights for practi-
tioners and researchers in the field.

Table 11. Comparisons of performance across several architectures.

Mask R-CNN Yolov8 Yolov5 Yolov4

Time needed per iteration (in seconds) 0.92 0.75 0.81 0.87
AP50 on clear weather 92.5% 91.3% 87.8% 83.4%

3.5. Challenges in Implementation

During the experimental process of implementing the proposed method, several
challenges surfaced, reflecting the complex nature of pavement management and reflec-
tive cracking detection. One significant obstacle pertained to the variability of environ-
mental conditions and pavement types across different regions. The diverse weather pat-
terns and road surface characteristics posed challenges in training and validating the mod-
els effectively. To address this, a rigorous approach to data collection was adopted, incor-
porating a wide range of weather data and pavement images from various sources. Addi-
tionally, fine-tuning the deep learning algorithms required iterative experimentation and
parameter optimization to ensure robust performance across different scenarios. Moreo-
ver, the interpretability of the models and the potential biases in the training data pre-
sented ongoing challenges in achieving generalizability and reliability. To mitigate these
issues, comprehensive sensitivity analyses and model validations were conducted, lever-
aging cross-validation techniques and external validation datasets where possible. De-
spite these challenges, the study underscores the importance of continuous refinement
and validation of predictive models in real-world applications, paving the way for more
accurate and reliable pavement management strategies in the future.
Buildings 2024, 14, 1808 21 of 24

4. Conclusions
This study investigates predictive and detection methods for reflective cracking in
pavement infrastructure, combining machine learning and advanced image detection
techniques. By employing algorithms such as linear regression and Mask R-CNN, predic-
tive models and precise detection methods were developed. These approaches aim to en-
hance pavement management systems by enabling automatic monitoring and assessment
of pavement conditions, ultimately improving maintenance strategies.
 The refined multilinear regression model exhibited improved predictive perfor-
mance for reflective cracking occurrences. By integrating data from weather, traffic
volume, and reflective cracking surveys spanning 2014 to 2018, the model achieved
heightened accuracy. Standardization of variables was crucial for accuracy enhance-
ment, particularly given the diverse ranges in traffic and temperature. Comparative
analysis between analytical and empirical approaches further validated the model’s
efficacy, as it successfully forecasted reflective cracking numbers for 2019, a period
not included in its initial training.
 The comprehensive evaluation of prediction models highlights the competitive per-
formance of both empirical and analytical approaches across the NH19 and NH23
districts. For the NH19 district, the empirical approach yielded RMSE, MAE, and
MSE values of 0.5355, 0.4217, and 0.2867, respectively, while the analytical approach
demonstrated slightly improved values of 0.5743, 0.4182, and 0.3299, indicating a
stronger fit to the data.
 The findings highlight the effectiveness of image classification techniques in catego-
rizing reflective cracking across various pavement types and weather conditions, as
evidenced by precision rates derived from extensive datasets. Notably, for concrete
pavement, image classification achieved remarkable precision rates of 95.9% under
clear weather and 91.4% under various weather scenarios. Conversely, for asphalt
pavement, slightly lower but still impressive average precision scores of 92.7% under
clear weather and 82.6% under multiple weather conditions were a ained. Concrete
pavement’s superior detection effectiveness can be a ributed to several factors, in-
cluding its high contrast with cracks against the background, smoother surface tex-
ture aiding segmentation, and durability leading to well-defined cracks.
 The Mask R-CNN model showed strong performance in detecting reflective cracking,
maintaining a total loss below 0.3 and a precision above 0.9. While convergence typ-
ically happens after 2000 iterations, datasets with varied weather conditions may re-
quire up to 4000 iterations for optimal training. The impact of climate data on training
is significant, with diverse weather conditions correlating with lower results. The
model’s effectiveness was validated by achieving over 80% accuracy in all testing sce-
narios, despite slight performance variations, indicating its reliability across different
conditions.
 The AP50 results illustrate segmentation accuracy for reflective cracking identifica-
tion, revealing varied reliability across contexts. “Clear se ings” exhibited the high-
est reliability, while the “multiple weather” scenario displayed the least reliability.
Mean accuracies for these conditions were 94.7% and 82.4%, respectively, for IOU =
0.5 (AP50). The integration of the “black” color of asphalt pavement poses challenges,
potentially leading to mistaken identification. These findings underscore the practi-
cal difficulty in recognizing reflective cracking.
 Mask R-CNN and Yolov8 exhibited top performance in pavement damage detection,
with AP50 scores of 92.5% and 91.3%, respectively, under clear weather conditions
for asphalt pavement.

Author Contributions: S.-P.S. and T.H.M.L.: conceptualization, methodology, writing—original

draft. K.K., S.-P.S., and T.H.M.L.: visualization, investigation, writing—review, and editing. S.-P.S.
and T.H.M.L.: data curation, software. All authors have read and agreed to the published version of
the manuscript.
Buildings 2024, 14, 1808 22 of 24

Funding: This research received no external funding.

Data Availability Statement: The original contributions presented in the study are included in the
article, further inquiries can be directed to the corresponding author.
Conflicts of Interest: The authors declare no conflicts of interest, financial or otherwise.

References
1. Dhakal, N.; Elseifi, M.A.; Zhang, Z. Mitigation Strategies for Reflection Cracking in Rehabilitated Pavements—A Synthesis. Int.
J. Pavement Res. Technol. 2016, 9, 228–239. https://doi.org/10.1016/j.ijprt.2016.05.001.
2. Tam, A.B.; Park, D.W.; Le, T.H.M.; Kim, J.S. Evaluation on Fatigue Cracking Resistance of Fiber Grid Reinforced Asphalt Con-
crete with Reflection Cracking Rate Computation. Constr. Build. Mater. 2020, 239, 117873.
https://doi.org/10.1016/j.conbuildmat.2019.117873.
3. Fallah, S.; Khodaii, A. Reinforcing Overlay to Reduce Reflection Cracking; An Experimental Investigation. Geotext. Geomembr.
2015, 43, 216–227. https://doi.org/10.1016/j.geotexmem.2015.03.002.
4. Ji, R.Y.; Mandal, T.; Yin, H. Laboratory Characterization of Temperature Induced Reflection Cracks. J. Traffic Transp. Eng. 2020,
7, 668–677. https://doi.org/10.1016/j.jtte.2019.01.002.
5. Doh, Y.S.; Baek, S.H.; Kim, K.W. Estimation of Relative Performance of Reinforced Overlaid Asphalt Concretes against Reflec-
tion Cracking Due to Bending More Fracture. Constr. Build. Mater. 2009, 23, 1803–1807.
https://doi.org/10.1016/j.conbuildmat.2008.09.027.
6. Khodaii, A.; Fallah, S.; Moghadas Nejad, F. Effects of Geosynthetics on Reduction of Reflection Cracking in Asphalt Overlays.
Geotext. Geomembr. 2009, 27, 1–8. https://doi.org/10.1016/j.geotexmem.2008.05.007.
7. Chen, Y.; Zhu, Z.; Lin, Z.; Zhou, Y. Building Surface Crack Detection Using Deep Learning Technology. Buildings 2023, 13, 1814.
https://doi.org/10.3390/buildings13071814.
8. Xi, Y.F.; Ren, S.J.; Chen, B.L.; Yang, B.; Lee, J.; Zhu, G.H.; Zhou, T.C.; Xu, H. Application of Steel-Fiber-Reinforced Self-Stressing
Concrete in Prefabricated Pavement Joints. Buildings 2023, 13, 2129. https://doi.org/10.3390/buildings13092129.
9. Lu, X.; Yan, G. A Quasi-2D Exploration of Mixed-Mode Fracture Propagation in Concrete Semi-Circular Chevron-Notched
Disks. Buildings 2023, 13, 2633. https://doi.org/10.3390/buildings13102633.
10. Di Mascio, P.; Moretti, L. Implementation of a Pavement Management System for Maintenance and Rehabilitation of Airport
Surfaces. Case Stud. Constr. Mater. 2019, 11, e00251. https://doi.org/10.1016/j.cscm.2019.e00251.
11. Zhao, Y.; Goulias, D.; Peterson, D. Recycled Asphalt Pavement Materials in Transport Pavement Infrastructure: Sustainability
Analysis & Metrics. Sustainability 2021, 13, 8071. https://doi.org/10.3390/su13148071.
12. Moradi, M.; Assaf, G.J. Building an Augmented Reality Experience on Top of a Smart Pavement Management System. Buildings
2022, 12, 1915. https://doi.org/10.3390/buildings12111915.
13. Shu, X.; Wang, Z.; Basheer, I.A. Large-Scale Evaluation of Pavement Performance Models Utilizing Automated Pavement Con-
dition Survey Data. Int. J. Transp. Sci. Technol. 2022, 11, 678–689. https://doi.org/10.1016/j.ijtst.2021.09.003.
14. Justo-Silva, R.; Ferreira, A.; Flintsch, G. Review on Machine Learning Techniques for Developing Pavement Performance Pre-
diction Models. Sustainability 2021, 13, 5248.
15. Pei, L.; Sun, Z.; Xiao, L.; Li, W.; Sun, J.; Zhang, H. Virtual Generation of Pavement Crack Images Based on Improved Deep
Convolutional Generative Adversarial Network. Eng. Appl. Artif. Intell. 2021, 104, 104376. https://doi.org/10.1016/j.en-
gappai.2021.104376.
16. Que, Y.; Dai, Y.; Ji, X.; Kwan Leung, A.; Chen, Z.; Tang, Y.; Jiang, Z. Automatic Classification of Asphalt Pavement Cracks Using
a Novel Integrated Generative Adversarial Networks and Improved VGG Model. Eng. Struct. 2023, 277, 115406.
https://doi.org/10.1016/j.engstruct.2022.115406.
17. Patel, T.; Guo, B.H.W.; van der Walt, J.D.; Zou, Y. Effective Motion Sensors and Deep Learning Techniques for Unmanned
Ground Vehicle (UGV)-Based Automated Pavement Layer Change Detection in Road Construction. Buildings 2023, 13, 5.
https://doi.org/10.3390/buildings13010005.
18. Tabatabai, H.; Aljuboori, M. A Novel Concrete-Based Sensor for Detection of Ice and Water on Roads and Bridges. Sensors 2017,
17, 2912. https://doi.org/10.3390/s17122912.
19. Rhee, J.Y.; Park, K.T.; Cho, J.W.; Lee, S.Y. A Study of the Application and the Limitations of Gpr Investigation on Underground
Survey of the Korean Expressways. Remote Sens. 2021, 13, 1805. https://doi.org/10.3390/rs13091805.
20. Ramanna, S.; Sengoz, C.; Kehler, S.; Pham, D. Near Real-Time Map Building with Multi-Class Image Set Labeling and Classifi-
cation of Road Conditions Using Convolutional Neural Networks. Appl. Artif. Intell. 2021, 35, 803–833.
https://doi.org/10.1080/08839514.2021.1935590.
21. Zhang, L.; Yang, F.; Daniel Zhang, Y.; Zhu, Y.J. Road Crack Detection Using Deep Convolutional Neural Network. In Proceed-
ings of the 2016 IEEE International Conference on Image Processing—ICIP, Phoenix, AZ, USA, 25–28 September 2016; pp. 3708–
3712. https://doi.org/10.1109/ICIP.2016.7533052.
22. Lee, S.Y.; Le, T.H.M.; Kim, Y.M. Prediction and Detection of Potholes in Urban Roads: Machine Learning and Deep Learning
Based Image Segmentation Approaches. Dev. Built Environ. 2023, 13, 100109. https://doi.org/10.1016/j.dibe.2022.100109.
Buildings 2024, 14, 1808 23 of 24

23. Dong, C.Z.; Catbas, F.N. A Review of Computer Vision–Based Structural Health Monitoring at Local and Global Levels. Struct.
Health Monit. 2021, 20, 692–743. https://doi.org/10.1177/1475921720935585.
24. Harrou, F.; Zeroual, A.; Hittawe, M.M.; Sun, Y. Recurrent and Convolutional Neural Networks for Traffic Management. Road
Traffic Model. Manag. 2022, 197–246. https://doi.org/10.1016/b978-0-12-823432-7.00011-2.
25. Harrou, F.; Zeroual, A.; Hittawe, M.M.; Sun, Y. Road Traffic Modeling and Management; Elsevier: Amsterdam, The Netherlands,
2022. https://doi.org/10.1016/c2019-0-05283-1.
26. He, K.; Zhang, X.; Ren, S.; Sun, J. Deep Residual Learning for Image Recognition. In Proceedings of the IEEE Conference on
Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778.
https://doi.org/10.1109/CVPR.2016.90.
27. Hittawe, M.M.; Langodan, S.; Beya, O.; Hoteit, I.; Knio, O. Efficient SST Prediction in the Red Sea Using Hybrid Deep Learning-
Based Approach. In Proceedings of the 2022 IEEE 20th International Conference on Industrial Informatics (INDIN), Perth, Aus-
tralia, 25–28 July 2022; pp. 107–114. https://doi.org/10.1109/INDIN51773.2022.9976090.
28. Hittawe, M.M.; Sidibé, D.; Beya, O.; Mériaudeau, F. Machine Vision for Timber Grading Singularities Detection and Applica-
tions. J. Electron. Imaging 2017, 26, 063015. https://doi.org/10.1117/1.jei.26.6.063015.
29. Zhang, J.; Qian, S.; Tan, C. Automated Bridge Surface Crack Detection and Segmentation Using Computer Vision-Based Deep
Learning Model. Eng. Appl. Artif. Intell. 2022, 115, 105225. https://doi.org/10.1016/j.engappai.2022.105225.
30. Xu, Y.; Zhou, Y.; Sekula, P.; Ding, L. Machine Learning in Construction: From Shallow to Deep Learning. Dev. Built Environ.
2021, 6, 100045. https://doi.org/10.1016/j.dibe.2021.100045.
31. Bochkovskiy, A.; Wang, C.-Y.; Liao, H.-Y.M. YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv 2020,
arXiv:2004.10934.
32. Li, C.; Li, L.; Jiang, H.; Weng, K.; Geng, Y.; Li, L.; Ke, Z.; Li, Q.; Cheng, M.; Nie, W.; et al. YOLOv6: A Single-Stage Object
Detection Framework for Industrial Applications. arXiv 2022, arXiv:2209.02976.
33. Li, C.; Li, L.; Geng, Y.; Jiang, H.; Cheng, M.; Zhang, B.; Ke, Z.; Xu, X.; Chu, X. YOLOv6 v3.0: A Full-Scale Reloading. arXiv 2023,
arXiv:2301.05586.
34. Wang, C.-Y.; Bochkovskiy, A.; Liao, H.-Y.M. YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object
Detectors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA,
18–24 June 2022.
35. He, K.; Gkioxari, G.; Dollár, P.; Girshick, R. Mask R-CNN. In Proceedings of the IEEE International Conference on Computer
Vision, Venice, Italy, 22–29 October 2017.
36. Ansari, S.; Rennie, C.D.; Clark, S.P.; Seidou, O. IceMaskNet: River Ice Detection and Characterization Using Deep Learning
Algorithms Applied to Aerial Photography. Cold Reg. Sci. Technol. 2021, 189, 103324. https://doi.org/10.1016/j.coldre-
gions.2021.103324.
37. Zhang, H.; Qian, Z.; Tan, Y.; Xie, Y.; Li, M. Investigation of Pavement Crack Detection Based on Deep Learning Method Using
Weakly Supervised Instance Segmentation Framework. Constr. Build. Mater. 2022, 358, 129117.
https://doi.org/10.1016/j.conbuildmat.2022.129117.
38. Liu, F.; Liu, J.; Wang, L.; Al-Qadi, I.L. Multiple-Type Distress Detection in Asphalt Concrete Pavement Using Infrared Ther-
mography and Deep Learning. Autom. Constr. 2024, 161, 105355. https://doi.org/10.1016/j.autcon.2024.105355.
39. Baduge, S.K.; Thilakarathna, S.; Perera, J.S.; Ruwanpathirana, G.P.; Doyle, L.; Duckett, M.; Lee, J.; Saenda, J.; Mendis, P. Assess-
ment of Crack Severity of Asphalt Pavements Using Deep Learning Algorithms and Geospatial System. Constr. Build. Mater.
2023, 401, 132684. https://doi.org/10.1016/j.conbuildmat.2023.132684.
40. Xiong, X.; Meng, A.; Lu, J.; Tan, Y.; Chen, B.; Tang, J.; Zhang, C.; Xiao, S.; Hu, J. Automatic Detection and Location of Pavement
Internal Distresses from Ground Penetrating Radar Images Based on Deep Learning. Constr. Build. Mater. 2024, 411, 134483.
https://doi.org/10.1016/j.conbuildmat.2023.134483.
41. Al-Huda, Z.; Peng, B.; Algburi, R.N.A.; Al-antari, M.A.; AL-Jarazi, R.; Zhai, D. A Hybrid Deep Learning Pavement Crack Se-
mantic Segmentation. Eng. Appl. Artif. Intell. 2023, 122, 106142. https://doi.org/10.1016/j.engappai.2023.106142.
42. Song, Q.; Liu, L.; Lu, N.; Zhang, Y.; Muniyandi, R.C.; An, Y. A Three-Stage Pavement Image Crack Detection Framework with
Positive Sample Augmentation. Eng. Appl. Artif. Intell. 2024, 129, 107624. https://doi.org/10.1016/j.engappai.2023.107624.
43. Ounpraseuth, S.T. Gaussian Processes for Machine Learning. J. Am. Stat. Assoc. 2008, 103, 429.
https://doi.org/10.1198/jasa.2008.s219.
44. Shai, S.; Shai, B. Understanding Machine Learning from Theory to Algorithm; Cambridge University: Cambridge, UK, 2014.
45. Ali, R.; Chuah, J.H.; Talip, M.S.A.; Mokhtar, N.; Shoaib, M.A. Crack Segmentation Network Using Additive Attention Gate—
CSN-II. Eng. Appl. Artif. Intell. 2022, 114, 105130. https://doi.org/10.1016/j.engappai.2022.105130.
46. Vishwakarma, R.; Vennelakanti, R. CNN Model Tuning for Global Road Damage Detection. In Proceedings of the 2020 IEEE
International Conference on Big Data, Atlanta, GA, USA, 10–13 December 2020; pp. 5609–5615. https://doi.org/10.1109/Big-
Data50022.2020.9377902.
47. Pham, V.; Pham, C.; Dang, T. Road Damage Detection and Classification with Detectron2 and Faster R-CNN. In Proceedings of
the 2020 IEEE International Conference on Big Data, Atlanta, GA, USA, 10–13 December 2020; pp. 5592–5601.
https://doi.org/10.1109/BigData50022.2020.9378027.
48. Wu, Y.; Kirillov, A.; Massa, F.; Lo, W.; Girshick, R. Detectron2. https://github.com/facebookresearch/detectron2 (accessed on 15
May 2024)
Buildings 2024, 14, 1808 24 of 24

49. Mokhtar, M.M.; Morsy, M.; Taha, N.A.; Ahmed, E.M. Investigating the Mechanical Performance of Nano Additives Reinforced
High-Performance Concrete. Constr. Build. Mater. 2022, 320, 125537. https://doi.org/10.1016/j.conbuildmat.2021.125537.
50. Buslaev, A.; Iglovikov, V.I.; Khvedchenya, E.; Parinov, A.; Druzhinin, M.; Kalinin, A.A. Albumentations: Fast and Flexible Im-
age Augmentations. Information 2020, 11, 125. https://doi.org/10.3390/info11020125.
51. Lee, S.Y.; Jeon, J.S.; Le, T.H.M. Feasibility of Automated Black Ice Segmentation in Various Climate Conditions Using Deep
Learning. Buildings 2023, 13, 767. https://doi.org/10.3390/buildings13030767.
52. Singh, J.; Shekhar, S. Road Damage Detection and Classification in Smartphone Captured Images Using Mask R-CNN. arXiv
2018, arXiv:1811.04535
53. Ultralytics Ultralytics YOLOv8. Available online: https://docs.ultralytics.com/vi (accessed on 12 June 2024).
54. Zhang, Z.; Sabuncu, M.R. Generalized Cross Entropy Loss for Noisy Labels. Available online: https://neurips.cc/media/nips-
2018/Slides/12761.pdf (accessed on 12 June 2024).
55. Nar, K.; Ocal, O.; Sastry, S.S.; Ramchandran, K. Cross-Entropy Loss and Low-Rank Features Have Responsibility for Adversarial
Examples. arXiv 2019, arXiv:1901.08360
56. Figiel, Ł.; Kamiński, M. Numerical Probabilistic Approach to Sensitivity Analysis in a Fatigue Delamination Problem of a Two
Layer Composite. Appl. Math. Comput. 2009, 209, 75–90. https://doi.org/10.1016/j.amc.2008.06.039.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual au-
thor(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.