Pavement Surface Distress Detection Using Digital Image Processing Techniques

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Jurnal Kejuruteraan 35(1) 2023: 247-256

https://doi.org/10.17576/jkukm-2023-35(1)-24 247

Pavement Surface Distress Detection Using Digital Image Processing Techniques


Abdulsalam Basher Alayat* & Hend Ali Omar

Department of Civil Engineering, University of Tripoli, Tripoli, Libya

*Corresponding author: [email protected]

Received 1 March 2022, Received in revised form 1 July 2022


Accepted 2 August 2022, Available online 30 January 2023

ABSTRACT

Road safety and pavement condition are considered top priorities in our civilized societies, and it’s important that the
pavement condition remains in an excellent state for a long time. However, eventually, the pavement will get exposed
to different types of distresses as a result of traffic loads, rough environment conditions, soil conditions, and underline
subgrade. Therefore, to achieve the required standards for the pavement surface roads in our country and provide the
best performance: detection and measurements of distresses extension must be included in maintenance preparation. This
paper proposes a technique for crack detection based on digital image processing using a programming language called
Matrix Laboratory known as MATLAB. The main target is to estimate the pavement’s length, width, and area by capturing
the image using a digital camera with the required precautions and image implementation. Secondly, developing an image
pre-processing operation to eliminate environmental interference as much as possible and subsequently use the image
thresholding method to separate the pixels within the image into two groups to find the thresholding value for image
binarization. The method successfully detects and removes the presence of unwanted objects in an image, even in difficult
situations where surfaces are less visible. Verification showed good results with an excellent processing time, which can be
considered an indicator of pavement crack parameters.

Keywords: Digital image processing; Pavement evaluation; Crack detection; Parameters estimation

INTRODUCTION many sub-areas of civil engineering, such as engineering


document scanning, pavement distress assessment,
Pavement condition evaluation is significant for a good site evaluation via satellite imagery, studies of crack
pavement management system, effective maintenance, propagation and microstructure in cement-based materials,
rehabilitation, and reconstruction decision-making. and evaluation of soil fabric, etc (Dewangan 2016).
Pavements run in excellent conditions and remain so for The typical pavement management system consists
a few years without the need for maintenance. However, of data collection, data verification and data analysis. A
due to the factor of time, pavements will exhibit distress pavement condition survey and analysis provide much
and their performance will start declining. One of the of the information necessary for pavement management,
key components of pavement condition evaluation is and are vital in order to maintain a quantified condition of
the quantification of pavement distress such as cracking, network, more accurate and accessible information, track
potholes, and rutting. Efforts are presently harnessed to performance of treatments, forecast pavement performance,
minimize the impact of the distresses on the safety and anticipate maintenance and rehabilitation needs, establish
serviceability of paved roads, which is why it is necessary maintenance and rehabilitation priorities, and allocate
to accurately measure the width, depth and length of these funding. Therefore, it is critical to collect accurate pavement
distresses. Cracks are the type of distress that occur on the condition data in an efficient and safe manner and to design
surface of pavements. Manual measuring methods are both a reliable analysis system.
time and effort-consuming without giving the required The manual collection system is self- validating in the
accuracy needed in comparison to the automatic techniques sense that all data are collected by an expert. In automatic
that will take less time and effort during the process using collection, data needs to be verified by randomly sampling
digital image processing. a percentage of the data and validating it against the actual
Digital image processing has become economical in corresponding pavement. It is critical to go through the
many fields like signature recognition, iris recognition and alternative evaluation phase which leads to decision making.
face recognition, in automobile detection and in military The following Table (1) contains a comparison between
applications). Each of these applications has its basic the two methods of pavement data collection, manual and
requirements, which may be unique from the others. Image automatic (Rababaah 2005).
processing has been extensively and successfully used in
248

TABLE 1. Manual vs. automatic pavement surveys (Rababaah 2005)

Manual Automatic
Expensive Extra Expensive
Time consuming Impressively fast
Data sampling Fully survey
Subjective Objective
Difficult to manage Management system linked
Poor repeatability High repeatability

The image processing technique in pavement crack images are divided into non-overlapping blocks for entropy
detection has received remarkable attention since the computation. Afterward, second dynamic thresholding is
late 80s. Many algorithms have been provided to detect applied to the resulting entropy blocks matrix, used as the
cracks and used widely in image enhancement, Artistic basis for the identification of image blocks containing crack
effects, Medical visualization, Industrial inspection, Law pixels.
enforcement, human-computer interfaces etc. A quick (Ouyang, Luo & Zhou 2010) such as image
overview of related crack detection research over the years, enhancement, image segmentation and edge detection. The
(Cheng et al. 1999) proposed a pavement cracking noise in pavement crack images is effectively removed by
detection algorithm based on fuzzy logic. The main idea of median filtering. The histogram modification technique is a
the proposed method is based on determining how much useable segmentation approach, and canny edge detection
darker the pixels are than the surroundings by deciding the is an ideal identification approach for pavement distresses.
brightness membership function for gray levels in different
images. Second, map the fuzzified image into the crack (Lee et al. 2013) developed an image processing
domain by finding the crack membership values of the technique that automatically detects and analyses cracks
pixels. Third, check the connectivity of the darker pixels to in the digital image of concrete surfaces. The image
eliminate the pixels lacking in connectivity. processing technique automates the measurement of crack
(Subirats et al. 2006) presented a new approach in characteristics including the width, length, orientation and
automation for crack detection on pavement surface images, crack pattern.
and this method is based on the continuous wavelet. In the
first step, a separable 2D continuous wavelet transforms for
METHODOLOGY
several scales is performed and complex coefficient maps
are built. The angle and modulus information is used to
keep significant coefficients. Then, the wavelet coefficient’s BRIEF STATEMENT

maximal values are searched and their propagation through


This method addresses a subset of asphalt crack detection
scales is analyzed. Finally, post-processing concludes with a
and parameters estimation, specifically the four types of
binary image that indicates the presence or not of cracks on
cracks longitudinal, transverse, alligator and block cracks.
the pavement surface image.
By implementing the Otsu’s thresholding method in
(Oliveira and Correia 2009) presented a novel
different samples. The developed MATLAB code consists
framework for automatic crack detection and classification
of the following phases: image acquisition, preprocessing,
using survey images Acquired at high driving speeds. The
crack detection, feature extraction and image representation.
resulting images are pre-processed using morphological
The following flowchart in Figure 1 shows the proposed
filters for reducing pixel intensity variance, then dynamic
crack detection steps:
thresholding is applied to identify dark pixels in images,
as these correspond to potential crack pixels. Thresholded
249

FIGURE 1. The crack detection flowchart

EXPERIMENT PROCEDURE

In this data collecting procedure, the acquired images were


taken by using a digital camera during daylight with its
optical perpendicular to the road surface and its lateral edge
parallel to the axis. This process was done by using a digital
camera, from randomly selected streets and parking lots.
Then, we begin collecting colored pictures of different crack
types to keep the images in the required standards during the FIGURE 2. The image acquisition model
acquisition. To maintain consistent, reliable, and systematic
image collection:
Illumination: Illumination is important to the quality of
the captured images which affects the contrast between the
crack objects and the pavement background, therefore we
made sure the images were taken during a bright day and at
similar timing for consistency.
Distance from the camera to the pavement: Maintained the
same pavement distance to the camera as much as possible
for the entire collection process, since it affects the size of
the crack and the relative gaps between crack objects in the
acquired image, which is important for the classification
process.
Expert validation: the data collection technique was
consulted and approved by a domain expert.
For the pavement capture procedure, the model of the
image acquisition system was installed at the highest point
on a portable three-legged tripod (as shown in Figure 2) to
support the camera’s weight and maintain stability, where
the distance between the camera and the pavement was 1.2
FIGURE 3. The image capturing process
meters. The camera had a line of sight of 45 degrees and
was capable of coping with a 100x100 Cm square pavement
area. Figure 3 shows the image capturing process.
250

EXPERIMENT EQUIPMENT

The image capturing process was performed using Canon


EOS 4000D during evening from 7 Pm to 8 Pm to ensure
the best quality possible to minimize shadow interruption in
comparison to noon time. The original image taken by the
camera was flexible to split and resize to achieve balance
between real life scale and image dimensions. The captured
images had a size of 378×378 Pixels to be analyzed using
MATLAB using VAIO laptop, and the verification step was FIGURE 9. Preprocessing Steps
accomplished using a digital caliper and a ruler.
IMAGE PREPROCESSING

For pavement crack images, useful information exists only


in the region containing cracks, but this region is usually
quite small compared to the whole image, especially when
the pavement condition is in a generally good state. This
means that most parts of the image belong to the useless
background, which should be excluded from the processing
FIGURE 4. CANON EOS 4000D task to improve work efficiency. However, most of the time,
the interference from the image shooting environment brings
so much noise that it was difficult to distinguish between
the useful image regions and the useless background.
Therefore, it is necessary to develop an image preprocessing
operation to eliminate environmental interference as much
as possible. Basically, the images which are obtained
during image acquisition may not be directly suitable for
identification and detection purposes due to some factors
FIGURE 5. Three-Legged Tripod
such as noise, weather conditions, poor resolution, and
unwanted background which led us to adopt the established
techniques and study their performances. Figure 9 shows the
main preprocessing steps (Cubero-Fernandez et al. 2017).

IMAGE THRESHOLDING

FIGURE 6. Camera Stick The simplest method of segmenting images from a grayscale
image. Thresholding can be used to create binary images.
To make thresholding completely automated, it is necessary
for the computer to automatically select the threshold,
histogram shape-based methods in particular, but also other
different thresholding algorithms make certain assumptions
about the image intensity probability distribution. Automatic
thresholding is a great way to extract useful information
encoded into pixels while minimizing background noise.
FIGURE 7. Digital Calliper This is accomplished by utilizing a feedback loop to optimize
the threshold value before converting the original grayscale
image to binary. The idea is to separate the image into
two parts; the background and the foreground (Gonzalez,
Woods & Eddins 2010). The histogram is an approximate
representation of the distribution of numerical data to give
a rough sense of the density of the underlying distribution
of the data. In an image processing context the histogram of
FIGURE 8. Useful Instruments Used an image normally refers to histogram of the pixel intensity
values. This histogram is a graph showing the number of
pixels.
251

In an image at each different intensity value found in that MEASUREMENTS CONCEPT


image For an 8-bit grayscale image there are 256 different
possible intensities, and histogram will graphically display The idea is analogous to rotating a spring-loaded Vernier
256 numbers showing the distribution of pixels amongst caliper around the outside of a convex polygon, every
those grayscale values. Figure 10 illustrates the histogram time one blade of the caliper lies flat against an edge of the
graph. polygon, it forms an antipodal pair with the point or edge
touching the opposite blade. The complete “rotation” of
the caliper around the polygon detects all antipodal pairs as
shown in Figure 11.

FIGURE 10. Histogram graphical representation

Creating a histogram provides a visual representation


of data distribution. It can display the frequency of the data
values and show the number of pixels in an image at each FIGURE 11. Antipodal pair of irregular shape
different intensity value found in that image. A histogram is a
Jurnal Kejuruteraan
useful tool for thresholding since the information contained 35(1) 2023: xxx-xxx
The method of rotating calipers can be interpreted as the
https://doi.org/10.17576/jkukm-2023-35(1)-24
in the graph is a representation of pixel distribution that can projective dual of a sweep line algorithm in which the sweep
be analyzed for peaks and valleys to find threshold values, is across slopes of lines rather than across x or y coordinates
then beOtsu’s
used formethod is a commonly
image segmentation (Fisher etemployed
al. 2003). of points. Performing dimensional inspections on complex
image in the thresholding technique. The parts have been a tricky process that has involved everything
basic idea of thisOTSUapproach
METHOD is to separate the
from hand gage measurement to sophisticated optical
pixels within an image into two groups. The systems. This process can be difficult and time-consuming,
especially when holes, anchors, braces, and other items
Otsu’sseparated
method is object is featured
a commonly employed byimage
ω and σ
in the The method of rotating calipers can be
must be located precisely on surfaces with irregular shapes
whichtechnique.
thresholding are the ratio of theidea
The basic number
of this of pixelsis
approach interpreted
(Toussaint 1983). as the projective dual of a sweep
to separate the pixels
and the averagewithin an level
gray image(Otsu
into two groups.
1979). InThe
a line algorithm in which the sweep is across
separated
similar manner, the background of theof
object is featured by ω and σ which are the ratio slopes of lines rather than across x or y
the number of pixels and the average gray level (Otsu 1979).
image also has two parameters: ω1 and µ1.
In a similar manner, the background of the image also has
coordinatesRESULTS of ANDpoints.
DISCUSSION
Performing
The totalω1
two parameters: mean gray
and μ1. Thelevel
total of
meanthegray
image
levelisof dimensional inspections on complex parts
The research has been implemented in MATLAB version
defined in Equation (1), where t denotes
the image is defined in Equation (1), where t denotes a gray a have been a tricky process that has involved
9.8 (R2020a) with Image Processing Toolbox. The image
gray
level of level of the image
the image everything
processing from hand
was performed gage
on Intel measurement
Core i7-3537U 2.00toGHz
sophisticated
CPU with 6GB of RAM. optical systems. This process
" ( ) " "
𝜎𝜎 # 𝑡𝑡 = 𝜔𝜔) (𝑡𝑡)𝜎𝜎 ) (𝑡𝑡) + 𝜔𝜔+ (𝑡𝑡)𝜎𝜎 + (𝑡𝑡) (1) can be difficult and time-consuming,
especially when holes,SAMPLES
RENDERED anchors, braces, and
(1)
The aim is to find the threshold value where the sum other items must be located precisely on
of foreground and background spreads is at its minimum The surfaces method successfully detects and removes any presence
withinirregular shapes (Toussaint
by iterating over all the possible threshold values and of unwanted objects an image, even in difficult situations
1983).
The aim is to find the threshold value where where surfaces are less visible. The effectiveness of this
calculating the sum of weighted variances. The optimum
the will
threshold sumbe of foreground
the one and background
with the minimum within-class method was verified on four pavement images that were
spreads
variance. is atofits
The value theminimum
gray level tby iterating over
corresponding RESULTS
to the collected from local roads,AND DISCUSSION
and prepared to test the reliability
all value
maximal the ofpossible threshold
σw is selected values and
as the thresholding value of the proposed system, also used for the crack analysis and
calculating
for image the sum of weighted variances.
binarization. dimensions extraction. The result
The research has been implemented is reported in all testing
in
images figures, the crack pixels revealed by the proposed
The optimum threshold will be the one with technique MATLAB version 9.8 (R2020a) with
are clean and well separated from the surface
the minimum within-class variance. The structure. ImageExamples
Processing Toolbox.
of pavement imagesThe image
for longitudinal
value of the gray level t corresponding to cracking processing was performed
and the effective steps based ononIntel Core
Otsu’s i7-
algorithm.
the maximal value of σw is selected as the Figure 3537U 2.00the
12 shows GHzsitesCPU with
location 6GB
while of RAM.
Figures 13,14,15
thresholding value for image binarization. and 16 show the crack detection for each sample:
RENDERED SAMPLES
MEASUREMENTS CONCEPT
The method successfully detects and
The idea is analogous to rotating a spring- removes any presence of unwanted objects
loaded Vernier caliper around the outside of in an image, even in difficult situations
a convex polygon, every time one blade of where surfaces are less visible. The
the caliper lies flat against an edge of the effectiveness of this method was verified on
polygon, it forms an antipodal pair with the four pavement images that were collected
point or edge touching the opposite blade. from local roads, and prepared to test the
Jurnal Kejuruteraan 35(1) 2023: xxx-xxx
Jurnal Kejuruteraan 35(1) 2023: xxx-xxx
https://doi.org/10.17576/jkukm-2023-35(1)-24
https://doi.org/10.17576/jkukm-2023-35(1)-24
252

(a)
(a)

(a)
(a)

(b)
(b)

(b)
(b)

(c)
(c)

(c)
(c)

FIGURE 13. Crack detection for sample No.1


FIGURE
FIGURE13.13.
Crack
(a) detection
Crack imagefor
detection
Original sample
for
(b) sample
Initial No.1 (a)image
No.1
binary Original image
(b) Initial
(a) Original image binary
(b) image
Initial
(c) Hole (c) Hole
binary
filled filled image
image
image
(d) (c) Hole filled image
(d)
FIGURE 12. Site views for each sample
FIGURE FIGURE
(a) 12.
Sample 12. Hay
No.1
Site views Site views
for each for each
Al Andauls
sample sample
–(a)May 19th No.1
Sample – 07:14
HayPM
Al
(a) Andauls
Sample
(b) –No.1
MayHay
Sample Al– Andauls
No.2
19th 07:14 PM– (b)
University May 19th ––No.2
of Sample
Tripoli 07:14 PM – 05:56
MayUniversity
27th of PM
(b) Tripoli
Sample No.2
May University
(c) –Sample No.3
27th of PM
Janzur
– 05:56 Tripoli
– June –18th
May–27th
(c) Sample 07:08–Janzur
No.3 05:56 –PM
PM June
(c) 18th
Sample
(d) No.3 Janzur
Sample
– 07:08 PM No.4 –Al
June
(d) Sample 18th
No.4–Al
Seyaheyya 07:08 PM27th –– June
–Seyaheyya
June 07:5427th
PM –
(d) Sample No.4 Al Seyaheyya – June
07:54 PM27th – 07:54 PM
Jurnal Kejuruteraan 35(1) 2023: xxx-xxx
https://doi.org/10.17576/jkukm-2023-35(1)-24 Jurnal Kejuruteraan 35(1) 2023
https://doi.org/10.17576/jkukm-20

253

(a)
(a)

(b)
(b)

(c)
(c)
FIGURE 15. Crack detection for sample No.3 (a) Original image FIGURE
FIGURE
(b) Initial15. Crack
binary detection
image for sample
(c) Hole No.3
filled image (a) Orig
FIGURE 14. Crack detection for sample No.2 (a) Original image (b) Initial binary image
FIGURE 14. Crack detection for sample No.2 (a) Original image
(a) Original image (b) Initial binary image (c) Hole filled image
(b) Initial binary image (c) Hole filled image
(c) Hole filled image
TABLE 2. Parameters estimated
The compa
Sample No. 1 2 3 4
calculated
the values
urnal Kejuruteraan 35(1) 2023: xxx-xxx Crack Width
1.8 2.1 3.2 1.7 provided in
s://doi.org/10.17576/jkukm-2023-35(1)-24 (cm)

Crack Length T
254 54.9 55.4 69.8 53.6
(cm)
Sample No.
The parameters estimated from proposed technique for
Crack Area
98.8 119 53.6 95.1
(cm)
each sample are shown in Table 2:
Wc (cm)
TABLE 2. Parameters estimated
Verification

Sample No. 1 2 of this3 technique,


4 Wm (cm)
During the development
Crack Width several
(cm) tests were
1.8 performed
2.1 3.2 on crack
1.7
Crack Lengthimages
(cm) taken 54.9
with a55.4 digital69.8camera53.6 to Error (%)
demonstrate the98.8
Crack Area (cm) validity119 of the
53.6 proposed
95.1
algorithms. Four crack images were
examined VERIFICATION
in the previous section (Rendered
Samples) to a manual test performed on the The compa
During the development
crack width and of this
lengthtechnique,
in order toseveral
verifytests
the calculated
were performed on crack images taken with a
performance of the developed technique digital cameraby the values m
(a) to demonstrate the avalidity
using Digitalof Caliper
the proposed algorithms.
to measure the in Table 4 l
Four crack images
actual were
crackexamined
width at inseveral
the previous
points,section
and a
(Rendered Samples) to a manual test performed on the T
ruler to measure crack length. The accuracy
crack width and
of alength in order technique
measuring to verify theusing
performance
image Sample No.
of the developed
processing and a relevant deviceCaliper
technique by using a Digital depends to
measure the on actual
the resolution which is determined ata
crack width at several points, and
ruler to measure crack acquisition.
length. The accuracy Lc (cm)
image There of awas measuring
little
technique using image processing and a
difference in the test conditions andrelevant device
the
depends on the resolution which is determined at image
error between the widths and lengths was Lm (cm)
acquisition. There was little difference in the test conditions
defined by the following equations (2) and
and the error between the widths and lengths was defined by
(3):
the following equations (2) and (3):
Error (%)

|#7 8 #9 |
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = #9
(2)
(b) P
|:7 8 :9 |
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = :9
(3) Processing
it assists
Where:
detection. T
Wc = the Where:
crack width calculated by the proposed
technique.
Wm = the crack width measured by the digital caliper.
Lc = the crack length measured by the proposed
technique.
Lm = the crack length measured by the ruler.

The comparison between the crack width calculated by


(c) the proposed technique and the values measured by a digital
caliper is provided in the width comparison table:
FIGUREFIGURE 16.detection
16. Crack Crack detection for No.4
for sample sample
(a)No.4
Original image
le No.3 (a)
(b)Original imageimage
Initial binary (b) Initial binary
(c) Hole image
filled image
mage (c) Hole filled image
255

TABLE 3. Width comparison CONCLUSION

Sample No. 1 2 3 4
A large amount of error was incorporated in the measured
Wc (cm) 1.8 2.1 3.2 1.7
crack length due to the scale factor and the use of a ruler
Wm (cm) 2.6 2.2 2.3 1.6 which necessitates dividing the crack length into several
Error (%) 31 3 38 10 segments. On the other side, the crack width results obtained
by the proposed technique were reasonably close to those
measured by the manual method. However, the present
The comparison between the crack length calculated by results suggest that the proposed technique can provide
the proposed technique and the values measured by a Ruler sufficient accuracy for analyzing the widths of the pavement
is provided in Table 4 length comparison: surface. Hopefully, in the future, we’ll use artificial
intelligence (AI) programs to predict lengths and reduce
TABLE 4. Length comparison
the error percentage to 5-10% also further evolve the crack
Sample No. 1 2 3 4 detection algorithm by implying and upgrading electrical
Lc (cm) 54.9 55.4 69.8 55.6 motion sensors with the ability to measure depths and more
complex branched cracks.
Lm (cm) 100 100 100 100
Error (%) 45 45 30 46
ACKNOWLEDGEMENT

The authors would like to thank to University of Tripoli,


PROCESSING TIME
Libya for supporting this research.

Processing time is critical for accuracy, as it assists the DECLARATION OF COMPETING INTEREST
researcher in real-time detection. The average total time
conceived in the operation per sample is equal to 4.190 None
seconds.
REFERENCES

DISCUSSION Chambon, Sylvie & Jean-Marc Moliard. 2011. Automatic road


pavement assessment with image processing: Review and
As the evaluation of pavement condition is the base comparison. International Journal of Geophysics 2011.
of pavement management, the trend toward automatic Cheng, HD, Jim-Rong Chen, Chris Glazier & YG Hu. 1999. Novel
distresses assessment has been widely spread which is based approach to pavement cracking detection based on fuzzy set
theory. Journal of Computing in Civil Engineering 13 (4):
largely on image processing techniques. Since the digital
270-280.
images taken for crack analysis include various difficulties Cubero-Fernandez, A, Fco J Rodriguez-Lozano, Rafael Villatoro,
such as low contrast and uneven illumination for the image Joaquin Olivares & Jose M Palomares. 2017. Efficient
analyzing process, this study developed an image processing pavement crack detection and classification. EURASIP Journal
program using MATLAB for detecting cracks on the surface on Image and Video Processing 2017 (1):1-11.
of pavement roads, relying on the adobe Photoshop Dewangan, Shailendra Kumar. 2016. Importance & applications of
enhancement techniques followed by the Otsu thresholding digital image processing. International Journal of Computer
method. The method proved success and was able to analyze Science & Engineering Technology (IJCSET) 7 (7):316-320.
cracking images with good consistency and reliability. The Du, Zhenyu, Jie Yuan, Feipeng Xiao & Chamod Hettiarachchi.
2021. Application of image technology on pavement distress
study consumes not more than a processing time to detect
detection: A review. Measurement 184:109900.
cracks of pavements therefore, the capability of repeating Falamarzi, Amir, Sara Moridpour & Majidreza Nazem. 2019. A
the same procedure on hundreds of images without failure review on existing sensors and devices for inspecting railway
could be the main key in replacing human manual pavement infrastructure. Jurnal Kejuruteraan 31 (1):1-10.
cracks evaluation. A comparison of digital caliper and ruler Fan, Zhun, Huibiao Lin, Chong Li, Jian Su, Salvatore Bruno &
crack measurement with the results obtained by the proposed Giuseppe Loprencipe. 2022. Use of parallel resNet for high-
technique can accurately detect cracks, but it differs from performance pavement crack detection and measurement.
the actual measurements due to the following: Sustainability 14 (3):1825.
No.1 sample shows an undetected small crack along the Fisher, Robert, Simon Perkins, Ashley Walker & Erik Wolfart.
2003. Adaptive thresholding. online.
main crack
Gonzalez, Rafael C, RE Woods & SL Eddins. 2010. Digital image
No.2 sample is clearly distinguishable from meaningless processing using MATLAB® Rafael C. Gonzalez, Richard E.
small holes Woods, Steven L. Eddins. 2nd ed: New Delhi: McGraw-Hill.
No.3 sample shows meaningless small objects on the edges Khamil, KN, MFM Sabri, AM Yusop, R. Mohamed & M.S.
No.4 sample shows very distinct branch-shaped cracks Sharuddin. 2020. Modelling and simulation of the performance
analysis for Peltier module and Seebeck module using
MATLAB/Simulink. Jurnal Kejuruteraan 32 (2):231-238.
256

Kheradmandi, Narges & Vida Mehranfar. 2022. A critical Rababaah, Haroun. 2005. Asphalt pavement crack classification:
review and comparative study on image segmentation-based A comparative study of three AI approaches: Multilayer
techniques for pavement crack detection. Construction and perceptron, genetic algorithms and self-organizing maps.
Building Materials 321:126162. Indiana University South Bend.
Lee, Bang Yeon, Yun Yong Kim, Seong-Tae Yi & Jin-Keun Kim. Safaei, Nima, Omar Smadi, Arezoo Masoud, & Babak Safaei.
2013. Automated image processing technique for detecting and 2022. An automatic image processing algorithm based on crack
analysing concrete surface cracks. Structure and Infrastructure pixel density for pavement crack detection and classification.
Engineering 9 (6):567-577. International Journal of Pavement Research and Technology
Oliveira, Henrique & Paulo Lobato Correia. 2009. Automatic 15 (1):159-172.
road crack segmentation using entropy and image dynamic Subirats, Peggy, Jean Dumoulin, Vincent Legeay & Dominique
thresholding. 2009 17th European Signal Processing Barba. 2006. Automation of pavement surface crack detection
Conference. using the continuous wavelet transform. 2006 International
Otsu, Nobuyuki. 1979. A threshold selection method from gray- Conference on Image Processing.
level histograms. IEEE transactions on systems, man, and Toussaint, Godfried T. 1983. Solving geometric problems with the
cybernetics 9 (1):62-66. rotating calipers. Proc. IEEE Melecon.
Ouyang, Aiguo, Chagen Luo & Chao Zhou. 2010. Surface Tran, Thai Son, Van Phuc Tran, Hyun Jong Lee, Julius Marvin
distresses detection of pavement based on digital image Flores & Van Phuc Le. 2022. A two-step sequential automated
processing. International Conference on Computer and crack detection and severity classification process for asphalt
Computing Technologies in Agriculture. pavements. International Journal of Pavement Engineering 23
Oyesikua, Olukayode O, Olasunkanmi O Akinyemib, Solomon O (6):2019-2033.
Giwab, Nurudeen S Lawalc & Babatunde O Adetifac. 2019. Valença, J, D Dias-da-Costa & E Júlio. 2010. Development of a
Evaluation of rural transportation technology: A case study Method for Crack Characterization using Image Processing of
of bicycle and motorcycle trailers. Jurnal Kejuruteraan 31 Concrete Surface. 13th Structural Faults and Repair:15-17.
(1):11-18.

You might also like