You are on page 1of 16

Australian Journal of Electrical and Electronics

Engineering

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tele20

Efficient crack detection and quantification in


concrete structures using IoT

Ajay Nair, Hemalatha R, P Sangeetha, Harish Kumar K, Dinesh Kumar P,


Inakota Sai Sahith & S. Radha

To cite this article: Ajay Nair, Hemalatha R, P Sangeetha, Harish Kumar K, Dinesh Kumar P,
Inakota Sai Sahith & S. Radha (2021) Efficient crack detection and quantification in concrete
structures using IoT, Australian Journal of Electrical and Electronics Engineering, 18:1, 43-57,
DOI: 10.1080/1448837X.2021.1914905
To link to this article: https://doi.org/10.1080/1448837X.2021.1914905

Published online: 26 Apr 2021.

Submit your article to this journal

Article views: 116

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at


https://www.tandfonline.com/action/journalInformation?journalCode=tele20
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING
2021, VOL. 18, NO. 1, 43–57
https://doi.org/10.1080/1448837X.2021.1914905

Efficient crack detection and quantification in concrete structures using IoT


Ajay Naira, Hemalatha Ra, P Sangeethab, Harish Kumar Ka, Dinesh Kumar Pb, Inakota Sai Sahithb and S. Radhaa
a
Dept of Electronics and Communication, Sri Sivasubramaniya Nadar College of Engineering, Chennai, Tamilnadu, India; bDept of Civil, Sri
Sivasubramaniya Nadar College of Engineering, Chennai, Tamilnadu, India

ABSTRACT ARTICLE HISTORY


Now-a-days public transit in our country has a significant usage of bridges and tunnels. Received 1 September 2020
Maintaining the safety of such structures becomes the need of the hour. Cracking can Accepted 7 April 2021
invite sudden failures of concrete structures. Within recent years, there has been an increase KEYWORDS
in the use of image processing techniques as Non-Destructive Testing (NDT) method to Cracks; pattern classification;
detect defects and anomalies in such structures. Hence, this work presents an efficient image morphological processing;
proces- sing model for identifying and quantifying the cracks in common structures using NDT; IoT
two algo- rithms. The first algorithm uses pattern classification method to classify the
identified specimens as crack or not. This pattern classification algorithm is tested and
evaluated with the test specimens of structures having induced cracks. The second algorithm
called morpho- logical processing uses an efficient thresholding strategy to improve the
detection accuracy of the cracks. Here the aim is to detect even the faintest of cracks and
magnify them. Thus, any one of the algorithms can be adopted depending on the
application. The developed algo- rithms have been tested in real field at the final stage
and achieved a detection accuracy of nearly 96.37%and all the obtained images have
been stored digitally on a web server.

1. Introduction approach but requires adaptation to the surface for


Image Processing is a Non-Destructive Evaluation best results and involves a large number of compu-
(NDE) method and has been developed to detect tations. The image acquisition setup is also expen-
exist- ing surface deteriorations including sive. It is desired to obtain the best results using
delamination and voids in concrete. Traditionally, a minimal cost setup. The methods provided in
bridges have been inspected by hammer sounding (Zhang et al. 2014a) give an insight into simpler
and/or chain drag, and visual inspections by qualified crack detection and classification, with crack quan-
engineers and inspectors; however, these methods tification; however, pre-processing requires high-
require a lot of field labour and lane closures, performance hardware equipment. The author in
especially for bridge deck inspections (Yehia et al. (Hoang 2018) presented a novel thresholding
2007). Cracks are one of the major defects found in method to highlight the crack regions along with
civil structures. They can be viewed as a form of crack parameter estimation but failed to remove the
discontinuity that appears on the surface of the falsely detected artefacts. Simple edge detection
structure and correspond to darker regions of the ideas with false positives were also presented
image. Many of these cracks might be harmless. (Dongna et al. 2012). Moreover, curve fitting mod-
However, the failure to detect harmful cracks can els are used to identify the cracks. The Hough
become very costly over a period of time. As more transform is a good tool for identifying line seg-
and more civil structures get built over the years with ments in the data, and have been used in crack
older ones more prone to damage, it becomes detection systems (Nashat, Abdullah, and Abdullah
unfeasible to manually inspect each structure for 2011) but it does not work well when the cracks are
defects. Hence, it is essential to develop an effective not perfectly straight. The top hat transform is used
automatic crack detection model. to detect cracks with a simple segmentation method
Crack detection can be performed by choosing an in (Giakoumis, Nikolaidis, and Pitas 2006). Even
initial set of seed points and growing them (Oliveira misidentified cracks are also corrected. However,
and LobatoCorreia 2017) but it takes a large number the crack detection is not so efficient when they
of iterations to converge to the result. Moreover, it cross the border between two regions. The detec-
does not calculate the crack properties. The road sur- tion method used in (Landstrom and Thurley 2012)
face crack detection algorithm in (Gavilán et al. 2011) was found to yield good results when cracks are
provides good results for different road surfaces and long and connected in nature. It utilises the
performs connection operation by minimum distance

CONTACT Hemalatha R hemalathar@ssn.edu.in Dept of Electronics and Communication, Sri Sivasubramaniya Nadar College of Engineering,
Chennai, Tamilnadu, India
© Engineers Australia
44 A. NAIR ET
AL.

Figure 1. (a)Shrinkage cracks (b) Hairline crack.

method of filling the gaps between cracks but it can The paper is organised as follows. Information on
turn up with relatively more number of false posi- the different types of crack is given in section 2.
tives. Methods for more specific applications such Section 3 briefs about the proposed model. Section 4
as cracks in tunnels and underground pipes are provides the performance evaluation and section 5
also proposed which extract characteristic shapes explains the IoT (Internet of Things)-based monitoring.
of such cracks for detection (Sinha and Fieguth Finally, section 6 concludes the paper with future
2006) (Seung-Nam, Jang, and Han 2007). work.
An effective segmentation method to identify crack
was proposed by Long et al. (Long, Shen, and Chen
2012). It was enabled through grey level correction 2. Overview of cracks
and adaptive minimum error threshold but could not
Cracking can occur due to various reasons such as
classify cracks. Han et al. (Han et al. 2016) used an
thermal stresses, chemical reaction, corrosion of rein-
interactive genetic algorithm and determined the
forcement or simply poor construction practices such
edges of the cracks and calculated its geometric para-
as poor concrete mix or rapid curing. Cracks can be of
meters through intensive calculations. The combina-
different types such as alligator, transversal or long-
tion of Canny edge detector and support vector
itudinal (Ajagbe and Ojedele 2018) (Md. et al. 2018).
machine was used to detect cracks. This enhanced
The other types are shrinkage and hairline cracks as
the ability to extract contours but struggled with
shown in Figure 1, which may result in leakage.
cracks [Li et al. (Shi et al. 2016)]. Similarly, Canny
Horizontal, vertical and diagonal cracks are charac-
edge detection has been used along with fast Haar
terised by how they are oriented in the structure.
transform algorithm for crack detection but, suffered
Alligator cracks occur as a V shape. Settlement cracks
with false crack detection. In addition, the method
may occur if there are problems with respect to the
could not extract crack pixels. The width of the cracks
underlying ground.
was also measured using labelling based on Red-Green
Manual introduction of cracks is performed in the
-Blue (RGB) colour information (Hutchinson and
laboratory to generate a data base. This involves mak-
Chen 2006).
ing of the test specimen in the desired shape as shown
It is also to be noted that a single algorithm for in Figure 2.
crack detection is not always expected to give Curing is done for about a month. After a couple of
efficient results for all practical situations. Some hours of dryness it is kept in an oven for an hour for
structures might have a large number of cracks with hardening. Then, cracks are introduced by the univer-
lesser major ones for repair purposes. Other sal testing machine by applying the desired load as
structures may have few minute cracks but each crack shown in Figure 3.
might have to be detected to ensure the safety of that
structure. Hence, the two proposed crack detection
algorithms will cater correspondingly to the needs of 3. Proposed system model
the afore- mentioned problems. The combination of
The proposed system can be visualised based on a
algorithms proposed in this paper is more efficient in
flow diagram as shown in Figure 4.
detecting even the finest cracks compared to the
The proposed system can be segmented into
complex algo- rithms (Zhang et al. 2014a). Moreover,
three major sections: the image processing, IoT-
real-time mon- itoring methodologies are also
based monitoring and the crack measurements
illustrated to ensure seamless implementation.
and analysis section. The system has two algo-
rithms for crack detection either of which can
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 45

Figure 2. (a) Sample specimen (b) Specimen undergoing load test.

Figure 3. Vertical crack introduced (a) on cube (b) on cylinder.

Figure 4. Proposed system model.

be used depending on the application, along with


3.1.Pattern classification algorithm
an IoT-based monitoring system. The first algo-
rithm optimally detects the presence of true Pattern classification algorithm in Figure 5 is used in
cracks and highlights them. This algorithm aims this paper to classify a given specimen as crack or not
to eliminate maximum false positives generated a crack. It is applicable to a multi-class condition.
due to inappropriate lighting conditions. The distance of each specimen from each of the
The second algorithm highlights the presence of classes is measured and the class with minimum
even the finest of cracks. This is useful in appli- distance is where the specimen belongs.
cations where even the slightest of cracking could
The image database is created by collecting
be a potential threat for the future. This proposed
a large number of similar images of various regions
system has been implemented in real time by
of cracks. The acquired image is first converted to
hosting a web server that can monitor cracks
greyscale. The quality of input image can be sig-
from a remote location.
nificantly improved by preprocessing. It may
46 A. NAIR ET
AL.

Figure 5. Classification algorithm flow diagram.


AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 47

include the removal of noisy regions or improving pixel is replaced with the background pixel. This pro-
the lighting. Further denoising can be performed cess helps to remove foreground regions of the image.
using a median filter which essentially smoothens Dilation: The opposite process from erosion
the image. Median filtering preserves the image wherein if even one pixel of the structuring element
edges while removing the noise. A 3 × 3 kernel is overlaps with the foreground image, then the centre
used which takes the median of values in the pixel is replaced with the foreground pixel. This pro-
neighbourhood and replaces the centre pixel with cess helps to increase the number of foreground pixels.
it. A large kernel size is good for high noise den- Closing: Closing is dilation followed by erosion.
sities but will excessively blur the image. Next, This morphological algorithm is represented in
binarization is used to convert pixel values to bin- Figure 6.
ary as expressed in Equation (1). (1) Preprocessing: The image is filtered out by using
Gaussian blur to remove unwanted noise components.
fb(x) = �0ifx T It is convolved with a Gaussian filter rather than a box
(1)
1ifx ≤
>T filter which replaces each pixel with a weighted aver-
where T is the selected threshold and fb x is the age of the neighbouring pixels. It is used to remove
binarized image. ( ) the high-frequency components. Here, a 3 × 3
The selection of threshold is a critical step in order Gaussian filter governed by Equation (2) is traversed
to optimally detect and separate cracks from the over the image.
image. Here a statistical analysis is used to determine � �
1 —(x2 + y2)
G( x) = exp (2)
the appropriate threshold. The greyscale peak value 2σ2
2πσ2
(N) is obtained using the histogram and is taken as the
initial threshold. If the output is not satisfactory, the where G(x) is the expression for the Gaussian filter in
threshold is manually reduced by 1 pixel increments. two dimensions.
Generally, a threshold of around N-8 to N-10 was Larger the standard deviation (σ), greater is the
found to be suitable by experimentation. For a set of image blurring. It is important to preserve image qual-
images under a certain lighting, threshold will most ity along with reducing noise and so it is not advisable
likely not change. This is followed by image to go for larger values of σ. Here σ = 1 is found to suit
segmenta- tion and labelling. Segmentation involves our requirements.
partitioning the image into its constituent parts. (1) Bottom hat transform: This is the subtraction of
Labelling is done on the image to identify possible image closing by original image. This helps to obtain
regions of cracks. This requires extraction of features darkest regions of the image and remove most of the
pertaining to cracks. The features used for crack background components (Zhang et al. 2014b). The
detection are ratio of major to minor axis and number appropriate choice of the structuring element will
of pixels. The ratio of major to minor axis for a enhance the results. Generally, the shape of the struc-
typical crack is a large value. The input image is turing element is chosen as close as possible to the
given to the two-layer feed- forward network to shape that is to be highlighted. Hence, a rectangular
perform crack classification in MATLAB (Matrix structural element is used. If the size of this element is
Laboratory Simulation Tool). It is trained using scaled too small, it takes a longer time to completely traverse
conjugate gradient back propaga- tion. Data for the image thereby reducing its efficiency and can
classification is organised into the input matrix X and bring up more noise elements. When larger, the
the target matrix T. Here, a selection of 15 neurons desired regions are found to be thicker but can miss
was able to provide good results. Half of the samples out on some small-sized cracks. Here, a size of 15 ×
are used for training and remaining part for testing 15 is found suitable to meet the desired requirements
and validation. Its performance is moni- tored using of highlighting the crack region as a clear output
cross entropy and confusion matrices. The final image. The operation can be depicted by the
output is obtained after the phases of training, testing Equations (3) &
and prediction. (4) as follows.

A_ = A ● B = (A + B) — B (3)
3.2.Morphological processing algorithm
The advantage of morphological image processing for Anew= A_ — —A (4)
crack detection is that even the thinnest of cracks can Here A, A_ , Anew are the original, closed and output
be detected and magnified so that it can be easily images, respectively, and B is the structuring element.
treated. + and– represent dilation and erosion operations
Erosion: The structuring element is moved over the respectively.{\displaystyle\ominus} =
entire image. If even one pixel of the structuring ele- (2) Thresholding: The image thresholding is per-
ment overlaps with the background, then the centre formed using Otsu’s method (Yang et al. 2012). This
48 A. NAIR ET
AL.

Image Acquisition

Gray Scaling

Gaussian Blur

Bottom Hat Transform

Otsu Thresholding

Noise Removal
Figure 6. Morphological algorithm flow diagram.

method involves selecting a particular threshold and connected and seem isolated, they can easily be
checking the number of pixels on either side. The aim removed as they behave as noise components.
is to find the threshold value where the sum of fore-
ground and background spreads is at its minimum. It
is suitable for images with a bimodal histogram dis- 4. Performance evaluation
tribution. This helps to clearly visualise the crack
regions but along with it, a number of salt noises can Dice: Dice is the ratio of twice the intersection
also occur. between the segmentation image and the ground truth
to its summation. Equation (5) shows the formula for
(3) Noise Removal: The final image can be
the Dice metric (Umeha, Hemalatha, and Radha
obtained by removing all the noise components.
2018).
Removal of isolated or sparsely distributed pixels will
effectively remove the noise, which requires the 2(A ∩B)
Dice (5)
knowledge of pixel connectivity. The image is =
A+B
segmented into its constituent components. When
where A is the segmented image and B is the ground
pixels are hardly
truth.
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 49

Jaccard index: Jaccard index is the ratio between


4.2. Morphological processing algorithm
the intersection and union of the segmented result and
the ground truth. Equation (6) shows the formula for The results obtained using morphological proces-
the Jaccard index (Umeha, Hemalatha, and Radha sing algorithm are displayed as shown in Figure 8.
2018). Improper thresholding due to poor contrast in the
image has resulted in poor output image quality in the
A ∩B base algorithm. However, proposed morphological
Jaccard = (6)
A∪ processing gives a more magnified and lucid view of
the cracks. It is advantageous for detection of
where A is the segmented image and B is the ground even minute cracks. The algorithm can be tweaked to
truth. show even the smallest of defects. The drawback is
Emeasure: E-measure is defined by Equation (7) that the detection of false positives cannot be
and uses the enhanced alignment matrix φ to capture avoided easily. Figure 9 shows how the algorithm
the pixel-level matching and image-level statistics of can be tweaked to show varying levels of the
a binary map. defects on
a surface.
w h
1 X X
QFM = FM( x;y) (7)
w×h = = 5. Iot-Based Monitoring
ϕ x

1 y 1

where h and w are the height and the width of the A complete crack detection and estimation model can
binary map, respectively, and ϕFM is the enhanced be implemented using a Raspberry pi (RPi) setup. The
alignment matrix (Fan et al. 2018). monitoring system consists of an RPi and the Pi
These provide a measure of closeness of output camera (PiCam). The web server is hosted by the RPi
image from the neural network with the ground and can be accessed by the user on any device by
truth image, which can otherwise be termed as accu- connecting to its IP address. A python program is
racy of classification. used to implement morphological image processing
algorithm to process the images. The image of cracks
can be uploaded from the device or it can be taken
4.1.Pattern classification algorithm remotely using the RPi and uploaded to the server.
Crack detection has been performed using pattern The PiCam is also used to perform live streaming to
classification method and various experimental get a real-time view of the inspected structure. The
results are obtained. Figure 7 shows a portion of inputs as well as processed images are displayed on
images used in evaluating the system the web server which is acces- sible from any device
performance. connected to the internet.
From Figure.7 it is clear that superior results The Web server setup comprises of the following:
are obtained using the classification algorithm. (1) Web Server: Web Server is an integral part of
Large numbers of false positives are observed in the system being designed. A web server on RPi can be
the LDP (Local Directional Pattern) based imple- used to host a full website, or just used to display some.
mentation output and heavier noise corruption is A Web server can be accessed through the RPi
evident. Based on comparison with the ground remotely through a laptop by using SSH (Secure Shell).
truth image, the Dice, Jaccard and E-measure It is a cryptographic network protocol that ensures
metrics and the accuracy of the classification are secure data transfer between computer and RPi.
calculated and provided in Table 1. Apache is a popular web server, which is a program
From the pattern classification method of crack that listens for server access requests from Internet
detec- browsers and grants them if per- mitted. The web
tion, fairly good results are obtained with reasonable
servers can be accessed using its IP address.
accuracy. The primary advantage of this method is
(2) Internet Framework: Python is used to deploy
that it is suitable for detection of even a large number
the web server on the RPi. The bottle web framework
of cracks on the surface of the specimen with better
architecture is used to make and enable web requests.
accu- racy when compared to (Zhang et al. 2014a).
It is a simple and lightweight structure which comes
Also it enables a very low percentage of false-positive
with inbuilt servers and can be used to perform var-
detection. However, the disadvantage is that minute
ious routing functions.
cracks might not be detected with good accuracy. The
proposed algo- rithm is implemented in the on-board (3) Live Streaming: The RPi camera can be used to
processor with lesser complexity. scan the civil structure and the live video can be
accessed using any device by connecting to the RPi.
50 A. NAIR ET
AL.

Figure 7. A) Input Images b) LDP-based segmentation output c) Proposed classification algorithm-based segmentation output.
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 51

Table 1. Comparison of the performance evaluation metrics.


Evaluation Metrics
LDP Algorithm(Zhang et al. 2014a) Proposed Algorithm
Input Images Dice Jaccard Accuracy E-measure Dice Jaccard Accuracy E-measure
Image 1 0.6156 0.4446 80.48% 0.2830 0.9783 0.9576 95.12% 0.9976
Image 2 0.5265 0.3573 91% 0.3024 0.9598 0.9227 98.72% 0.9437
Image 3 0.5705 0.3991 84.84% 0.2803 0.9462 0.8980 93.94% 0.9273
Image 4 0.7872 0.6491 84% 0.6560 0.9268 0.8636 96% 0.9827
Image 5 0.5978 0.4263 80% 0.2942 0.9674 0.9368 96.67% 0.9862
Image 6 0.6459 0.4770 84.37% 0.8127 0.9481 0.9012 96.9% 0.9959
Image 7 0.6863 0.5224 91.17% 0.9303 0.8396 0.7235 94.1% 0.9825
Image 8 0.6971 0.5350 89.74% 0.8857 0.9034 0.8238 97.4% 0.9960
Image 9 0.0176 0.0089 78.94% 0.2542 0.8722 0.7733 94.7% 0.9772
Image 10 0.2338 0.1324 92.63% 0.4583 0.9274 0.8646 97.9% 0.9871

Figure 8. (a) Input Images (b) LDP-based segmentation output (c) Morphological processing algorithm-based segmentation
output.
52 A. NAIR ET
AL.

Figure 9. (a) Original greyscale image (b) Output after morphological processing (c) Output after noise reduction.

Figure 10. Experimental Setup of RPi with (a) PiCam and (b) an LCD Display.

Figure 11. (a) Upload webpage (b) Database webpage.

From the live video, one can capture photos instantly user. This image is sent to the web server for proces-
and then use these photos for processing. sing. The captured image as well as the processed
image is displayed as output on the web server. The
morpho- logical processing algorithm has been
5.1.Real-time implementation
implemented in Python in the RPi
The experimental setup consists of RPi and PiCam. (https://www.raspberrypi.org). The experimental
This setup can be placed on any civil structure for setup is done as shown in Figure 10.
continuous monitoring of the structure. The PiCam Sample of the web pages is shown in Figure 11.
can be used to take images whenever desired by the The upload page asks the user to upload the image to
be processed or the PiCam can be used. The
database
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 53

Figure 12. Live streaming.

Figure 13. (a) Segmented Output 1 (b) Crack measurement results.

Figure 14. (a) Segmented Output2 (b) Crack measurement results.

page displays the results. These operations can be


5.2.Measurements and analysis
done remotely from any location.
To obtain a real-time view of the structure to be Once the cracks have been identified and processed,
inspected, live-streaming feature has been provided the next step is to analyse it by performing certain
for continuous monitoring. This process is shown in calculations to determine the potential harm that the
Figure 12. crack could possibly impart to the structure under test.
54 A. NAIR ET
AL.

Figure 15. (a) Segmented output 1 (b) Confusion matrix.

Figure 16. (a) Segmented output2 (b) Confusion matrix.

Figure 17. (a) Segmented output 3 (b) Confusion matrix.

Also it provides the accuracy of crack detection. repaired. Such quantitative parameters obtained make
Thus, once all the parameters are obtained one has it easier to ensure a cost-effective treatment of the
to take a decision on when the structure becomes affected structure.
ready to be
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 55

Figure 18. (a) Segmented output 4 (b) Confusion matrix.

Figure 19. (a) Segmented output 5 (b) Confusion matrix.

Table 2. Metric values.


Metric Image 1 Image 2 Image 3 Image 4 Image 5 Image 6 Image 7 Image 8 Image 9 Image 10
Accuracy 95.12% 98.72% 93.94% 98.3% 96.6% 96.9% 94.1% 97.4% 94.7% 97.9%
True Positive 22 20 23 35 19 11 12 13 12 36
False Positive 2 1 2 0 1 0 2 1 1 2
False Negative 0 0 0 1 1 1 0 0 0 0
True Negative 17 57 7 24 39 20 20 25 6 57
Precision 0.917 0.952 0.92 1 0.95 1 0.86 0.93 0.92 0.95
Recall 1 1 1 0.972 1 0.92 1 1 1 1

5.2.1. Crack dimensions the various connected components. Distance between


Measurement of crack parameters is essential to get the two farthest points in a connected component
the crack characteristics and adopt a certain treatment gives the crack length. Crack Area is determined by
mechanism. The number of pixels is obtained from the num- ber of pixels in the crack. A large crack area
but a small
56 A. NAIR ET
AL.

length means that the width of the crack is large.


Funding
Crack Average width is given by the perpendicular
distance between the two end pixels of the midpoint This work was supported by the SSN Trust under Student
of the line joining the two farthest pixels. A larger Internal Funding Project;SSN Trust;
crack width usually implies that the associated
structure is more severely affected. Each of these Notes on contributors
measurements is done for the largest crack, which has
the potential to cause maximum harm. These Ajay Nair obtained his Bachelor’s degree in Department of
Electronics and Communication Engineering from Sri
measured parameters give an idea of how large the
Sivasubramaniya Nadar College of Engineering,
crack is and if it is of immediate concern. Figure 13 Tamilnadu, India.
and Figure 14 show the measure- ment results of the
R. Hemalatha is an Associate Professor in the Department
largest crack for the particular image.
of Electronics and Communication Engineering, Sri
Sivasubramaniya Nadar College of Engineering,
5.2.2. Evaluation metrics Tamilnadu, India. Her area of interest includes Wireless
The metrics which contribute to the overall analysis Multimedia Sensor Networks, Image Processing, and IoT.
of the crack detection method are calculated in this P. Sangeetha is an Associate Professor in the Department
section for the five images which are used in the of Civil Engineering, Sri Sivasubramaniya Nadar College
classification algorithm. The processed images for of Engineering, Tamilnadu, India. She completed her B.E
few samples are shown in Figures 15-19 along with (Civil), MS (Research), and Ph.D from college of
Engineering, Guindy, Anna University. Her area of interest
their confusion matrices. This will enable the cal- includes composite space truss, steel concrete composite,
culation of the metrics. Table 2 summarises the cold formed steel sections and finite element analysis.
metric values obtained from the test images.
Harish Kumar Kobtained his Bachelor’s degree
From Table 2 it is inferred that the proposed algo- in Department of Electronics and Communication
rithm achieves a detection accuracy of nearly 96.37% Engineering from Sri Sivasubramaniya Nadar College of
on an average with better precision. The accuracy is Engineering, Tamilnadu, India.
calculated with reference to the manually built ground P. Dinesh Kumar , obtained his Bachelor’s degree in
truth. Department of Civil Engineering from Sri
Sivasubramaniya Nadar College of Engineering,
Tamilnadu, India.
6. Conclusions and future work
Sai Sahith I, obtained his Bachelor’s degree in Department
The presented paper elaborated on two techniques of of Civil Engineering from Sri Sivasubramaniya Nadar
crack detection and backed it up with a real-time College of Engineering, Tamilnadu, India.
crack detection model. The first technique did crack S. Radha is the Professor and Head of the Department of
detec- tion by measuring how well the cracks Electronics and Communication Engineering, Sri
Sivasubramaniya Nadar College of Engineering,
correlate with standard parameters. This method
Tamilnadu, India. Her area of interest includes Wireless
ensures high crack detection accuracy and will Sensor networks, energy harvesting devices, MEMS,
significantly reduce false positives. The second NEMS, IoT and cognitive radios.
technique used morphological processing which
detects and highlights even the faint- est of cracks.
References
This method is particularly useful when even the
smallest of cracks is of due concern. Depending on Ajagbe, W. O., and O. S. Ojedele. 2018. “Structural
the application, either of the processing techniques Investigation into the Causes of Cracks in Building
can be used. Further the hosting of a web server and Solutions: A Case Study.” American Journal of
Engineering Research (AJER) 7 (8): 152–160.
enables remote monitoring of a structure using IoT.
Dongna, H., T. Tian, H. Yang, S. Xu, and X. Wang,“Wall
Also the quality of images that are taken can be Crack Detection Based on Image Processing” 2012 Third
improved with the help of a camera with higher reso- International Conference on Intelligent Control and
lution. This monitoring system can be integrated with Information Processing, July 15–17, Dalian, China,
a mobile robot to allow scanning of the entire struc- 2012, pp. 597-600, 10.1109/ICICIP.2012.6391474.
ture. Moreover, the correlation between the size and Fan, D.-P., C. Gong, Y. Cao, B. Ren, -M.-M. Cheng, and
A. Borji, “Enhanced-alignment Measure for Binary
type of cracks can be examined to comprehend the Foreground Map Evaluation”, in Proc. Twenty-Seventh
complexity and severity associated with it in real International Joint Conference on Artificial Intelligence
scenarios. (IJCAI-18),July 2018, Stockholm, Sweden, pp.698–704.
Gavilán, M., D. Balcones, D. F. Oscar Marcos, M. A. Llorca,
I. P. Sotelo, M. Ocaña, P. Aliseda, P. Yarza, and
Disclosure statement A. Amírola. 2011. “Adaptive Road Crack Detection

No potential conflict of interest was reported by the


author(s).
AUSTRALIAN JOURNAL OF ELECTRICAL AND ELECTRONICS ENGINEERING 57

System by Pavement Classification.” Sensors 11: 9628–


Refinement”, 25th European Signal Processing
9657. doi:10.3390/s111009628.
Conference (EUSIPCO), Kos island, Greece, 2017,
Giakoumis, I., N. Nikolaidis, and I. Pitas. 2006. “Digital
pp.2080-2084
Image Processing Techniques for the Detection and
Removal of Cracks in Digitized Paintings.” IEEE Seung-Nam, Y., J.-H. Jang, and C.-S. Han. 2007. “Auto
Transactions on Image Processing 15: 178–188. Inspection System Using a Mobile Robot for Detecting
doi:10.1109/TIP.2005.860311. Concrete Cracks in a Tunnel.” Automation in
Han, F., S. Zhao, L. Zhang, and J. Wu. 2016. “Survey of Construction 16: 255–261. doi:10.1016/j.
Strategies for Switching off Base Stations in autcon.2006.05.003.
Heterogeneous Networks for Greener 5G Systems.” Shi, Y., L. Cui, Z. Qi, F. Meng, and Z. Chen. 2016.
IEEEAccess 4 (August): 49594973. “Automatic Road Crack detection Using Random
Hoang, N.-D. 2018. “Detection of Surface Crack in Structured Forests.” IEEE Trans. Intell. Transp.Syst 17
Building Structures Using Image Processing Technique (12, December): 34343445. doi:10.1109/
with an Improved Otsu Method for Image TITS.2016.2552248.
Thresholding.” Hindawi Advances in Civil Engineering Sinha, S. K., and P. W. Fieguth. 2006. “Automated
2018, pp.1-10. Detection of Cracks in Buried Concrete Pipe Images.”
Accessed 15-April-2021 Automation in Construction 15 (January): 58–72.
https://www.raspberrypi.org doi:10.1016/j. autcon.2005.02.006.
Hutchinson, T. C., and Z. Chen. 2006. “Improved Image Umeha, M., R. Hemalatha, and S. Radha, “Structural Crack
Analysis for Evaluating Concrete Damage.” Journal of Detection Using High Boost Filtering Based Enhanced
Computing in Civil Engineering 20 (3, May): 210–216. Average Thresholding”, in the Proceedings of 2018
doi:10.1061/(ASCE)0887-3801(2006)20:3(210). International Conference on Communication and Signal
Landstrom, A., and M. J. Thurley. 2012. “Morphology-based Processing (ICCSP),Chennai, India, 2018, pp. 1026-
Crack Detection for Steel Slabs.” IEEE Journal of 1030, 10.1109/ICCSP.2018.8524284.
Selected Topics in Signal Processing 6: 866–875. Yang, X., X. Shen, J. Long, and H. Chen. 2012. “An
doi:10.1109/ JSTSP.2012.2212416. Improved Median-based Otsu Image Thresholding
Long, J. W., X. J. Shen, and H. P. Chen. 2012. “Interactive Algorithm.” AASRI Procedia 3: 468–473. doi:10.1016/
Document Image thresholding Segmentation Algorithm j.aasri.2012.11.074.
Based on Image Regions.” Journal of Computer Yehia, S., O. Abudayyeh, S. Nabulsi, and I. Abdelqader.
Research and Development 49(7):14201431. July. 2007. “Detection of Common Defects in Concrete
Md., S. A., B. M. AmrulKaish, C.-O. Woon, and Bridge Decks Using Non-destructive Evaluation
S. N. Raman. 2018. “Early-Age Cracking in Concrete: Techniques.” Journal of Bridge Engineering 12: 215–225.
Causes, Consequences, Remedial Measures, and doi:10.1061/(ASCE)1084-0702(2007)12:2(215).
Recommendations.” Applied Sciences 8: 1730. Zhang, W., Z. Zhang, D. Qi, and Y. Liu. 2014a. “Dapeng
doi:10.3390/app8101730. Qi and Yun Liu, “Automatic Crack Detection and
Nashat, S., A. Abdullah, and M. Abdullah, “A Robust Classification Method for Subway Tunnel Safety
Crack Detection Method for Non-uniform Distributions Monitoring”.” Sensors 14 (10): 19307–19328.
of Coloured and Textured Image,” inProc. IEEE Int. doi:10.3390/s141019307.
Conf. Imag. Syst. Tech. (IST), Batu Ferringhi, Malaysia, Zhang, W., Z. Zhang, D. Qi, and Y. Liu. 2014b.
May 2011, pp. 98–103. “Automatic Crack Detection and Classification
Method for Subway Tunnel Safety Monitoring.”
Oliveira, H., and P. LobatoCorreia, “Road Surface Crack Sensors 14: 19307–19328.
Detection: Improved Segmentation with Pixel-based

You might also like