Scribd
Upload
3 views28 pages

Seminar

The document discusses the application of digital image processing techniques to detect pavement failures, detailing the steps involved such as image acquisition, enhancement, preprocessing, thresholding, and failure detection. It presents case studies demonstrating the effectiveness of these methods in various locations, highlighting improvements in accuracy and efficiency compared to traditional assessment methods. The conclusion emphasizes the benefits of using digital image processing for proactive pavement maintenance and management.

Uploaded by

ashiqkn140
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views28 pages

Seminar

The document discusses the application of digital image processing techniques to detect pavement failures, detailing the steps involved such as image acquisition, enhancement, preprocessing, thresholding, and failure detection. It presents case studies demonstrating the effectiveness of these methods in various locations, highlighting improvements in accuracy and efficiency compared to traditional assessment methods. The conclusion emphasizes the benefits of using digital image processing for proactive pavement maintenance and management.

Uploaded by

ashiqkn140
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 28

DIGITAL IMAGE PROCESSING TO DETECT

PAVEMENT FAILURES

Guided By
Presented By
Mrs.Serin Sara Roy
Asst.Professor Hiba Nourin K A
Dept. of Civil Engineering
CONTENTS
 Introduction
 Digital image
 Types of image
 Steps to detect pavement failure
 Case studies
 Conclusion
Reference
INTRODUCTION
 Digital image processing means processing digital image by means of
a digital computer.
Use of computer algorithms, in order to get enhanced image either to
extract some useful information.
Processing of image data for storage,transmission and representation
for autonomous machine perception.
DIGITAL IMAGE
 An image is defined as a two dimensional function F(x,y) , where x
and y are spatial coordinates and the amplitude of F at any pair of
coordinate (x,y) is called the intensity of that image at that point.
 When x,y and amplitude values of F are finite it is called digital image
 A digital image can be represented as a matrix.
 Digital image is composed of a finite number of elements, each of
which elements have a particular value at particular location.
 These elements are referred to as pixels or image elements.
TYPES OF IMAGE
 Binary Images- Contain only two pixel elements i.e 0 & 1 ,where 0
refers to black and 1 refers to white. The image is also known as
Monochrome.
TYPES OF IMAGE
(Contd…)
8 bit colour format – It has 256 different shades of colours and
commonly known as Grayscale image.In this format 0 stands for Black
and 255 stands for white and 127 stands for gray.
 16 bit colour format – It has 65,536 different colours. This format is
divided into three further formats which are Red, Green, Blue. That
famous RGB format.
STEPS TO DETECT TO
PAVEMENT FAILURES

1. Image Acquisition
2. Image Enhancement
3. Image Preprocessing
4. Image Thresholding
5. Failure Detection & Parameter Estimation
Image Acquisition

 Image of the pavement were taken by using a digital camera during


daylight.
Camera optical should be perpendicular to the road surface and its
lateral edge parallel to the aixs.
Image captured on a bright days for better contrast.
Ensured images met required standards during acquisition.
Maintained consistent distant from the camera to the pavement
(ateast 1 m)
Image Enhancement
 To improve the visual quality and features of the image captured
from the pavement
 Enhance features such as cracks, potholes, and surface wear, making
them more distinct and easier to analyze.
Enhance images so that automated algorithms can more accurately
identify and assess pavement conditions.
 Provide clearer images to help in the planning of maintenance and
repair activities.
Methods of Image Enhancement
1. Image Filtering :-
• Apply filters like Gaussian blur, median filter, or bilateral filter to
reduce noise and smooth out images. This helps in making the
pavement features clearer.
• Enhance edges and details using filters like the Laplacian filter or
unsharp mask to make cracks and defects more visible.
2. Gamma Correction: -
Adjust the brightness of the image to correct for underexposure or
overexposure, improving the visibility of pavement features.
4. De-blurring:-
Use algorithms to correct blurriness caused by motion or focus issues,
making pavement features clearer.
Image Preprocessing
 It necessary to develop preprocessing operations to eliminate
environmental interference.
 Useful information exist only in the region containing cracks.
 Raw images may not be directly suitable for identification and
detection
 Factors affecting raw images include noise, weather condition, poor
resolution, and unwanted background.
Image Thresholding
 The simplest method of segmenting image from a
grayscale.
 Thresholding create binary images.
 Automated thresholding extract useful
information from pixels while
minimizing background noise.
 It utilizes a feedback loop to optimize the
threshold value before converting the grayscale
image to binary
 Histogram represent the distribution of pixel
intensity values in an image
Failure Detection & Parameter
Estimation
 The failure were classified based on the features extracted from the
image.
 Extracted features that describe the cracks such as length, width,
area and its orientation.
 Using machine learning algorithms to classify the detected cracks.
CASE STUDY 1
 The study were conducted in the “University of Tripoli”.
 Image capturing process was performed by using Canon EOS 4000D during
evening from 7 pm to 8 pm.
 4 samples of image were collected.
 The captured image has a size of 378x378 pixels to be analysed using MATLAB.
 The distance between the pavement and camera is 1.2 m
CASE STUDY 1
(Contd…)

1) Hay Aldauls – 07 : 14 pm
2) University of Tripoli - 05: 56 pm
c) Janzur – 07: 08 pm 3) Janzur – 07 : 08 pm
4) Al Seyaheyya – 07 : 54 pm
CASE STUDY 1
(Contd…)
Otsu’s Method
 Otsu’s method is used for thresholding.
 The approach separates the pixels within an image into two groups.
 ω and σ represent the ratio of the number of pixels and the average gray level for
the object. ω 1​ and μ 1​ represent the parameters for the background.
 Total Mean Gray Level: Defined by the equation
𝜎^2(𝑡)=𝜔0(𝑡)𝜎^2(𝑡)+𝜔1(𝑡)𝜎^2(𝑡)
where 𝑡 denotes a gray level of the image.
 In threshold value where the sum of foreground and background spreads is
minimized.
 The gray level 𝑡 corresponding to the maximal value of σ w​ is chosen for image
binarization.
CASE STUDY 1
(Contd…)

Crack detection of sample No 1 Parameter estimated


CASE STUDY 1 (Contd…)

Error Percentage =|Wc-Wm|/Wm


Error Percentage = |Lc-Lm|/Lm
Where:
Wc = the 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.
CASE STUDY 1
(Contd…)

Inference
 There is a noticeable error between the proposed technique and the traditional
method.
 It may be due to the scale and ruler segmentation.
 The technique is noted for its efficient processing time , with an average total
time per sample being around 4.190 seconds
CASE STUDY 2
 This study were conducted in Baghdad ,Iraq.
 The image were captured from a GoPro 8 hero camera
 Camera is positioned with an angle of 90 degree and a height of 1.10 m.
 The image must be in jpg format.
8 samples were collected.
CASE STUDY 2 (Contd…)
 The noise reduction is done by Gaussian Blur.
 Gaussian Blur - It is a non-uniform low pass filters
that preserves low spatial frequency.
 The image is convert from RGB to grayscale to
neutralize colors.
 Resized to 200x200 pixels.
 Hysteresis Thresholding – it requires minval &
maxval
 Pixels with intensity gradient of the edges greater
than the maxval are guaranteed to the edges
 The intensity gradient less than minval are non-
edges and it is eliminated
CASE STUDY 2 (Contd…)
 The manual assesmet was conducted by a
committee of three expert engineers in the
maintenance department of the Mayoralty of
Baghdad on a highway in the Baghdad city.
 The percentage of accuracy result of distress
detection for this case study is 88.4% .
 The time needed to finish the manual assessment
was about 23 hrs and automated way was about 3
hrs.
CASE STUDY 2
(Contd…)
Inference
 The automated system showed higher accuracy and consistency in detecting
pavement distress when compared to manual assessments.
 Implementing AEOP can improve the efficiency and safety of pavement
condition assesments.
The system provides a real- time results , which can help in timely maintenance
and repairs ,reducing the overall cost and enhancing road safety.
CONCLUSION
 The application of digital image processing to detect pavement failure
represents a significant advancement in the field of infrastructure maintenance
and management.
 It provides a non-destructive, efficient, and cost-effective means to continuously
monitor and assess pavement conditions.
The ability to analyze large volumes of data quickly and accurately ensures timely
identification of failures, enabling proactive maintenance and reducing the risk of
severe deterioration and costly repairs.
The integration of machine learning and artificial intelligence enhances the
capability of digital image processing systems, allowing for the automatic
detection and classification of different types of pavement distresses.
REFERENCES
[1] Abdulsalam Basher Alayat & Hend Ali Omar , “Pavement Surface Distress
Detection Using Digital Image Processing Techniques”, Jurnal Kejuruteraan
35(1) ,247-256, 2023.
[2] Tariq N Ataiwe , “Using Image Processing for Automatic Detection of
Pavement , Al salam Journal for Engineering and Technology, Vol . 2.
No.1.(2023) p.46-52, 2022.
[3] Abbas & Ismael , “ Automated Pavement Distress Detection Using Image
Processing Techniques” , Vol. 11, No.5, 7702 -7708.
[4] N. Safaei , O Smadi, Masoud ,B. J. I. J. o. P . R. Safaei , and Technology ,
“An automatic image processing algorithm based on crack pixel density for
Pavement crack detection and classification” , Vol. 15, no. 1, pp.159-172,
2022.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy