1

?

Abstract — Most of industries around us make use of iron
machines and tools for manufacturing their products.
On the other hand corrosion is a natural process that
deteriorates the integrity of iron surface. Therefore
rusting of iron take place. To avoid this it is necessary to
detect rusting in earlier stage, so that it can be
prevented. Digital image processing for the detection of
the rusting provides fast, accurate and objective results.
Thi s paper provides a brief description of the past and
present technologies for rust detection. The proposed
method attempted to create a program that is capable of
detecting rust through image processing. Image
processing is known for the manipulation of i mage
through quantizing the image itself in matrix form.
Through this quantization, it gives opportunity to not
only manipulate the image but also detect a particular
subject on the image as well, such as rust. Through
setting the threshold values and the use of edge detection
and segmentation, rusts on the image can be detected.
The threshold values will set the parameters and
characterize what a rust is. The edge detection will check
for the sudden changes of colors in the images. The
segmentation will th en determine the colors on the
image. The results in the edge detection and
segmentation will be integrated to determine the rust on
the image.

Keywords -— Segmentation;thresholding;edge detection;
MATLAB

1. INTRODUCTION

Iron machines and materials are used in most of industries
for manufacturing products. In industries these iron
materials come in contact with humidity and pollution,
therefore increases the rusting of iron. Corrosion takes place
when the mechanical materials come in contact with
humidit y and pollution in industries. Due to the attack of the
corrosion, these mechanical materials undergo the fatigue
that affects the integrity of the metallic surfaces. This
rusting caused by corrosion causes wastage of iron
materials, reduction in eff icienc y and costly maintenance .
Different departments make use of materials that are made

up of iron. In Civil department, for maintaining the good
quality of steel bridges, it is important to detect rust defects
in advance. By detecting rust defects in advan ce, bridge
managers can make important decisions whether to paint
bridges immediately or later. Electricity department makes
use of crossarms that are made up of iron. These crossarms
undergo the process of rusting. Depending on the rust
present on these c rossarms, the decisions are made by
electricity department whether to reuse these crossarms or
not. For making such kind of decisions of classification
Support Vector Machine plays an important role . To detect
the rusting on the metal surfaces of aircraft s, texture analysis
using image processing is done . To use digital image
processing for the detection of rusting of metals is fast,
convenient, a ccurate and very much objective . In this pap er,
object detection will be used in detecting rust. The process
requires segmented images of the material to be processed
in MATLAB for rust detection. Advantages of using image
processing are the accuracy of reading, cost effective, faster,
objective and consistent.

2. COMM ON STEPS FOR RUST DETECTION

General block diagram for rust detection as shown in fig1.
General steps for rust detection include,

A. Automatic capturing of images of materials using
digital camera :
A Digital Camera is used for automatically taking the
imag es of the iron made materials. These captured images
are then processed for the detection of the rust. It is to be
ensured in these images whether there is presence of rust or
not.

B. Apply rust detection techn ique to ensure presence
of rust:
In this step t he captured images are processed to ensure
whether they contain rust or not. This is the most important
step for the detection of the rust. There are different types of
rust detection techniques. Each rust detection technique has
its own series of steps fo r ensuring the presence of rust.
When rust detection technique is applied on the captured
images then it is ensured whether these images contain rust
or not.

RUST DETECTION USING IMAGE
PROCESSING: A REVIEW

DHANYA P , M.TECH Signal Processing , College Of Engineering Thalassery ,
[email protected]
Dr. RINI JONES S B , Associate Professor & HEAD OF ECE DEPARTMENT , College Of Engineering
Thalassery

2

C. If image is rusted the n calculating total area to be
rusted::
In this step of detecting rust in i mages, when images are
found rusted then the total area that i s found rusted is
calculated . This step is generally performed to make sure
that the images is partially rusted or totally rusted.
Depending on the area found rusted important decisions are
to b e made. Therefore, calculation of total rusted area is
done.

D. To make decision on the basis of total rusted area :
In this step decision is made depending on the area that is
found rusted in the previous step.

Fig 1. General steps for rust detection

3. PREVIOUS WORKS FOR RUST DETECTION

Rust defect assessment is important in order to maintain a
good quality of steel bridge painting. Bridge managers can
more realistically develop long -term cost -effective
maintenance programs if they have depen dable coating
condition data. Also, they can make decisions as to whether
a bridge shall be painted again immediately or later. Taking
digital images of the steel bridge surface with a
conventional camera to evaluate its painting surfaces offers
the advant ages of being inexpensive, accurate, objective,
fast, and consistent. The proposed algorithm calculates the
percentage of rust rather than just classifying an image as
defective and non defective. In “wavelet domain detection
of rust in steel bridge images ” proposed by Sindhu Ghanta
and Tanja krap 1 is based on the concept of wavelet
transform. This technique provides entropy minimization for
illumination correction in the images. This is done as a
preprocessing step for completely eliminating shading
effects. In this technique directly colored image s are
processed, therefore, there is no loss of information . The
algorithm for detecting the rust defects has three steps that
are Feature Vectors Extraction, Training and Detection.
Extraction of Feature Vectors, Training and Detection In
this technique two feature vectors are used for classification:
entropy and energy. After applying one level of wavelet
transform to all the three color planes (RGB) of the image,
the entropy and energy values are calculated in each sub
band. By using the feature vectors extracted in this way it is
ensured whether the image is rusted or not.

Detection of Steel Defect Using the Image Processing
Algorithms proposed by M. Sharifzadeh &S. Sadri;
detection and classification of steel surface defects were
investigated 2 . Imag e processing algorithms are applied for
detecting four popular kind of steel defects, i.e., hole,
scratch, Coil break and rust. A set of 250 steel defect images
were used for testing. Some of common operation for
defect detectingare thresholding, noise re moval, edge
detection and segmentation. thresholding is the first step for
hole and scratch detection. The second step is hough
transform. Experimental results shows that hough transform
of the holedefect has Gaussian function with large ? and
scratch defec t has a small ?. For coil break detetion
pixelshave been distributed over the wide range of steel
sheet. Experimental result shows an evident difference
between the histogram of this defect image and the other
defects. In this method the first step findin g the rust defects
is segmentation. For segmentation, image has been
thresholded. For thresholding, many methods such as
Maximum Entropy Sum Method, Entropic Correlation
Method and Renyi Entropy are reported . However, in this
research Renyi Entropy is use d.

In a study conducted by Huwang, N., Son, H., Kim, C., &
Kim, a rust detection program was created to detect rust and
determine the on which area the robot is going to do the grit –
blasting procedure. The first step in their program is the
conversion of the RGB colors to HIS 3 . This procedure
was done to eliminate the probability of false reading. After
such, the image of the rust will then undergo to the process
of classification, to determine what technique or process to
be used in analyzing the rust. The study offered six
categories of techniques. The purpose of these techniques is
to classify whether the pixel belongs to the background or
the rusted area. In these techniques, the neighboring pixels
were also checked for comparison on whether is a rus t or
part of the background. Although, this study still needs
further testing on its rust detection part, since its more on a
comparative study.

3

Towards Corrosion Detection System proposed by B.B.
Zaidan etc; which introduced a method for rust detection
with the concept of texture analysis 4 . Texture analysis
plays an important role in detecting the isolated data aand
reducing the error and improving the classification results.
The method uses texture segmentation with the aid of edge
detection. Edge de tection is the process of finding sharp
contrast in intensities in an image. This process significantly
reduces the amount of data in the image while preserving
the most important structured feature of that image. The
proposed solution has focused on the r ough texture of the
corrosion areas, and identifies the simple texture as non –
corrosion area. The test has shows a good result in term of
detecting visible corrosion.

Many studies has been done to tackle image processing
based rust detection such as the s tudy done by Sharma V. &
Tejinder T.,the techniques used for rust detection were
discussed. The first step in doing rust detection using image
processing is through obtaining the data, which is obtaining
the image of the object. In this study the data was obtained
automatically through a camera fixed on the object .that is
being monitored. The next step proposed is the detection of
rust. In this step it was proposed that different rust detection
techniques should be done, which might be due to different
typ es and levels of rust. It was emphasized that different
techniques have different steps to follow. The third step is
calculating the area of rust on the image. This is to
determine whether the object is either partially rusted or
totally rusted. An additio nal feature was added in their
study, which to determine on what maintenance should be
done to the object 5

4. PROPOSED MTECH PROJECT

This method is aimed to create a rust detection program with
a 90% success rate. It is implemented by using MATLAB.
To execute the whole program, three functions were created,
namely the thresholding, edge detection , and segmentation.
Segmentation is the process of dividing the image into
pieces. Thresholding is the process if converting theimage
into binary. The assigning of binary bits to the matrix of
image is depend on the intensity of the background in the
image. Edge detection is the process of recognizing the
boundary of an object in the image processing. The output
of thresholding was set to be the input of edge det ection
function. Edge detection works through checking the
neighboring process if their range of value within the
acceptable values in their cluster. These methods set the
parameters of a rust and detects them through their matrix
values.
Block diagram of rust detection as shown in fig 2. The rust
detection method used is based on image segmentation and
image thresholding. The image are segregated into red,
green, and blue channels. The red channel image is stored
into a 2D matrix. The grayscale of the image is then
acquired. Thresholding is then applied to the image. The
thresholding method used is Shannon entropy method. From
the method, the threshold value is acquired and is used to
binarize the image. After the binarization, the images are
classified into non -rust images and rust images. For rust
images, the black pixel represents the rusted area and by
computing the number of pixels and dividing it to the total
number of pixels, the approximate percentage of rust is
detected.The result of the program yields a 90% success rate
in detecting rust on images and 100% in detecting non -rust
images.

Fig 2.Block diagram for rust detection

5. ONCLUSION

This paper described about rust detection techniques to
classify rusted and non rusted images. In this paper we made
a comparison analysis on different existing rust detection
techniques and methods for classification of rusted and non –
rusted images. So we can say that this paper can help those
researchers who are planning to research in this field . T he
proposed method can be implemented using MATLAB.

4

This proposed method yield a success rate 90% in detecting
rust on image with rusts and did not obtain any errors on
images with no rust.

REFERENCE

1″Wavelet domain detection of rust in steel bridge
images,” In 2011 IEEE International Conference on
Acoustics, Speech and Signal Processing (ICASSP).
2 M. Sharifzadeh, S. Alirezaee, R. Amirfattahi and S.
Sadri, “Detection of steel defect using the im age processing
algorithms,” 2008 IEEE International Multitopic
Conference, Karachi, 2008, pp. 125 -127.
3 N. Huwang, H. Son, C. Kim, and *C. Kim, “Rust
Surface Area Determination of Steel Bridge Component for
Robotic Grit -Blast Machine,” In 2013 Proceedin gs of the
30th ISARC, Montréal, 2013, pp 1148 -1156.
4 B. B. Zaidan, A. A. Zaidan, H. O. Alanazi, and R.
Alnaqeib, “Towards Corrosion Detection System,”
International Journal of Computer Science Issues, vol. 7, pp.
33 -36, 2010
5 A. Sharma, and T. Tejin der, “Techniques for Detection
of Rusting of Metals using Image Processing: A Survey,”
International Journal of Emerging Science and Engineering,
vol. 1, 2013, pp 60 -62 .

5

A NON REFERENCE BASED IMAGE QUALITY PREDICTOR USING DEEP CONVOLUTIONAL NEURAL NETWORK:A REVIEW Page 6