See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/319623148

Identification of Brain Tumor using Image Processing Techniques

Technical Report · September 2017

CITATIONS READS

2 18,112

1 author:

Praveen Gamage
University of Moratuwa
4 PUBLICATIONS   4 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

FileX-teamXcaliburs View project

Independent Study View project

All content following this page was uploaded by Praveen Gamage on 11 September 2017.

The user has requested enhancement of the downloaded file.

 Independent Study Final Report

Level 3

Identification of Brain Tumor using Image Processing

Techniques

Gamage P. T. - 144056U

Supervised by: Dr. Lochandaka Ranathunga

Faculty of Information Technology

University of Moratuwa

2017
 Independent Study Final Report

Level 3

Identification of Brain Tumor using Image Processing

Techniques

Gamage P. T. - 144056U

Supervised by : Dr. Lochandaka Ranathunga

Signature: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Faculty of Information Technology

University of Moratuwa

2017
 Identification of Brain Tumor using Image Processing
Techniques
Gamage P.T.
Faculty of Information Technology, University of Moratuwa
praveenthrilanka@gmail.com

Abstract - At present, processing of medical images is a compared to other medical imaging techniques, Magnetic
developing and important field. It includes many different types Resonance Imagine (MRI) is majorly used and it provides greater
of imaging methods. Some of them are Computed Tomography contrast images of the brain and cancerous tissues. Therefore, brain
scans (CT scans), X-rays and Magnetic Resonance Imaging tumor identification can be done through MRI images. This paper
(MRI) etc. These technologies allow us to detect even the focuses on the identification of brain tumor using image processing
smallest defects in the human body. Abnormal growth of tissues techniques.
in the brain which affect proper brain functions is considered The rest of the paper describes, background of identification of
as a brain tumor. The main goal of medical image processing is brain tumor using image processing (.Section II) and .Section III
to identify accurate and meaningful information using images presents the related work. .Section IV provides a description about
with the minimum error possible. MRI is mainly used to get methodology while Section V gives an acknowledgement of the
images of the human body and cancerous tissues because of its research.
high resolution and better quality images compared with other
imaging technologies. Brain tumor identifications through MRI II. BACKGROUND
images is a difficult task because of the complexity of the brain.
MRI images can be processed and the brain tumor can be Brain Tumor is described as abnormal development of tissues in the
segmented. These tumors can be segmented using various brain. Nowadays the prevalence of tumors is growing fast. In 2016,
image segmentation techniques. The process of identifying an estimated 23,800 adults. (10,350 women and 13,450 men) in the
brain tumors through MRI images can be categorized into four US will be identified with the harmful tumors of brain as well as
different sections; pre-processing, image segmentation, feature spinal code [2]. Analysis of brain tumors is somewhat problematic
extraction and image classification. as the varied shape, size, tumor location and the presence and
Keywords - Brain Tumor, Image Processing, MRI, CT, X-Ray, appearance of tumor in brain. It’s hard to detect brain tumors in
Pre-Processing, Segmentation, Feature Extraction beginning stage because the accurate measurement of tumor can’t
be found. But once the brain tumor is identified at the very
beginning, the proper treatments can be done and it may be curable.
At present, visual representation of the interior of the body is
I. INTRODUCTION processed using medical imaging technique for clinical analysis and
Brain tumors affect the humans badly, because of the abnormal medical researches. MRI is the most effective and extensively used
growth of cells within the brain. It can disrupt proper brain function technique for brain tumor detection. Current diagnosis techniques
and be life-threatening. Two types of brain tumors have been are performed using the conventional methods based on human
identified as Benign tumors and Malignant tumors. Benign tumors experience and this increases the possibility of false detection when
are less harmful than malignant tumors as malignant are fast identifying brain tumors. Present tools and methods to analyze
developing and harmful while benign are slow growing and less tumors and their behavior have become more prevalent. Image
harmful. processing technique can be used to identify brain tumors. Image
processing methods converts images into digital and do operations
Medical imaging technique is used to create visual representation on them, in order to get better and enhanced images [3]. This study
of interior of the human body for medical purposes and noninvasive will focus how to identify brain tumors using image processing
possibilities can be diagnosed by this technology. The various types techniques.
of medical imaging technologies based on noninvasive approach
like; MRI, CT scan, Ultrasound, SPECT, PET and X-ray [1]. When
 III. RELATED WORK processed successfully. The initial step of image processing is
Image Pre-Processing [8].
In recent years, image processing has applied to process images in
medical stream, in cooperating cell detection. In 2012, ‘S.
Mokhled’[4] introduced several identification steps, including
segmenting images to extract the object from the background
through the threshold. This feature was introduced with the ‘Gabor
filter’ in order to do more classification into cancer cells.

In 2013, ‘H. G. Zadeh’ [5] proposed further steps, which is image

extraction and segmentation of images for diagnosing cancer cells.
The Gaussian smoothing concept was introduced as a filtering Original Image Enhanced Image
purpose, previous to applying the ‘Fast Fourier Transform’ (FFT). Figure 2. Image Enhancement [8]
Machine Learning. for tumor detection: ‘NN’, ‘Fuzzy. C-mean’
algorithms was introduced for the identification of tumorous cells Pre-Processing involves processes like conversion to greyscale
[6]. This takes lower computational time but the accuracy also image, noise removal and image reconstruction. Conversion to
lower. greyscale image is the most common pre-processing practice [9].
After the image is converted to greyscale, then remove excess noise
In 2014, ‘X. Chen’ introduces gene counting technology[7]. But
using different filtering methods.
this technology is appropriate only for the complex formation of
gene selection. A..Median Filter
From the above-mentioned techniques and using other This the most common technique which used for noise elimination.
technologies, in this research paper focuses on the identification of It is a ‘non-linear’ filtering technique. This is used to eliminate ‘Salt
brain tumor using image processing techniques. and Pepper noise’ form the greyscale image [10]. Median filter is
based on average value of pixels. The advantages of median filter
IV. .METHODOLOGY
are efficient in reducing Salt and Pepper noise and Speckle noise.
According to the following steps, Brain tumors can be detected Also, the edges and boundaries are preserved. The main
using Image Processing techniques. disadvantages are complexity and time consumption as compared
to mean filter.

Salt and pepper noise Median Filter

Figure 3. Applying Median filter [12]

B. Mean Filter

This filter is also a de-noising filter that is based on average value

of pixels. Advantages of mean filter are it reduces Gaussian noise
and the response time is fast. Main disadvantage is it distorted
Figure 1. Steps for image processing
boundaries and edges.

C. Wiener Filter
4.1 Image Pre-Processing.
The Wiener Filter is also a de-noising filter that is based on the
It is very difficult to process an image. Before any image is inverse filtering in the frequency domain. Efficient to eliminate
processed, it is very significant to remove unnecessary items it may images in the form of blur is the main advantage of the Wiener
hold. After removing unnecessary artifacts, the image can be Filter. Because of working in the frequency domain, it has low
speed and is not suitable for Speckle noise.
D. Hybrid Filter. processing seeks to achieve the goals of eliminating these defects
by accounting for image shape and structure. Generally, this
The Hybrid filter consists both Median filter and Wiener filter. It denotes recognizing objects or boundaries within the image.
can eliminate Speckle noise, Impulse noise and blurring effects Morphological operations are logical conversions based on
from images. But the complexity and time consumption is the main comparison of pixel neighborhoods with a pattern. Usually,
disadvantage of the Hybrid filter. morphological operations are implemented on binary images under
E. Modified Hybrid. Median Filter the pixel values; 0 or 1. Many of the morphological operations
target on binary images.
This filter is also a de-noising filter which consists both Mean and
Median filter. It is very efficient to eliminate Speckle noise, Salt C. K-Means.algorithm
and Pepper noise and the Gaussian noise. But the main disadvantage Most image processing techniques use K-Means algorithm for
of this filter is high time consumption compared to the simple image segmentation. It is very useful for large images with poor
Median filter. contrast. But it has been realized that K-Means is susceptible to
F. Morphology. Based De-noising selection of samples and establishments of fuzzy sets [12] [13].

This filter is based on Morphological operations of opening and

closing. Producing results better than other de-noising filters and
the efficiency are the main advantage of this filter.

4.2 Image.Segmentation

‘Image Segmentation’ is the procedure of distributing an image into ----------- (1)

minor portions. It creates several sets of pixels within same image.
Assigns a tag to every pixel in an image and the pixels with the
Table 1 represents the different segmentation methods with
similar label share particular features [11]. Segmenting makes it
different characteristics.
easier to further analyze and recognize important information form
a digital image. Segmentation
Usage Susceptible To
A..Threshold Segmentation Technique

‘Segmentation’ is the technique that has been introduced to divide Use gradient Hard to be used
a digital image into number of segments that include sets of pixels magnitude to find for images with
Threshold method
and set of super pixels. Objectives to be accomplished through the the potential edge poor contrast.
process of segmentation are simplifying and changing the format pixels [15].
of representation of an image in a way that it will become more
Correctly separate Noise may lead to
detailed , meaningful and easy for the process of analysis. Placing
regions according quality of final
of objects and boundaries in images such as lines, curves could be Region Based
to the similarity of result
performed through Image segmentation. Throughout the procedure
properties [15].
of image segmentation, every pixel in an image is assigned a label
and the pixels consist of same label share certain visual features. Useful for large Sample selection
Each pixel in the region is similar in relation to some features or Fuzzy C Means, K
images with poor and establishing
computed properties, such as color, intensity or texture. Adjoining Means and Level
contrast. fuzzy sets may be
regions are particularly different in regard to the same features. Set Techniques
difficult [16].
Thresholding methodology is the simplest technique of image
segmentation. This technique involves a threshold value that is used TABLE 1. VARIOUS IMAGE SEGMENTATION TECHNIQUES
to converting a gray-scale featured image to a binary image. The
4.3 Feature Extraction.
major advantage of this method is selecting the threshold value to
be used. Accurate tumor extraction is a critical task in the case of brain tumor
due to the complex structure of the brain. There are some criterions
B..Morphological Based Segmentation
that are being considered to extract features such as configuration,
‘Morphology’ refers to describing the properties of the shape and form (shape), size and image location. With respect to the results
structure of any entity. Binary images may comprise many defects. retrieved from extract features the process of tumor classification is
Particularly, the binary regions constructed by simple thresholding performed.
are deformed by texture and noise. Morphological image
A. Edge Detection. V. DISCUSSION.
The most significant thing in image processing is image
An edge happens when there is an sudden and unexpected intensity
segmentation, while diagnosing brain tumor from a digital image.
modification of the image. Whenever it is detected an abrupt
Table 2 represents the different segmentation methods with
modification or a chane in the intensity of a certain image, the
different characteristics such as accuracy and complexity.
associated pixel would be treated as an edge pixel. The algorithm
that has been put forward for the detection of edge pixel supports in
identifying the quality of the edge. But sometimes these edges are
not displayed in the final result. Hence the algorithms are adjusted
to determine the edges[17].

1. “Prewitt” edge detection.

The “Prewitt Mask” is considered as a distinct differentiation

operation. Accordingly, approximated derivative values in both the
directions, such that horizontal and vertical, are calculated using
two 3 × 3 masks [17]. Prewitt masks approximates both horizontal
derivative and the vertical derivative.

2. “Robert edge” detection.

Through the “Roberts edge” detection operation, the image gradient

is estimated via distinct differentiation. In addition, “Robert Mask”
Main goal of Pre-Processing is the edge preservation of the image.
is a matrix and the regions of high spatial frequency are highlighted,
Among the edge detection mechanisms, Sobel is the best option
that often correspond to edges in the image[17].
then both the Gaussian and the Median filter. The different edge
3. “Sobel edge” detection. detection algorithms like ‘Sobel’, ‘Prewitt’ and ‘Robert’ are shown
in the Table 3 with different characteristics such as computation
The “Sobel Mask” is mostly work as the “Prewitt mask”. It can only time in seconds and computation number of flip-flops used.
be distinguish as the Sobel operator has values; ‘2’ and ‘-2’ which
are allocated in the center of 1st and the 3rd columns of the
horizontal mask and 1st and 3rd rows of the vertical mask. Hence
it provides high edge intensity[17].

B. “Histogram of Oriented Gradient” Feature Extraction.

The extraction process of the “Histogram of Oriented Gradient”

(HOG) is having following calculations. First, the pre-processed
cell image will be distributed into “32 × 32” pixels. The intensity of
each pixel is ‘0’ or ‘1’. Then the result will be added to “HOG” [18].
Figure. 4 shows the architecture of “HOG” feature. Then the image
will be distributed into “8 × 8” pixels that is called box. Here, the
box will be already added into a single block. Again each box will
be distributed into 9 bins which is “3 × 3”. Pixel gradient is used for
the creation of the feature in each and every bin[18]. Therefore there
are 9 features, it will lead to “9 × 4” characteristics for each block. TABLE 3. PERFORMANCE OF EDGE DETECTION
In the all “32 × 32” pixels, “HOG” feature extraction allows to
create ‘9 blocks’ and finally, it will having “9 × 9 × 4” features in The following figures show the steps of brain tumor detection using
single dimension or “1 × 324” in the vector image. image processing techniques. That is original MRI image, grayscale
image, filtering image using Median filter, segmenting using
threshold method, morphological operation applied image and
finally diagnosed tumor from MRI image.

Figure 4. Architecture of HOG feature (elongated cell) [18]

 segmentation techniques. It easy to implement and widely used
these days. When the contrast between foreground object and
background object is comparatively high, threshold technique
works well.

VII. ACKNOWLEDGEMENT.
MRI image of tumor Grayscale image I would like to take this great opportunity to thank Dr. Lochandaka
Ranathunga, Head of the Department, Faculty of Information
Technology for his advices and valuable guidance during my
independent study.
I should also thank the Faculty of Information Technology for
giving me the opportunity and also thank all those who gave me the
valuable advices that allowed me to make this independent study a
success.

Median Filtered Threshold Segmented REFERENCE

[1] International Journal of Emerging Trends & technology in Computer
Science (IJETTCS), ISSN 2278-6856, Volume: 04, Issue: 5(2),
September-October 2015.
[2] Brain Tumor: Statistics - Approved by the Cancer.Net Editorial Board,
06/2016. [Online] Available: http://www.cancer.net/cancer-types/brain-
tumor/statistics [Accessed: 06- May- 2017].
[3] V. Shinde, P. Kine, S. Gadge and S. Khatal, “Brain Tumor Identification
using MRI Iamges”, ISSN 2331-8169, Volume: 2 Issue:10, October
2014.
[4] M. S. Al-Tarawnwneh, “Lung Cancer Detection Using Image Processing
Morphological operation Final Tumor detected Techniques,” Leonardo Electro. J. of Practices and Technol., no. 20, pp.
applied image image 147-158, 2012.
[5] H. G. Zadeh, S. Janianpour, and J. Haddadnia, “Recognition and
Figure 5. Steps of brain tumor detection [20] Classification of the Cancer Cells by Using Image Processing and
LabVIEW,” Int. J. of Comput. Theory and Engr., vol. 5, no. 1, pp. 104-
107, 2013.
VI. CONCLUSION. [6] Y. Hu, K. Ashenayi, R. Veltri, G. O'Dowd, G. Miller, R. Hurst, and R.
Abnormal growth of tissues in the brain which affect proper brain Bonner, “A Comparison of Neural Network Classification,” in Proc. of
IEEE International Neural Networks, vol. 6, pp. 3461-3466, 1994.
functions is considered as a brain tumor. The main goal of medical
[7] X. W. Chen, “Gene Selection for Cancer Classification Using
image processing is to identify accurate and meaningful Bootstrapped Genetic Algorithms and Support Vector Machines,” in
information using images with the minimum error possible. Brain Proc. of IEEE Bioinformatics Conference, pp. 504-505, 2013.
tumor identifications through MRI images is a difficult task because [8] Amruta Pramod Hebli, Sudha Gupta “Brain Tumor Detection Using
of the complexity of the brain. These tumors can be segmented Image Processing: A Review”, in 65th IRF International Conference,
978-93-86291-38-7, 2016.
using various image segmentation techniques. The process of
[9] Luxit Kapoor, Prof. Sanjeeev Thakur, “A Survey on Brain Tumor
identifying brain tumors through MRI images can be categorized Detection Using Image Processing Techniques” in Proc. of IEEE
into four different sections; pre-processing, image segmentation, International Cloud computing, Data Science & Engineering, 978-1-
feature extraction and image classification. 5090-3519-9, 2017
Median filter is the most commonly used filtering technique among [10] Rajesh C. patil, A.S. Bhalchandra, “Brain tumor extraction from MRI
images Using MAT Lab”, IJECSCSE, ISSN: 2277-9477, Volume 2,
various filtering techniques. Less complexity and the efficiency in
issue1.
eliminating ‘Salt and Pepper noise’ are the main advantages of [11] Priyanka, balwinder Singh, “A review on Brain tumor detection using
median filter. Not like Gaussian filter, it is a non-linear filter, segmentation”, IJCSMC, Vol. 2, Issue, pg. 48-54. 7, July 2013.
Median filter is an edge preserving filter. Also, Gaussian filter is a [12] Gopal,N.N. Karnan, M. , ―Diagnose brain tumor through MRI using
low pass filter hence the edge information will be lost and edges image processing clustering algorithms such as Fuzzy C Means along
getting displaced and blurred. Although, less complexity and the with intelligent optimization techniques Page(s): 1 – 4, Computational
Intelligence and Computing Research (ICCIC), 2010 IEEE International
cheapness to implement than the Median filter are the main
Conference, 28-29 Dec. 2010.
advantages of Gaussian filter. Another advantage is the Gaussian [13] M. Schmidt, I. Levner, R. Greiner, et al. Segmenting brain tumors using
filter is very applicable in smoothening Gaussian noise. alignment-based features[C]. Proceedings of the Fourth International
Thresholding is the best and easiest approach among image Conference on Machine Learning and Applications, 2005, 6 pp.
[14] Ehab F. Badran, Esraa Galal Mahmoud, and Nadder Hamdy, “An
Algorithm for Detecting Brain Tumors in MRI Images”. IEEE
International Conference on Computer engineering and
Systems(ICCES), pp: 368 - 373, 2010.
[15] Vipin Y. Borole, Sunil S. Nimbhore, Dr. Seema S. Kawthekar, “Image
Processing Techniques for Brain Tumor Detection: A Review”, in
International Journal of Emerging Trends & Technology in Computer
Science (IJETTCS), ISSN 2278-6856, 2015.
[16] Luxit Kapoor, Prof. Sanjeeev Thakur, “A Survey on Brain Tumor
Detection Using Image Processing Techniques” in Proc. of IEEE
International Cloud computing, Data Science & Engineering, 978-1-
5090-3519-9, 2017.
[17] William Thomas H. M., S. C. Prasanna Kumar, “A review of
segmentation and edge detection methods for real time image processing
used to detect brain tumor” in IEEE International Conference on
Computational intelligence and Computing Research, 978-1-4799-7849-
6, 2015.
[18] Warintorn Phusomsai, Wiyada Punjaruk “Brain Tumor Cell Recognition
Schemes using Image Processing with Parallel ELM Classifications on
GPU” in 13th IEEE International Joint Conference on Computer Science
and Software Engineering (JCSSE),978-1-5090-2033-1, 2016.
[19] N. Dalal and B. Triggs, “Histogram of Oriented Gradients for Human
Detection”, in Proc. of Int. Conference on Computer Vision & Pattern
Recognition, volume 1, pp. 886-893, 2005.
[20] Natarajan P., Krishnan N., Natasha Sandeep Kenkre, Shraiya Nancy,
Bhuvanesh Pratap Singh, “Tumor Detection using threshold operation in
MRI Brain Images”, in IEEE International Conference on
Computational Intelligence and Computing Research, 2012.

View publication stats