SlideShare a Scribd company logo
International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016]
Infogain Publication (Infogainpublication.com) ISSN: 2454-1311
www.ijaems.com Page | 411
Brain Image Fusion using DWT and Laplacian
Pyramid Approach and Tumor Detection using
Watershed Segmentation
Susan Mary Olakkengil1
, Mrs. Prathima. M.N2
1
Department of Information Science & Eng, MS Ramaiah Institute of Technology, Bengaluru, India
2
Associate Professor, Department of Information Science & Eng, MS Ramaiah Institute of Technology, Bengaluru, India
Abstract— Image fusion is the process of combining
important information from two or more images into a
single image. The resulting image will be more enhanced
than any of the input pictures. The idea of combining
multiple image modalities to furnish a single, more
enhanced image is well established, special fusion methods
have been proposed in literature. This paper is based on
image fusion using laplacian pyramid and Discreet Wavelet
Transform (DWT) methods. This system uses an easy and
effective algorithm for multi-focus image fusion which uses
fusion rules to create fused image. Subsequently, the fused
image is obtained by applying inverse discreet wavelet
transform. After fused image is obtained, watershed
segmentation algorithm is applied to detect the tumor part
in fused image.
Keywords— Image Fusion, Laplacian Pyramid Algorithm,
Discreet Wavelet Transform.
I. INTRODUCTION
Image fusion is the system of combining significant
information from two or more images into a single image.
The resultant image will be more enhanced than any of the
input images. The concept of fusion image has been utilized
in wide style of functions like remedy, remote sensing,
machine vision, automotive changes detection, bio metrics
etc. With the emergence of various image capturing devices,
it's not possible to acquire an image with all the information.
Generally, an entire image is probably not consistently
feasible since optical lenses of imaging sensor primarily with
long focal lengths, only have a restricted depth of field.
Image fusion helps to receive an image with all the expertise.
Image fusion is a concept of combining multiple images into
composite products, by way of which extra information than
that of individual input image will be revealed. A brain
tumour is a group (or mass) of abnormal cells in the brain.
The skull is very rigid and the brain is enclosed, so any
progress within the sort of limited area can cause problems.
Brain tumours can also be cancerous (malignant) or non-
cancerous (benign). Symptoms of a brain tumour can also be
general or specific. A common symptom is caused by the
strain of the tumour on the mind or spinal cord. Particular
symptoms are cased when a designated a part of the mind is
not working mainly in view that of the tumour. To determine
the sort of tumour, imaging of the brain is predominant. The
reply to which imaging modality is better for imaging the
brain is based on the rationale of the examination.
The paper uses wavelet based image fusion procedure for
brain tumor detection. Wavelet analysis is commonly used
method for solving difficult issues in mathematics, physics
and engineering in the present day times. It has diverse
applications within the fields of wave propagation, data
compression, signal processing, image processing, pattern
recognition and medical imaging technologies. Wavelet
analysis decomposes complex information to fundamental
varieties at different positions and scales, which can be
conveniently reconstructed with high precision. Wavelet
transform is a powerful tool within the analysis of indicators
in comparison with Fourier transform, given that the later
approach fails in the evaluation of non stationary signals.
The paper is organized as follows: Section 2: explains
literature survey, different methodologies on brain tumor
detection using DWT and Laplacian Pyramid algorithm;
Section 3: contains methodology of proposed system which
includes pre-processing, fused image creation, post
processing and segmentation to detect the brain tumor.
Section 4: depicts the experimental results obtained from the
evaluation of the proposed methods. Section 5: finally,
conclusions are drawn in.
II. RELATED WORK
Ambily P.K, Shine P.James and Remya R.Mohan proposed
a system which detects the brain tumor part in a fused image.
Fused image and neural network is segmented to extract the
tumor part in image. Initially MRI image and CT images are
pre-processed and de-compose the image using discreet
wavelet transform. The fused image is segmented to detect
the tumor part. In this paper [2], Vivek Angoth, CYN Dwith
and Amarjot Singh proposed an efficient tumor detection
algorithm which uses complementary and redundant
International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016]
Infogain Publication (Infogainpublication.com) ISSN: 2454-1311
www.ijaems.com Page | 412
information from the computer tomography (CT) and
Magnetic resonance imaging (MRI).
S.L. Jany Shabu, Dr.C. Jayakumar and T. Surya reviewed
different image fusion techniques [3] for brain tumor
detection. This paper presents the review and comparison of
various image fusion methods. They have find out various
issues in different techniques and to remove that Genetic
Algorithm is used. A. P. Jamesa and B. V. Dasarathyb [4]
have surveyed on image fusion approach, they Characterize
the medical image fusion study based on (1) the extensively
used image fusion ways, (2) imaging modalities, and (3)
imaging of organs which are under study. This review
concludes that despite the fact that there exists a number of
open ended technological and scientific challenges, the
fusion of scientific images has proved to be useful for
advancing the scientific reliability of using medical imaging
for medical diagnostics and analysis, and is a scientific self-
discipline that has the expertise to vastly develop in the
coming years.
S. Anbumozhi and P. S. Manoharan [5] proposed a
methodology to classify the tumors such as benign and
malignant. This paper integrates two images of same type to
get the fused image; fuzzy logic is used to fuse the images.
Multilevel Adaptive Neuro was used as a classifier. S.M.
Mukane, Y.S. Ghodake and P.S. Khandagle [6] described a
methodology to enhance the image by combining wavelet
transform and Laplacian Pyramid algorithm. Performance of
image fusion technique is measured by mean square error,
normalized absolute error and peak signal to noise ratio.
From the performance analysis it has been observed that
MSE is decreased in case of both the methods where as
PSNR increased, NAE decreased in case of laplacian
pyramid where as constant for wavelet transform method.
III. PROPOSED SYSTEM
Accurate detection of size and location of brain tumor plays
a vital role in the diagnosis of tumor. An efficient wavelet
based algorithm is used for tumor detection which utilizes
the complementary and redundant information from the
Computed Tomography (CT) image and Magnetic
Resonance Imaging (MRI) images. Hence this algorithm
effectively uses the information provided by the CT image
and MRI images there by providing a resultant fused image
which increases the efficiency of tumour detection. Figure 1
shows the architecture of proposed system. Block-wise
explanation of the system is given below.
Fig.1: Block Diagram of Proposed Work
The algorithm decomposes the input image using 2D-
Disreet wavelet transform. The lower approximations are
subjected to Laplacian pyramid algorithm. The SF
algorithm combined with wavelet fusion algorithm is used
for higher approximations. The new sets of detailed and
approximate coefficients from each image are then added
to get the new fused coefficients. The final step performs
Inverse DWT with the new coefficients to construct the
fused image. The two main components of the proposed
algorithm are the Laplacian Pyramid algorithm and the
wavelet algorithm and are explained in the following sub-
sections.
A. Pre-processing
A MRI and a CT image is pre processed and taken as
source images A and B. Both the images are subjected to
wavelet decomposition which decomposes into
approximation, horizontal detail, vertical detail and
diagonal detail respectively. This is a stage where the input
image taken and resized to 265x256 standard size. Input
International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016]
Infogain Publication (Infogainpublication.com) ISSN: 2454-1311
www.ijaems.com Page | 413
image is in RGB color format; such images have to be
converted to gray scale images using rgb2gray command.
B. Discreet Wavelet Transform
The mathematical basis of the wavelet transform is the
Fourier transform. In the wavelet analysis, the size of the
window is constant while the shape is changeable, as well
as the time window and the frequency windows. As a
result, wavelet analysis has respectively better resolution
but yet worse time resolution in low frequency band,and
vice versa. Discrete wavelet transform (DWT) is likely
one of the most widespread methods for the
decomposition of an image, which has been widely used in
a large number of researches.
The preceding result is applied to the next stage of 2-D
DWT decomposition and this process is going to be
repeated recursively in each stage. The low-pass filter h
and high-pass filter g correspond to a particular type of
wavelet used.
Image fusion using wavelet scheme decomposes source
images ( , ) and ( , ) into approximation and
detailed coefficients at required level using DWT. The
approximation and detailed coefficients of both images are
combined using fusion rule. The fused image could be
obtained by taking the inverse discrete wavelet transform
(IDWT) as eq. (1):
( , ) =
( , ) + ( , )
2
The fusion rule used here is simply averages the
approximation coefficients and picks the detailed
coefficient in each sub band with the largest magnitude.
(a)
G- High Level Filter
H- Low Level Filter
C. Laplacian Pyramid Algorithm
The Laplacian Pyramid [7] implements a pattern selective‖
process for image fusion, so that the composite image is
built not based on a pixel at a time, but a feature at a time.
The fundamental suggestion is to participate in a pyramid
decomposition on every source image, then combine all
these decompositions to form a composite representation,
and subsequently reconstruct the fused image by way of
performing an inverse pyramid transform.
(b)
Fig.2: Decomposition of Image using DWT
The overall algorithm is proven in figure 1. The first step
is to assemble a pyramid for each source image. The
fusion is then applied for each and every level of the
pyramid using a feature decision mechanism. The feature
selection process selects essentially the most salient
sample from the source and copies it to the composite
pyramid, at the same time discarding the least significant
salient pattern. In this way, the entire locations the place
the source images are distinctly selected. The salient
aspect is selected using Equation (1).
( , ) =
( , ) | ( , )| > | ( , )|
( , ) ℎ !" #
$ (1)
Where A and B are input images and F is the fused
image and 0 ≤ ≤ ( − 1
D. Segmentation: Watershed Algorithm
This is one of the best methods to group pixels of an
image on the basis of their intensities. Pixels falling under
identical intensities are grouped together. It is an excellent
segmentation approach for dividing an image to separate a
tumor from the image Watershed is a mathematical
morphological operating device. Watershed is normally
used for checking output rather than using as an input
segmentation method since it in general suffers from over
segmentation and under segmentation [8].
For using watershed segmentation specific approaches are
used. Two common principle approaches are given under:
1) the computed local minima of the image gradient are
chosen as a marker. In this system an over segmenta.tion
happens. After making a choice on marker region merging
is completed as a second step; 2) Watershed
transformation using markers utilizes particular defined
marker positions. These positions are either outlined
explicitly by using a user or they can be decided
mechanically with the help of utilizing morphological
tools.
International Journal of Advanced Engineering, Management and Science (IJAEMS)
Infogain Publication (Infogainpublication.com
www.ijaems.com
IV. RESULTS AND DISCUSSION
The experimental results of our proposed work
in the figure below.
(a)
Fig.3 (a) Input Image A (b) Input Image
(a)
Fig.4 (a) Fused Image, (b) Segmented Image
Fig.5:Tumor Detected Image
V. CONCLUSION
An efficient methodology is proposed to detect tumor part
in an image using fusion based methods. Initially we take
two images and produce the fused image using dis
wavelet transform and laplacian pyramid algorithm.
Inverse DWT applied to get back fused
watershed algorithm is applied to detect the tumor part in
an MRI image.
REFERENCES
[1] Ambily P.K., Shine P.James and Remya R.Mohan,
“Brain Tumor Detection using Image Fusion and
Neural Network”, Volume 3, Issue 2, 2015.
[2] Vivek Angoth, CYN Dwith and Amarjot Singh, “
Novel Wavelet Based Image Fusion for Brain Tumor
Detection”, 2013.
Journal of Advanced Engineering, Management and Science (IJAEMS)
Infogainpublication.com)
DISCUSSION
of our proposed work are shown
(b)
Input Image B
(b)
Segmented Image
Tumor Detected Image
is proposed to detect tumor part
Initially we take
two images and produce the fused image using discreet
wavelet transform and laplacian pyramid algorithm.
fused image. Finally
watershed algorithm is applied to detect the tumor part in
Ambily P.K., Shine P.James and Remya R.Mohan,
“Brain Tumor Detection using Image Fusion and
Neural Network”, Volume 3, Issue 2, 2015.
Amarjot Singh, “A
Image Fusion for Brain Tumor
[3] L. Jany Shabu, Dr.C. Jayakumar and T. Surya,
“Survey of Image Fusion Techniques for Brain
Tumor Detection”, Volume 3, Issue 2, 2013.
[4] A. P. James and B. V. Dasarathy, “
Fusion: A survey of the state of the art
[5] S. Anbumozhi and P. S. Manoharan
Analysis of Brain Tumor Detection
Fusion”, 2014.
[6] S.M.Mukane, Y.S.Ghodake and P.S.Khandagle,
“Image enhancement using fusion by
transform and laplacian
[7] Wencheng Wang and Faliang Chang, “
Image Fusion Method Based on
Volume 6, issue 12, 2011
[8] Gang Li, “Improved Watershed Segmentation
Optimal Scale Based on Ordered Dither Halftone
Mutual Information”, pp 296
[Vol-2, Issue-5, May- 2016]
ISSN: 2454-1311
Page | 414
L. Jany Shabu, Dr.C. Jayakumar and T. Surya,
Survey of Image Fusion Techniques for Brain
Volume 3, Issue 2, 2013.
A. P. James and B. V. Dasarathy, “Medical Image
Fusion: A survey of the state of the art”, 2014.
S. Anbumozhi and P. S. Manoharan, “Performance
Analysis of Brain Tumor Detection Based On Image
S.M.Mukane, Y.S.Ghodake and P.S.Khandagle,
Image enhancement using fusion by wavelet
pyramid”.
Wencheng Wang and Faliang Chang, “A Multi-focus
Image Fusion Method Based on Laplacian Pyramid”,
12, 2011.
Improved Watershed Segmentation with
n Ordered Dither Halftone and
on”, pp 296-300, 2010.

More Related Content

Brain Image Fusion using DWT and Laplacian Pyramid Approach and Tumor Detection using Watershed Segmentation

  • 1. International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016] Infogain Publication (Infogainpublication.com) ISSN: 2454-1311 www.ijaems.com Page | 411 Brain Image Fusion using DWT and Laplacian Pyramid Approach and Tumor Detection using Watershed Segmentation Susan Mary Olakkengil1 , Mrs. Prathima. M.N2 1 Department of Information Science & Eng, MS Ramaiah Institute of Technology, Bengaluru, India 2 Associate Professor, Department of Information Science & Eng, MS Ramaiah Institute of Technology, Bengaluru, India Abstract— Image fusion is the process of combining important information from two or more images into a single image. The resulting image will be more enhanced than any of the input pictures. The idea of combining multiple image modalities to furnish a single, more enhanced image is well established, special fusion methods have been proposed in literature. This paper is based on image fusion using laplacian pyramid and Discreet Wavelet Transform (DWT) methods. This system uses an easy and effective algorithm for multi-focus image fusion which uses fusion rules to create fused image. Subsequently, the fused image is obtained by applying inverse discreet wavelet transform. After fused image is obtained, watershed segmentation algorithm is applied to detect the tumor part in fused image. Keywords— Image Fusion, Laplacian Pyramid Algorithm, Discreet Wavelet Transform. I. INTRODUCTION Image fusion is the system of combining significant information from two or more images into a single image. The resultant image will be more enhanced than any of the input images. The concept of fusion image has been utilized in wide style of functions like remedy, remote sensing, machine vision, automotive changes detection, bio metrics etc. With the emergence of various image capturing devices, it's not possible to acquire an image with all the information. Generally, an entire image is probably not consistently feasible since optical lenses of imaging sensor primarily with long focal lengths, only have a restricted depth of field. Image fusion helps to receive an image with all the expertise. Image fusion is a concept of combining multiple images into composite products, by way of which extra information than that of individual input image will be revealed. A brain tumour is a group (or mass) of abnormal cells in the brain. The skull is very rigid and the brain is enclosed, so any progress within the sort of limited area can cause problems. Brain tumours can also be cancerous (malignant) or non- cancerous (benign). Symptoms of a brain tumour can also be general or specific. A common symptom is caused by the strain of the tumour on the mind or spinal cord. Particular symptoms are cased when a designated a part of the mind is not working mainly in view that of the tumour. To determine the sort of tumour, imaging of the brain is predominant. The reply to which imaging modality is better for imaging the brain is based on the rationale of the examination. The paper uses wavelet based image fusion procedure for brain tumor detection. Wavelet analysis is commonly used method for solving difficult issues in mathematics, physics and engineering in the present day times. It has diverse applications within the fields of wave propagation, data compression, signal processing, image processing, pattern recognition and medical imaging technologies. Wavelet analysis decomposes complex information to fundamental varieties at different positions and scales, which can be conveniently reconstructed with high precision. Wavelet transform is a powerful tool within the analysis of indicators in comparison with Fourier transform, given that the later approach fails in the evaluation of non stationary signals. The paper is organized as follows: Section 2: explains literature survey, different methodologies on brain tumor detection using DWT and Laplacian Pyramid algorithm; Section 3: contains methodology of proposed system which includes pre-processing, fused image creation, post processing and segmentation to detect the brain tumor. Section 4: depicts the experimental results obtained from the evaluation of the proposed methods. Section 5: finally, conclusions are drawn in. II. RELATED WORK Ambily P.K, Shine P.James and Remya R.Mohan proposed a system which detects the brain tumor part in a fused image. Fused image and neural network is segmented to extract the tumor part in image. Initially MRI image and CT images are pre-processed and de-compose the image using discreet wavelet transform. The fused image is segmented to detect the tumor part. In this paper [2], Vivek Angoth, CYN Dwith and Amarjot Singh proposed an efficient tumor detection algorithm which uses complementary and redundant
  • 2. International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016] Infogain Publication (Infogainpublication.com) ISSN: 2454-1311 www.ijaems.com Page | 412 information from the computer tomography (CT) and Magnetic resonance imaging (MRI). S.L. Jany Shabu, Dr.C. Jayakumar and T. Surya reviewed different image fusion techniques [3] for brain tumor detection. This paper presents the review and comparison of various image fusion methods. They have find out various issues in different techniques and to remove that Genetic Algorithm is used. A. P. Jamesa and B. V. Dasarathyb [4] have surveyed on image fusion approach, they Characterize the medical image fusion study based on (1) the extensively used image fusion ways, (2) imaging modalities, and (3) imaging of organs which are under study. This review concludes that despite the fact that there exists a number of open ended technological and scientific challenges, the fusion of scientific images has proved to be useful for advancing the scientific reliability of using medical imaging for medical diagnostics and analysis, and is a scientific self- discipline that has the expertise to vastly develop in the coming years. S. Anbumozhi and P. S. Manoharan [5] proposed a methodology to classify the tumors such as benign and malignant. This paper integrates two images of same type to get the fused image; fuzzy logic is used to fuse the images. Multilevel Adaptive Neuro was used as a classifier. S.M. Mukane, Y.S. Ghodake and P.S. Khandagle [6] described a methodology to enhance the image by combining wavelet transform and Laplacian Pyramid algorithm. Performance of image fusion technique is measured by mean square error, normalized absolute error and peak signal to noise ratio. From the performance analysis it has been observed that MSE is decreased in case of both the methods where as PSNR increased, NAE decreased in case of laplacian pyramid where as constant for wavelet transform method. III. PROPOSED SYSTEM Accurate detection of size and location of brain tumor plays a vital role in the diagnosis of tumor. An efficient wavelet based algorithm is used for tumor detection which utilizes the complementary and redundant information from the Computed Tomography (CT) image and Magnetic Resonance Imaging (MRI) images. Hence this algorithm effectively uses the information provided by the CT image and MRI images there by providing a resultant fused image which increases the efficiency of tumour detection. Figure 1 shows the architecture of proposed system. Block-wise explanation of the system is given below. Fig.1: Block Diagram of Proposed Work The algorithm decomposes the input image using 2D- Disreet wavelet transform. The lower approximations are subjected to Laplacian pyramid algorithm. The SF algorithm combined with wavelet fusion algorithm is used for higher approximations. The new sets of detailed and approximate coefficients from each image are then added to get the new fused coefficients. The final step performs Inverse DWT with the new coefficients to construct the fused image. The two main components of the proposed algorithm are the Laplacian Pyramid algorithm and the wavelet algorithm and are explained in the following sub- sections. A. Pre-processing A MRI and a CT image is pre processed and taken as source images A and B. Both the images are subjected to wavelet decomposition which decomposes into approximation, horizontal detail, vertical detail and diagonal detail respectively. This is a stage where the input image taken and resized to 265x256 standard size. Input
  • 3. International Journal of Advanced Engineering, Management and Science (IJAEMS) [Vol-2, Issue-5, May- 2016] Infogain Publication (Infogainpublication.com) ISSN: 2454-1311 www.ijaems.com Page | 413 image is in RGB color format; such images have to be converted to gray scale images using rgb2gray command. B. Discreet Wavelet Transform The mathematical basis of the wavelet transform is the Fourier transform. In the wavelet analysis, the size of the window is constant while the shape is changeable, as well as the time window and the frequency windows. As a result, wavelet analysis has respectively better resolution but yet worse time resolution in low frequency band,and vice versa. Discrete wavelet transform (DWT) is likely one of the most widespread methods for the decomposition of an image, which has been widely used in a large number of researches. The preceding result is applied to the next stage of 2-D DWT decomposition and this process is going to be repeated recursively in each stage. The low-pass filter h and high-pass filter g correspond to a particular type of wavelet used. Image fusion using wavelet scheme decomposes source images ( , ) and ( , ) into approximation and detailed coefficients at required level using DWT. The approximation and detailed coefficients of both images are combined using fusion rule. The fused image could be obtained by taking the inverse discrete wavelet transform (IDWT) as eq. (1): ( , ) = ( , ) + ( , ) 2 The fusion rule used here is simply averages the approximation coefficients and picks the detailed coefficient in each sub band with the largest magnitude. (a) G- High Level Filter H- Low Level Filter C. Laplacian Pyramid Algorithm The Laplacian Pyramid [7] implements a pattern selective‖ process for image fusion, so that the composite image is built not based on a pixel at a time, but a feature at a time. The fundamental suggestion is to participate in a pyramid decomposition on every source image, then combine all these decompositions to form a composite representation, and subsequently reconstruct the fused image by way of performing an inverse pyramid transform. (b) Fig.2: Decomposition of Image using DWT The overall algorithm is proven in figure 1. The first step is to assemble a pyramid for each source image. The fusion is then applied for each and every level of the pyramid using a feature decision mechanism. The feature selection process selects essentially the most salient sample from the source and copies it to the composite pyramid, at the same time discarding the least significant salient pattern. In this way, the entire locations the place the source images are distinctly selected. The salient aspect is selected using Equation (1). ( , ) = ( , ) | ( , )| > | ( , )| ( , ) ℎ !" # $ (1) Where A and B are input images and F is the fused image and 0 ≤ ≤ ( − 1 D. Segmentation: Watershed Algorithm This is one of the best methods to group pixels of an image on the basis of their intensities. Pixels falling under identical intensities are grouped together. It is an excellent segmentation approach for dividing an image to separate a tumor from the image Watershed is a mathematical morphological operating device. Watershed is normally used for checking output rather than using as an input segmentation method since it in general suffers from over segmentation and under segmentation [8]. For using watershed segmentation specific approaches are used. Two common principle approaches are given under: 1) the computed local minima of the image gradient are chosen as a marker. In this system an over segmenta.tion happens. After making a choice on marker region merging is completed as a second step; 2) Watershed transformation using markers utilizes particular defined marker positions. These positions are either outlined explicitly by using a user or they can be decided mechanically with the help of utilizing morphological tools.
  • 4. International Journal of Advanced Engineering, Management and Science (IJAEMS) Infogain Publication (Infogainpublication.com www.ijaems.com IV. RESULTS AND DISCUSSION The experimental results of our proposed work in the figure below. (a) Fig.3 (a) Input Image A (b) Input Image (a) Fig.4 (a) Fused Image, (b) Segmented Image Fig.5:Tumor Detected Image V. CONCLUSION An efficient methodology is proposed to detect tumor part in an image using fusion based methods. Initially we take two images and produce the fused image using dis wavelet transform and laplacian pyramid algorithm. Inverse DWT applied to get back fused watershed algorithm is applied to detect the tumor part in an MRI image. REFERENCES [1] Ambily P.K., Shine P.James and Remya R.Mohan, “Brain Tumor Detection using Image Fusion and Neural Network”, Volume 3, Issue 2, 2015. [2] Vivek Angoth, CYN Dwith and Amarjot Singh, “ Novel Wavelet Based Image Fusion for Brain Tumor Detection”, 2013. Journal of Advanced Engineering, Management and Science (IJAEMS) Infogainpublication.com) DISCUSSION of our proposed work are shown (b) Input Image B (b) Segmented Image Tumor Detected Image is proposed to detect tumor part Initially we take two images and produce the fused image using discreet wavelet transform and laplacian pyramid algorithm. fused image. Finally watershed algorithm is applied to detect the tumor part in Ambily P.K., Shine P.James and Remya R.Mohan, “Brain Tumor Detection using Image Fusion and Neural Network”, Volume 3, Issue 2, 2015. Amarjot Singh, “A Image Fusion for Brain Tumor [3] L. Jany Shabu, Dr.C. Jayakumar and T. Surya, “Survey of Image Fusion Techniques for Brain Tumor Detection”, Volume 3, Issue 2, 2013. [4] A. P. James and B. V. Dasarathy, “ Fusion: A survey of the state of the art [5] S. Anbumozhi and P. S. Manoharan Analysis of Brain Tumor Detection Fusion”, 2014. [6] S.M.Mukane, Y.S.Ghodake and P.S.Khandagle, “Image enhancement using fusion by transform and laplacian [7] Wencheng Wang and Faliang Chang, “ Image Fusion Method Based on Volume 6, issue 12, 2011 [8] Gang Li, “Improved Watershed Segmentation Optimal Scale Based on Ordered Dither Halftone Mutual Information”, pp 296 [Vol-2, Issue-5, May- 2016] ISSN: 2454-1311 Page | 414 L. Jany Shabu, Dr.C. Jayakumar and T. Surya, Survey of Image Fusion Techniques for Brain Volume 3, Issue 2, 2013. A. P. James and B. V. Dasarathy, “Medical Image Fusion: A survey of the state of the art”, 2014. S. Anbumozhi and P. S. Manoharan, “Performance Analysis of Brain Tumor Detection Based On Image S.M.Mukane, Y.S.Ghodake and P.S.Khandagle, Image enhancement using fusion by wavelet pyramid”. Wencheng Wang and Faliang Chang, “A Multi-focus Image Fusion Method Based on Laplacian Pyramid”, 12, 2011. Improved Watershed Segmentation with n Ordered Dither Halftone and on”, pp 296-300, 2010.