Development and Comparison of Image Fusion Techniques for CT&MRI Images

Fatima. Int. Journal of Engineering Research and Application www.ijera.com
ISSN : 2248-9622, Vol. 6, Issue 7, ( Part -5) July 2016, pp.59-62
www.ijera.com 59|P a g e
Development and Comparison of Image Fusion Techniques for
CT&MRI Images
Fatima*, Anitha Kulkarni**
*(Department of Electronics and Instrumentation, VNR VJIET, Hyderabad
** (Associate Professor, Department of Electronics and Instrumentation, VNR VJIET, Hyderabad
ABSTRACT
Image processing techniques primarily focus upon enhancing the quality of an image or a set ofimages to derive
the maximum information from them. Image Fusion is a technique of producing a superior quality image from a
set of available images. It is the process of combining relevant information from two or more images into a
single image wherein the resulting image will be more informative and complete than any of the input images. A
lot of research is being done in this field encompassing areas of Computer Vision, Automatic object detection,
Image processing, parallel and distributed processing, Robotics and remote sensing. This project paves way to
explain the theoretical and implementation issues of seven image fusion algorithms and the experimental results
of the same. The fusion algorithms would be assessed based on the study and development of some image
quality metrics.
Keywords: Average Difference (AD), Laplacian Mean Square Error (LMSE), Maximum Difference (MD),
Mean Square Error (MSE), Normalized Absolute Error (NAE), Normalized Cross Correlation (NCC), Peak
Signal to Noise Ratio (PSNR), Principal Component Analysis Method (PCA), Structural Content (SC),
Structural Similarity Index Metric (SSIM).
I. INTRODUCTION
Any piece of information makes sense
only when it is able to convey the content across.
The clarity of information is important. Image
Fusion is a mechanism to improve the quality of
information from a set of images. By the process of
image fusion the good information from each of the
given images is fused together to form a resultant
image whose quality is superior to any of the input
images. This is achieved by applying asequence of
operators on the images that would make the good
information in each of the image prominent. The
resultant image is formed by combining such
magnified information from the input images into a
single image. Image Fusion finds it application in
vast range of areas. It is used for medical
diagnostics and treatment [1]. A patient’s images in
different data formats can be fused. These forms
can include magnetic resonance image (MRI),
computed tomography (CT), and positron emission
tomography (PET). In radiologyand radiation
oncology, these images serve different purposes.
For example, CT images are used more
Often to ascertain differences in tissue density
while MRI images aretypically used to diagnose
brain tumors[5]. Image fusion is also used in the
field of remote sensing wherein multivariate
images like thermal images, IR Images, UV
Images, ordinary optical image etc. can be fused
together to get a better image taken from a satellite
[8]. The project mainly required the study and
implementation of the following 4 algorithms of
Image Fusion [1] [2] [3].
Averaging method
Select Maximum method
Select Minimum method
Principal Component Analysis Method
The project also required the development of the
following 9 Image Quality Metrics to assess the
quality of the fused images with respect to a sample
perfect image for a given pair of input images [10].
Mean Square Error (MSE)
Peak Signal to Noise Ratio (PSNR)
Average Difference (AD)
Normalized Cross Correlation (NCC)
Maximum Difference (MD)
Normalized Absolute Error (NAE)
Laplacian Mean Square Error (LMSE)
Structural Content (SC)
Structural Similarity Index Metric (SSIM)
II. IMAGE FUSION ALGORITHMS
Image Fusion method can be divided into
two groups. 1. Spatial domain fusion and 2.
Transform domain fusion.
Spatial domain fusion directly deals with
pixels of input images [4]. The fusion methods
such as simple maximum, simple minimum,
average and principal component analysis (PCA)
fall under spatial domain approaches.
RESEARCH ARTICLE OPEN ACCESS

a) Select Maximum Method: In this method, the
resultant fused image is obtained by selecting the
maximum intensity of corresponding pixels from
both the input images.
𝐹 𝑖, 𝑗 = 𝑚𝑎𝑥𝐴(𝑖, 𝑗)𝐵(𝑖, 𝑗)
𝑛
𝑗 =0
𝑚
𝑖=0
(1)
where A (i , j) and B (i , j) are two input images
and F(i ,j ) is fused image.
b) Select Minimum Method: In this method, the
resultant fused image is obtained by selecting the
minimum intensity of corresponding pixels from
both the input images.
𝐹 𝑖, 𝑗 = 𝑚𝑖𝑛𝐴(𝑖, 𝑗)𝐵(𝑖, 𝑗)
𝑛
𝑗 =0
𝑚
𝑖=0
(2)
where A (i , j) and B (i , j) are two input images
and F(i ,j ) is fused image.
c) Simple Average Method: In this method the
resultant fused image is obtained by taking the
average intensity of corresponding pixels from both
the input images.
𝐹 𝑖, 𝑗 = 𝐴 𝑖, 𝑗 + 𝐵 𝑖, 𝑗 /2 (3)
d) Principal Component Analysis (PCA):
Principal Component Analysis is a vector space
transform often used to reduce multidimensional
data sets to lower dimensions for analysis. It is the
simplest and most useful of the true Eigen vector
based multivariate analyses, because itsoperation is
to reveal the internal structure of data in an
unbiased way. It is mostly used as a tool
in exploratory data analysis and for
making predictive models [6] [7].
III. FUSION RESULTS
Source Images
a)CT Image b)MRI Image
c)Maximum Method d)Minimum Method

e)Average Method f)PCA Method
IV. IMAGE QUALITY METRICS
The general requirement of an image
fusion process is to preserve all valid and usefull
information from the source images, while at the
same time it should not introduce any distortion in
resultant fused image [9] [10].
a) Mean Square Error (MSE): Mean square error
is one of the most commonly used error projection
method where, the error value is the value
difference between the actual data and the resultant
data.b) Peak Signal to Noise Ratio (PSNR):
Defined as log of the ratio between the square of
the peak value to the Mean Square Error multiplied
to the value 10. This basically projects the ratio of
the highest possible value of the data to the error
obtained in the data.
c) Average Difference (AD):Average Difference,
as explained by the term itself, is the average value
of the differencebetween the actual/ideal data and
the obtained/resultant data.
d) Structural Content (SC): Here the ratio
between the content of the both the expected and
the obtained data. Practically, it is the ratio between
the net sum of the square of the expected data and
the net sum of square of the obtained data.
e) Normalized Cross Correlation (NCC): Here a
cross correlation is performed between the
expected data and the obtained data and normalized
with respect to the expected data.
f) Maximum Difference (MD): Maximum
Difference is a very simple metric that gives us the
information of the largest of thecorresponding pixel
error.
g) Normalized Absolute Error (NAE): This is a
metric where the error value is normalized with
respect to the expected or the perfect data. That is,
the net sum ratio between the error values and the
perfect values is calculated. The net sum of the
error value which is the difference between the
expected values and the actual obtained values is
divided by the net sum of the expected values.
i) Laplacian Mean Square Error (LMSE):
Laplacian Mean Square Error, as explained by the
term, is the normal mean square errorcalculation.
But the difference here is that the mean square
error is calculated not based on the expected and
obtained data but basedon the Laplacian value of
the same.
j) Structural Similarity Index Metric (SSIM):
The Structural Similarity Index measures the
similarity between two images.
V. CONCLUSION
The four image fusion techniques were
implemented using MATLAB 2016. Thefusion was
performed on a set of input pair of images. The
fused images were verified for their quality based
on a perfect image ineach of the sets. A set of 9
image metrics were developed to assess the fused
image quality.
In the total of four image fusion
techniques, three very basic fusion techniques
wereAveraging Method, Maximum Selection
Method and Minimum Selection Method and a
Principal Component Analysis (PCA) Method. By
the means of the 9 image metrics developed - MSE,
PSNR,SC, NCC, AD, MD, NAE, LMSE and
SSIM, the Principal Component Method was
assessed as the fusion algorithm producing a fused
image of superior quality compared to the other
three.
The project does hold scope for further
advancements as a lot of research ishappening in
the field. The following are some proposed
practical advancements possible in the project:
Multi Wavelets based image fusion can be
performed to achieve a better image fusion quality.
The image fusion quality has been assessed based
on optical image sets with a perfect image.
Image Registration has not been incorporated in
the project. Image Registration /Image Alignment
will certainly enhance the efficiency of the project
as vast set of even unregistered images can be
considered as set of input images. It would also
help in possibility of more set of sample
test/perfect images made available for assessing the
image fusion algorithms.
REFRENCES
[1]. Kusum Rani, Reecha Sharma. “Study of
Different Image Fusion Algorithm”.
International Journal of Emerging
Tevhnology and Advanced
Engineering(IJETAE). ISSN 2250-2459,
ISO 9001:2008 Certified Journal, Volume
3, Issue 5, May 2013
[2]. Mingjing Li, Yubing Dong. “Review on
Technology of Pixel-level Image Fusion”.
International Conference on Measurement,
Information and Control. 978-1-4799-
1392-3/13. IEEE 2013
[3]. Deepak Kumar Sahu , M.P.Parsa.
“Different Image Fusion Techniques –A

Critical Review”. International Journal of
Modern Engineering Research (IJMER)
www.ijmer.com Vol. 2, Issue. 5, Sep.-Oct.
2012 pp-4298-4301 ISSN: 2249-6645
[4]. Deepali A.Godse, Dattatraya S. Bormane
(2011) “Wavelet based image fusion using
pixel based maximum selection rule”.
International Journal of Engineering
Science and Technology (IJEST), Vol. 3
No. 7 July 2011, ISSN : 0975-5462
[5]. Anjali Malviya, S. G. Bhirud . ”Image
Fusion of Digital Images” International
Journal of Recent Trends in Engineering,
Vol 2, No. 3, November 2009
[6]. Jonathon Shlens, “A Tutorial on Principal
Component Analysis”. Center for Neural
Science, New York University New York
City, NY 10003-6603 and Systems
Neurobiology Laboratory, Salk Institute
for Biological Studies La Jolla, CA 92037
[7]. Ujwala Patil, Uma Mudengudi. “Image
Fusion using hierarchical PCA”.
International Conference on Image
Information Processing (ICIIP 2011). 978-
1-61284-861-7/11 IEEE 2011.
[8]. Z.Wang, Y. Tie and Y. Liu, “Design and
Implementation of Image Fusion System”.
International Conference on Computer
Application and System Modelling
(ICCASM), 2010
[9]. Dr. M. Sumathi, R. Barani. “Qualitative
Evaluation of Pixel Level Image Fusion
Algorithms”. IEEE transaction on Pattern
Recognition, Informatics and Medical
Engineering, March 21-23, 2012
[10]. Z. Wang, A. C. Bovik, H. R. Sheikh, and
E. P. Simoncelli, "Image quality
assessment: From error visibility to
structural similarity“, IEEE Transactions
on Image Processing,vol. 13, no. 4,
pp.600-612, Apr. 2004

Development and Comparison of Image Fusion Techniques for CT&MRI Images

More Related Content

Development and Comparison of Image Fusion Techniques for CT&MRI Images