International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
Directive Contrast Based Multimodal Medical Image
Fusion in NSCT with DWT Domain
Senthil V PG scholar
Prof. B. Rajesh Kumar M.E.,(Ph.D.),
RVS College of Engineering and Technology
RVS College of Engineering and Technology
Coimbatore
Coimbatore
Tamilnadu
Tamilnadu.
Abstract—Multi modal medical image fusion, as a
powerful tool for the clinical applications, has developed
with the advent of various imaging modalities in
medical imaging. The main motivation is to capture
most relevant information from sources into a Single
output, which plays an important role in medical
diagnosis. In this paper, a novel fusion framework is
proposed for multimodal medical images based on non
Sub sampled contour let transform (NSCT). The source
medical images are first transformed by NSCT followed
by combining low- and high-frequency components.
Two different fusion rules based on phase congruency
and directive contrast are Proposed and used to fuse
low- and high-frequency coefficients. Finally, the fused
image is constructed by the inverse NSCT with all
composite, coefficients. Experimental results and
comparative study show that the proposed fusion frame
work Provides an effective way to enable more accurate
analysis of multimodality images. Further, the
applicability of the proposed framework is carried out
by the three clinical examples of persons affected with
Alzheimer, subacute stroke and recurrent tumor.
Keywords—Low frequency fusion, high frequency
fusion, Inverse NSCT implementation.
INTRODUCTION
Medical imaging has attracted increasing
attention due to its critical role in health care.
However, different types of imaging techniques such
as X-ray, computed tomography(CT), magnetic
ISSN: 2231-5381
resonance imaging (MRI), magnetic resonance
angiography (MRA), etc.,
Provide
information
where
some
information is common, and some are unique. For
example, X-ray and computed tomography (CT) can
provide dense structures like bones and implants with
less distortion, but it cannot detect physiological
changes. Similarly, normal and pathological soft
tissue can be better visualized by MRI image whereas
PET can be used to provide better information on
blood flow and flood activity with low spatial
resolution. As a result, the anatomical and functional
medical images are needed to be combined for a
compendious view. For this purpose, the multimodal
medical image fusion has been identified as a
promising solution which aims to integrating
information from multiple modality images to obtain
a more complete and accurate description of the same
object. Multimodal medical image fusion not only
helps in diagnosing diseases, but it also reduces the
storage cost by reducing storage to a single fused
image instead of multiple-source images. So far,
extensive work has been made on image fusion
technique with various techniques dedicated to
multimodal medical image fusion These techniques
have been categorized into three categories according
to merging stage. These include pixel level, feature
level and decision level fusion where medical image
fusion usually employs the pixel level fusion due to
the advantage of containing the original measured
quantities, easy implementation and computationally
efficiency .Hence, in this paper, we concentrate our
http://www.ijettjournal.org
Page 288
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
efforts to pixel level fusion, and the terms image
fusion or fusion are intently used for pixel level
fusion. The well-known pixel level fusion are based
on principal component analysis (PCA), independent
component analysis (ICA), contrast pyramid (CP),
gradient pyramid (GP) filtering, etc. Since, the image
features are sensitive to the human visual system
exists in different scales. Therefore, these are not the
highly suitable for medical image fusion . Recently,
with the development of multi scale decomposition,
wavelet transform has been identified ideal method
for image fusion. However, it is argued that wavelet
decomposition is good at isolated discontinuities, but
not good at edges and textured region. As a result, the
anatomical and functional medical images are needed
to be combined for a compendious view. For this
purpose, the multimodal medical image fusion has
been identified as a promising solution which aims to
integrating information from multiple modality
images to obtain a more complete and accurate
description of the same object. Multimodal medical
image fusion not only helps in diagnosing diseases,
but it also reduces the storage cost by reducing
storage to a single fused image instead of multiplesource images . These techniques have been
categorized into three categories according to
merging stage These include pixel level , feature
level and decision level fusion where medical image
fusion usually employs the pixel level fusion due to
the advantage of containing the original measured
quantities , easy implementation and computationally
efficiency we concentrate our efforts to pixel level
fusion, and the terms image fusion or fusion are
intently used for pixel level fusion . The well-known
pixel level fusion are based on principal component
analysis (PCA), independent component analysis
(ICA), contrast pyramid (CP), gradient pyramid (GP)
filtering, etc. Since, the image features are sensitive
to the human visual system exists in different scales.
Therefore, these are not the highly suitable for
medical image fusion. Image registration, also
referred to as image fusion , superimposition ,
matching or merge , is the process of attaching two
images so that corresponding coordinate points in the
two images correspond to the same physical region of
the scene being imaged . It is a classical problem in
many applications where it is necessary to match two
or more images of the same scene. The images to be
registered might be acquired with different sensors
(e.g. sensitive to different parts or different tissues in
human body) or the same sensor at different times.
Image registration has application in many fields;
medical image registration is one of the most
successful fields that image registration is used.
Medical images are widely used in healthcare and
ISSN: 2231-5381
biomedical research. Their applications occur not
only within clinical diagnostic settings, but also
prominently so in the area of planning,
consummation, and evaluation of surgical and radio
therapeutical procedures. There is a wide range of
imaging modalities available, which include X-ray
Computed Tomography (CT), Single Photon
Emission Computed Tomography (SPECT), Positron
Emission Tomography (PET), Magnetic Resonance
Imaging, Nuclear Medicine, Ultrasonic Imaging,
Endoscopy and surgical microscopy, etc. These and
other imaging technologies provide rich information
on the physical properties and biological function of
the issues.
2METHODOLOGY
A
newimagefusion
framework
for
multimodal medical images which relies on the
NSCT domain. Two different fusion rules are
proposed for combining low-and high-frequency
coefficients. For fusing the low-frequency
coefficients, the phase congruency based model is
used. The main benefit of phase congruency is that it
selects and combines contrast- and brightnessinvariant representation contained in the lowfrequency coefficients. in the contrary, a new
definition of directive contrast in NSCT domain is
proposed and used to combine high-frequency
coefficients. Using directive contrast, the most
prominent texture and edge information are selected
from high-frequency coefficients and combined in
the fused ones. The definition of directive contrast is
consolidated by in corporating a visual constant to
the SML based definition of directive contrast which
provides a richer representation of the contrast.
LOW FREQUENCY ALGORITHM
The coefficients in the low-frequency subimages represent the approximation Component of
the source images. The simplest way is to use the
conventional averaging methods to produce the
composite bands. However, it cannot give the fused
low- frequency component of high quality for
medical image be-cause it leads to the reduced
contrast in the fused images. Therefore, a new
criterion is proposed here based on the phase
congruency. The complete process is described as
follows.
Phase congruency is a measure of feature
perception in the images which provides a
illumination and contrast invariant feature extraction
http://www.ijettjournal.org
Page 289
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
method This approach is based on the Local Energy
Model, which postulates that significant features can
be found at points in an image where the Fourier
components are maximally in phase. Furthermore,
the angle at which phase congruency occurs signifies
the feature type. The phase congruency approach to
feature perception has been used for feature
detection. First, logarithmic Gabor filter banks at
different discrete orientations are applied to the
image and the local amplitude and phase at a point
are obtained. The main properties, which acted as the
motivation to use phase congruency for multimodal
fusion, are as follows.
• The phase congruency is invariant to different pixel
intensity mappings. The images captured with
different modalities have significantly different pixel
mappings, even if the object is same. Therefore, a
feature that is free from pixel mapping must be
preferred.
the intensity contrast rather than the intensity value
itself. Generally, the same intensity value looks like a
different intensity value depending on intensity
values of neighboring pixels. Where is the local
luminance and is the luminance of the local
background. Generally, is regarded as local low
frequency and hence, is treated as local high
frequency. This definition is further extended as
directive contrast for multimodal image fusion These
contrast extensions take high-frequency as the pixel
value in multi resolution do-main. However,
considering single pixel is insufficient to determine
whether the pixels are from clear parts or not. Therefore, the directive contrast is integrated with the summodified-Laplacian to get more accurate salient
features.
• The phase congruency feature is invariant to
illumination and contrast changes. The capturing
environment of different modalities varies and
resulted in the change of illumination and contrast.
Therefore, multimodal fusion can be benefitted by an
illumination and contrast invariant feature.
• The edges and corners in the images are identified
by collecting frequency components of the image that
are in phase. As we know, phase congruency gives
the Fourier components that are maximally in phase.
Therefore, phase congruency provides the improved
localization of the image features, which lead to
efficient fusion. First the features are extracted from
low-frequency pca,pcb Sub-images using the phase
congruency extractor denoted by and respectively.
Fuse the low-frequency sub-images
HIGH FREQUENCY ALGORITHM
The coefficients in the high-frequency subimages usually include detailscomponent of the
source image. It is noteworthy that the noise is also
related to high-frequencies and may cause
miscalculation of sharpness value andthereforeeffect
the fusion performance.Therefore, a new criterion is
proposed here based on directive contrast. The whole
process is described as follows.First, the directive
contrast for NSCT high-fre-quency sub-images at
each scale and orientation using Equations, denoted
by Dca,Dcb Fuse the high-frequency sub-image.The
contrast feature measures the difference of the
intensity value at some pixel from the neighboring
pixels. The human visual system is highly sensitive to
ISSN: 2231-5381
Fig 1.0 multimodal medical image fusion framework.
2 a.EXISTING SYSTEM
Multi resolution (MR) transforms are a
widespread tool for image fusion. MR image fusion
as compared to other wavelet families. Additionally,
it revise the novel Multi size Windows (MW)
technique as a general approach for MR frameworks
that exploits advantages of different window sizes.
2 b.PROPOSED SYSTEM
Decomposition using NSCT
ANon-Sub sampled Contour let Transform
(NSCT)NSCT, based on the theory of CT, is a kind
of multi-scale and multi-direction computation
framework of the discrete images It can be divided
into two stages including non-sub sampled pyramid
(NSP) and non- sub sampled directional filter
bank(NSDFB). The former stage ensures the multi
scale property by using two-channel non-sub sampled
filter bank, and one low-frequency image and one
http://www.ijettjournal.org
Page 290
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
high-frequency image can be produced at each NSP
decomposition level. The subsequent NSP
decomposition stages are carried out to decompose
the low-frequency component available iteratively to
capture the singularities in the image .As a result,
NSP can resul tin sub-images ,which consists of one
low- and high-frequency images having the same size
as the source image where denotes the number of
decomposition levels. With levels. The NSDFB is
two-channel non-sub sampled filter banks which are
constructed by combining the directional fan filter
banks. NSDFB allows the direction decomposition
with stages in high-frequency images from NSP at
each scale and produces directional sub-images with
the same size as the source image. Therefore, the
NSDFB offers the NSCT with the multi-direction
property and provides us with more precise
directional details information.
Phase Congruency
Phase congruency is a measure of feature
perception in the images which provides a
illumination and contrast invariant fea-ture extraction
method This approach is based on the Local Energy
Model, which postulates that significant features can
be found at points in an image where the Fourier
components are maximally in phase. Furthermore,
the angle at which phase congruency occurs signifies
the feature type.
The phase congruency approach to feature
perception has been used for feature detection. First,
logarithmic Gabor filter banks at different discrete
orientations are applied to the image and the local
amplitude and phase at a point are obtained. The
main properties, which acted as the motivation to use
phase congruency for multimodal fusion, are as
follows.
• The phase congruency is invariant to different pixel
intensity mappings. The images captured with
different modalities have significantly different pixel
mappings, even if the object is same. Therefore, a
feature that is free from pixel mapping must be
preferred.
• The phase congruency feature is invariant to
illumination and contrast changes. The capturing
environment of different modalities varies and
resulted in the change of illumination and contrast.
Therefore, multimodal fusion can be benefitted by an
illumination and contrast invariant feature.
ISSN: 2231-5381
• The edges and corners in the images are identified
by collecting frequency components of the image that
are in phase. As we know, phase congruency gives
the Fourier components that are maximally in phase.
Therefore, phase congruency provides the improved
localization of the image features, which lead to
efficient fusion. The visual demonstrates that the
proposed algorithm can enhance the details of the
fused image, and can improve the visual effect with
much less information distortion than its competitors.
These statistical assessment findings agree with the
visual assessment.
Directive contrast
The contrast feature measures the difference
of the intensity value at some pixel from the
neighboring pixels. The human visual system is
highly sensitive to the intensity contrast rather than
the intensity value itself. Generally, the same
intensity. value looks like a different intensity value
depending on intensity values of neighboring pixels.
where is the local luminance and is the luminance of
the local background Generally, is regarded as local
low frequency and hence, is treated as local high
frequency. This definition is further extended as
directive contrast for multimodal image fusion .These
contrast extensions take high-frequency as the pixel
value in multi resolution do-main. However,
considering single pixel is insufficient to determine
whether the pixels are from clear parts or not. Therefore, the directive contrast is integrated with the summodified-Laplacian to get more accurate salient
features.
Inverse NSCT implementation
Perform n level inverse NSCT on the fused
low-frequency and high-frequency sub images, to
get the fused image .
3 RESULT AND DISCUSSION
In this Method, a image fusion framework is
proposed for multi-modal medical images, which is
based on non sub sampled contour let transform and
directive contrast. For fusion, two different rules are
used by which more information can be preserved in
the fused image with improved quality. The low frequency bands are fused by considering phase
congruency where as directive contrast is adopted as
the fusion measurement for high - frequency bands.
In our experiment, two groups of CT/MRI and two
groups of MR-T1/MR-T2 image s are fused using
http://www.ijettjournal.org
Page 291
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
convention al fusion algorithms and the proposed
frame work.
3 a.SCREENSHOT
Figure 3a.3 CT –Low freq image
The input CT image is divided into two
segments i.e. High level and low level Figure 3a.3
describes the low level segment.
Figure 3a.1 The input image for NSCT( human brain).
The input image for NSCT describes the
image of the human brain,which gets divided into
two segments i.e. High level and low level.
Figure 3a.4 CT -High freq image
The input CT image is divided into two
segments i.e. High level and low levelFigure3a.4
describes the high level segment.
Figure 3a.2 Input MRI images
The input MRI image describes the image of
human brain which gets divided into two segments
i.e. High level and low level.
Figure 3a.5 MRI-low freq image
ISSN: 2231-5381
http://www.ijettjournal.org
Page 292
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
The input MRI image is divided into two
segments i.e. High level and low level Figure 3a.5
describes the low level segment.
frequency bands are fused by considering phase
congruency where as directive contrast is adopted as
the fusion Measurement for high-frequency bands.
In our experiment, two groups of CT/MRI and two
groups of MR-T1/MR-T2 images are fused using
conventional fusion algorithms and the proposed
framework. The visual and statistical comparisons
demonstrate that the proposed algorithm can enhance
the details of the fused image, and can improve the
visual effect with much less information distortion
than its competitors. These statistical assessment
findings agree with the visual assessment. Further, in
order to show the practical applicability of the
proposed method, three clinical example are also
considered which includes analysis of diseased
person’s brain with alzheimer, subacute stroke and
recurrent tumor.
REFERENCES
Figure 3a.6 MRI-high freq image
The input CT image is divided into two
segments i.e. High level and low level Figure3.6
describes the high level segment.
[1]Jan Rexilius, “ Sum modified-Laplacian based Multi focus
Image Fusion Method in Cycle Spinning Sharp Frequency
Localized Contourlet Transform Domain”, master's Thesis,
Medical University of Luebeck, Germany, 2001.
[2]Christensen G. E. and Johnson H. J., “Consistent Image
Registration”, IEEE Transactions on medical imaging, vol. 20, No.
7, 2001.
[3]Woods R. P., Mazziotta J. C., and Cherry S. R., “MRI-PET
registration with automated algorithm,” J. Comput. Assist.
Tomogr., 17, pp. 536–546, 1993.
[4]Wang, M. Y., Fitzpatrick, J. M., and Maurer, C. R. “Design of
fiducials for accurate registration of CT and MR volume
images”.InLoew, M. (ed.), Medical imaging, Vol. 2434, pp. 96–
108.SPIE, 1995.
[5]Peters, T., Davey, B., Munger, P., Comeau, R., Evans, A., and
Olivier, A. “Threeguid-dimensional multimodal image-ance for
Figure 3a.7 Fusion images
Figure 3a.7 describes the fusion of all the
input images. The low level segment of both input
images is fused together and the same is done with
high level segment. Then finally both the fused
images is inverted and fusion is done.
CONCLUSION
A novel image fusion framework is
proposed for multi-modal medical images, which is
based on non-sub sampled contourlet transform and
directive contrast. For fusion, two different rules are
used by which more information can be preserved in
the fused image with improved quality. The low
ISSN: 2231-5381
neurosurgery.”IEEE Transactions on medical imaging, 15(2), 121–
128, 1996.
[6]K.S. Arun, T.S. Huang, S.D. Bostein, “Least squares fitting of
two 3D point sets.” IEEE
[7]M.J. Fitzpatrick, J. West and C. Maurer Jr., “Predicting error in
rigid-body, point-based registration.” IEEE Transactions in
Medical Imaging, vol. 17, pp. 694-702, 1998.
[8]C.A. Pelizzari, G.T.Y Chen, D.R Spelbring, R.R.
Weichselbraum, C. Chen, “Accurate three dimensional registration
of CT, PET and/or MR images of the brain.” Journal of Computer
Assisted Tomography vol. 13, pp. 20-26, 1989.
[9]P.J. Besl and N.D. McKay, “A method for registration of 3D
shapes.” IEEE Transactions in Pattern Analysis and Machine
http://www.ijettjournal.org
Page 293
International Journal of Engineering Trends and Technology (IJETT) – Volume 9 Number 6 - Mar 2014
Vision.Vol.14, pp. 239-256, 1992.aph.Image Processing, vol. 27,
pp. 321–345, 1984.
[10]C. T. Huang and O. R. Mitchell, “A Euclidean distance
transform using grayscale morphology decomposition,” IEEE
Trans. Pattern Anal. Mach. Intell., vol. 16, pp. 443–448, 1994.
[11]K. J. Friston, J. Ashburner, J. B. Poline, C. D. Frith, J. D.
Heather, and F. R.S.J., “Spatial registration and normalisation of
images,” Human Brain Mapping, vol. 2, pp. 165–189, 1995.
[12]Lemieux L and Barker G J, “Measurement of small inter-scan
fluctuations in voxel dimensions in magnetic resonance images
using registration”, Med. Phys. 25 1049–54, 1998
[13] R. P. Woods, S. T. Grafton, C. J. Holmes, S. R. Cherry, and J.
C. Mazziotta, “Automated image registration: I. General methods
and intrasubject, intramodality validation,” J. Com-put. Assist.
Tomogr., vol. 22, pp. 139–152, 1998.
[14]Van Den Elsen P A, Pol E-J D, Sumanaweera T S, Hemler P
F, Napel S and Adler J R, “Grey value correlation techniques used
for automatic matching of CT and MR brain and spine images”,
Proc. SPIE 2359227–37, 1994
[15]J.P.W. Pluim, J.B.A. Maintz, M.A. Viergever, "Mutual
information based registration of medical images: a survey", IEEE
Transactions on Medical Imaging, 2003.
[16]Buzug T. M. and Weese J., “Image registration for DSA
quality enhancement”, Computerized Imaging Graphics 22103
1998.
[17]W. M. Wells, III, P. Viola, H. Atsumi, S. Nakajima, and R.
Kikinis, “Multi-modal volumeregistration by maximization of
mutual information,” Med. Image Anal., vol. 1, pp. 35–51, 1996.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 294