4. Image fusion is the process of
combining relevant information from
two or more images into a single
image.
The resulting image will be more
informative than any one of the input
image.
5. Remote sensing.
Medical diagnosis.
Battle field surveillance.
Guidance & control of autonomous
vehicle.
Automated target recognition.
7. In remote sensing fields, it contains
color information which is produced by
three sensors covering the red, green
and blue spectral wavelength.
It has poor spatial resolution.
8. In contrast to MS image, it has high
spatial resolution but without color
information.
Image fusion can combine the
geometric detail of PAN image & the
color information of the MS image to
produce a high resolution MS image.
9. Fused image may be created from multiple images
from the same imaging modality or by combining
information from multiple modalities such as:
• MRI
• CT
• PET
• SPECT
14. CT provides good spatial resolution (the
ability to distinguish two structures an
arbitrarily small distance from each other as
separate).
MRI provides comparable resolution with
far better contrast resolution (the ability to
distinguish the differences between two
arbitrarily similar but not identical tissue).
15.
16. For accurate diagnosis, radiologists must
integrate information from multiple image
formats.
Fused images are beneficial in diagnosing
and treating cancer.
17. Method of image fusion based on Discrete
Wavelet Transform & Self Organizing Feature
Mapping (SOFM) neutral network.
The proposed method is a feature level fusion
method.
Two-Dimensional DWT is used to decompose
into various detail at different levels to extract
useful feature and SOFM neural network is
used to recognize complementary features.
18. This transform is most appropriate for non-
stationary signals.
Wavelet analysis can be used to divide
information of an image into approximation
and detail sub-signal.
The approximation sub-signal shows
general trend of pixel values and three
detail sub-signal on the horizontal, vertical
and diagonal details.
20. SOFM neural network is used to recognize & extract
features.
Features can be intensity of the pixels or edge &
texture of an object.
This is done by training & simulating the network
for corresponding details of each level of all images.
At the end of training & simulation the coefficient of
detail set depicting features are clustered.
21.
22. The application is very helpful in various fields like
military, security, surveillance & medical areas,
especially in change detection of organs & tumors, &
in remote sensing for monitoring land or forest
exploitation.
One of the major drawbacks of DWT is that the
transformation does not provide shift invariance.
23.
24. Q.P. Zhang, W.J. Tang, L.L. Lai, W.C. Sun, K.P. Wong , ” Medical diagnostic image data fusion
based on wavelet transformation & self organizing features mapping neutral networks” ,
Proceedings of the Third International Conference on Machine Learning & Cybemetics ,
Shanghai, 26-29 August 2004, IEEE.
Q.P. Zhang, M. liang, w.c. sun ,” medical diagnostic image fusion based on feature mapping
wavelet neural network” proceeding of third international conference on image and
graphics, IEEE
Debasish patnaik , “ biomedical image fusion using wavelet transform and SOFM neural
network, IEEE