Final Year IEEE Projects for BE, B.Tech, ME, M.Tech,M.Sc, MCA & Diploma Students latest Java, .Net, Matlab, NS2, Android, Embedded,Mechanical, Robtics, VLSI, Power Electronics, IEEE projects are given absolutely complete working product and document providing with real time Software & Embedded training......
JAVA, .NET, NS2, MATLAB PROJECTS:
Networking, Network Security, Data Mining, Cloud Computing, Grid Computing, Web Services, Mobile Computing, Software Engineering, Image Processing, E-Commerce, Games App, Multimedia, etc.,
EMBEDDED SYSTEMS:
Embedded Systems,Micro Controllers, DSC & DSP, VLSI Design, Biometrics, RFID, Finger Print, Smart Cards, IRIS, Bar Code, Bluetooth, Zigbee, GPS, Voice Control, Remote System, Power Electronics, etc.,
ROBOTICS PROJECTS:
Mobile Robots, Service Robots, Industrial Robots, Defence Robots, Spy Robot, Artificial Robots, Automated Machine Control, Stair Climbing, Cleaning, Painting, Industry Security Robots, etc.,
MOBILE APPLICATION (ANDROID & J2ME):
Android Application, Web Services, Wireless Application, Bluetooth Application, WiFi Application, Mobile Security, Multimedia Projects, Multi Media, E-Commerce, Games Application, etc.,
MECHANICAL PROJECTS:
Auto Mobiles, Hydraulics, Robotics, Air Assisted Exhaust Breaking System, Automatic Trolley for Material Handling System in Industry, Hydraulics And Pneumatics, CAD/CAM/CAE Projects, Special Purpose Hydraulics And Pneumatics, CATIA, ANSYS, 3D Model Animations, etc.,
CONTACT US:
ECWAY TECHNOLOGIES
23/A, 2nd Floor, SKS Complex,
OPP. Bus Stand, Karur-639 001.
TamilNadu , India.Cell: +91 9894917187.
Website: www.ecwayprojects.com | www.ecwaytechnologies.com
Mail to: ecwaytechnologies@gmail.com.
A technique for simultaneous visualization and segmentation of hyperspectral data
1. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
A TECHNIQUE FOR SIMULTANEOUS VISUALIZATION AND
SEGMENTATION OF HYPERSPECTRAL DATA
By
A
PROJECT REPORT
Submitted to the Department of electronics &communication Engineering in the
FACULTY OF ENGINEERING & TECHNOLOGY
In partial fulfillment of the requirements for the award of the degree
Of
MASTER OF TECHNOLOGY
IN
ELECTRONICS &COMMUNICATION ENGINEERING
APRIL 2016
2. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
CERTIFICATE
Certified that this project report titled “A TECHNIQUE FOR SIMULTANEOUS
VISUALIZATION AND SEGMENTATION OF HYPERSPECTRAL DATA ” is the
bonafide work of Mr. _____________Who carried out the research under my supervision Certified
further, that to the best of my knowledge the work reported herein does not form part of any other
project report or dissertation on the basis of which a degree or award was conferred on an earlier
occasion on this or any other candidate.
Signature of the Guide Signature of the H.O.D
Name Name
3. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
DECLARATION
I hereby declare that the project work entitled “A TECHNIQUE FOR SIMULTANEOUS
VISUALIZATION AND SEGMENTATION OF HYPERSPECTRAL DATA ” Submitted to
BHARATHIDASAN UNIVERSITY in partial fulfillment of the requirement for the award of the
Degree of MASTER OF APPLIED ELECTRONICS is a record of original work done by me the
guidance of Prof.A.Vinayagam M.Sc., M.Phil., M.E., to the best of my knowledge, the work
reported here is not a part of any other thesis or work on the basis of which a degree or award was
conferred on an earlier occasion to me or any other candidate.
(Student Name)
(Reg.No)
Place:
Date:
4. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
ACKNOWLEDGEMENT
I am extremely glad to present my project “A TECHNIQUE FOR SIMULTANEOUS
VISUALIZATION AND SEGMENTATION OF HYPERSPECTRAL DATA ” which is a
part of my curriculum of third semester Master of Science in Computer science. I take this
opportunity to express my sincere gratitude to those who helped me in bringing out this project
work.
I would like to express my Director,Dr. K. ANANDAN, M.A.(Eco.), M.Ed., M.Phil.,(Edn.),
PGDCA., CGT., M.A.(Psy.)of who had given me an opportunity to undertake this project.
I am highly indebted to Co-OrdinatorProf. Muniappan Department of Physics and thank from
my deep heart for her valuable comments I received through my project.
I wish to express my deep sense of gratitude to my guide
Prof. A.Vinayagam M.Sc., M.Phil., M.E., for her immense help and encouragement for
successful completion of this project.
I also express my sincere thanks to the all the staff members of Computer science for their kind
advice.
And last, but not the least, I express my deep gratitude to my parents and friends for their
encouragement and support throughout the project.
5. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
ABSTRACT:
In this paper, we propose an optimization-based method for simultaneous fusion and
unsupervised segmentation of hyperspectral remote sensing images by exploiting redundancy in
the data. The hyperspectral data set is visualized as a single image obtained by weighted addition
of all spectral points at each pixel location in the data set. The weights are optimized to improve
those statistical characteristics of the fused image, which invoke an enhanced response from a
human observer. A piecewise-constant smoothness constraint is imposed on the weights instead
of the fused image by minimization of its 3-D total-variation norm, thus preventing the fused image
from blurring. The optimal recovery of the weight matrix additionally provides useful information
in segmenting the hyperspectral data set spatially. We provide ample experimental results to
substantiate the usefulness of the proposed method.
6. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
INTRODUCTION:
Hyperspectral imaging captures the reflectance map of a scene at various wavelengths of
light, typically in the range of 0.4–2.5 μm with a resolution of 10 nm. Such a high spectral
resolution leads to a very large number of image bands that suffer from a lot of redundancy of data.
The problem of efficient mining of information from such data sets has drawn considerable
attention over the past few decades. Image fusion and segmentation are two important operations
performed on remote sensing hyperspectral data sets. Image fusion offers the first step in
visualizing a scene in a meaningful way for a human observer, whereas segmentation offers an
object level description of the scene. In previous studies, many different techniques have been
developed to perform these operations. Although the end results of the solutions to these two
problems, i.e., visualization and segmentation, are different, they have a certain common structure
associated with them. In both the problems, some measure of spatial and/or spectral distance
between the pixels can be used to operate on the data set to obtain the output. By exploiting this
similarity of structure, we suggest here an alternative approach that can perform these tasks
simultaneously and in an interdependent manner.
Most image fusion algorithms aim at extracting the salient features from the hyperspectral
data set and combining them into a single image for observer interpretation. Traditional methods
include the use of techniques such as principle and independent component analysis (ICA) to
reduce the dimensionality of a data set to produce a small number of bands that capture statistically
significant information from the data. Multiresolution methods have also been effectively used to
capture the significant spatial changes in the contiguous bands and render the scene radiance in a
single image while preserving the continuity of edges in the fused image, Some techniques
involving information theoretic measures such as mutual information and entropy try to identify
significant bands in the data set and discard the rest, before performing image fusion , Although
accurate and well established, these techniques are derived from general image fusion techniques
and do not scale very well functionally and computationally for the increasingly large number of
bands encountered in hyperspectral data sets.
Recently proposed optimization-based image fusion techniques have expressed the pixels
in the fused image as a weighted sum of corresponding pixels in the different spectral bands, and
the weights are calculated to improve certain statistical characteristics in the fused image, which
are desirable for better visual interpretation , Due to the smoothly varying nature of hyperspectral
7. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
data, at each pixel, it is expected that the contribution by, and hence, the value of weights assigned
to, neighboring bands will be nearly equal except at those consecutive bands, where there is a
significant change in the reflectance value. Hence, the weights value in the 3-D space (x, y, z) is
expected to be a piecewise-constant function. However, none of the existing methods enforce such
a constraint, In our technique, along with the optimization of desired statistical characteristics of
the fused image, we also optimize the weights to obtain such a piecewise-constant nature.
Furthermore, we demonstrate that due to the imposed constraint on the weight function, the
estimated weight imbibes the salient information from the reflectance data required to
meaningfully segment the hyperspectral image. Furthermore, unlike most of the existing methods
(to be discussed in the next section), we optimize over the choice of weight array w(x, y, λ) as
opposed to the fused image directly.
Another contribution of this paper is to identify the right set of desired characteristics of
the fused image and provide a simple formulation to achieve the same. Higher Shannonentropy is
one such desired quality of the fused image but it is difficult to set it up as an optimization problem.
Thus, the entropy maximization problem has been framed as a kurtosis minimization problem. In
addition, previously suggested objective functions have not considered the preservation of edge
features while optimizing desired statistical features of the fused image, which leads to a loss of
salient information. Imposing l2-norm minimization-based smoothness constraint on the image for
solving this problem leads to overblurring of the image. This problem has been solved in our
approach by optimizing the fused image to achieve a “target-gradient,” which preserves the salient
spatial features. This target gradient is calculated from the data set by calculating direction and
magnitude of the maximum gradient at each pixel out of all the constituent bands using a structural
tensor, as proposed by Socolinksy et al. and Piella .
Although hyperspectral data is not sparse in itself, the gradient of the data contains many
near-zero values. This is due to the gradually varying nature of the reflectance spectrum of any
material. Minimization of total-variation (TV) norm of a vector leads to a representation whose
gradient is sparse. Our proposed approach exploits this model of hyperspectral data to estimate a
weight array having a sparse gradient by minimizing its TV norm. However, we need to minimize
the 3-D TV norm as opposed to the popular 2-D TV norm used in solving the image restoration
problem , The difficulty in the minimization of the TV norm due to its non-differentiability is
overcome using the majorization-minimization algorithmic approach suggested by Oliveira et al.
8. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
in which approximates the TV-norm function by an upper bound quadratic function and then
minimizes it. Imposing the piecewise-constant smoothness constraint on the weights helps in
preserving and enhancing the edges. Optimization of the overall cost function is posed as an
unconstrained optimization problem and solved by a simple gradient descent algorithm.
Having estimated the weight array, we obtain the fused image using a weighted sum of all
the spectral bands and the segmentation map by using the k-means clustering algorithm on the
weight array. A point to note is that this formulation is very different from end-member unmixing
and does not seek to express the pixel spectra as a weighted sum of standard spectra of elemental
materials. It requires no training or learning. The organization of this paper is as follows. In Section
II, we look at some recently suggested algorithms for the problem of visualization of hyperspectral
images. Section III presents in detail the formulation of the proposed algorithm. In Section IV, we
present test results of our algorithm on standard hyperspectral data sets and contrast our results
with the currently existing algorithms using standard quality metrics for both fusion and
segmentation process. Finally, in Section V, we provide the conclusions drawn from the results
obtained in the previous section.
9. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
CONCLUSION:
We have presented a new approach for fusion of hyperspectral images that also facilitates
a better segmentation. The image fusion has been performed by obtaining an optimal weight at
each spatial location and at each spectral band. The contribution of this paper lies in formulating a
refined set of qualitative goals that ensure an enhanced fused image ready to be visualized on a
display device. These goals have been achieved by associating them with quantitative quality
measures of the fused image and optimizing the weights to improve them. The problem was posed
as an unconstrained optimization problem and solved using a simple gradient descent algorithm.
The significant improvements that were obtained in our method over previous methods of similar
construction include the optimization of the entropy of the fused image, which was solved by
posing the problem as a kurtosis minimization problem,
Extracting the spatial features in the bands using the closeness of the gradient of the fused
image to the “target-gradient” cost term, and, most importantly, the edge-enhancement and noise
removal, which was achieved by imposing a piecewise-constant smoothness constraint on the
weight array by minimization of its 3-D TV norm. The resulting weight array is also seen to be
better suited for segmentation than the hyperspectral data set itself because of its piecewise
constant nature. The effectiveness of the suggested algorithm was demonstrated with multiple data
sets and its competitiveness with the state-of-the-art methods was proven using various quality
metrics suggested in the literature.
10. OUR OFFICES @CHENNAI/ TRICHY / KARUR / ERODE / MADURAI / SALEM / COIMBATORE / BANGALORE / HYDRABAD
CELL: +91 9894917187 | 875487 1111 / 2111 / 3111 / 4111 / 5111 / 6111
ECWAY TECHNOLOGIES
IEEE SOFTWARE | EMBEDDED | MECHANICAL | ROBOTICS PROJECTS DEVELOPMENT
Visit: www.ecwaytechnologies.com | www.ecwayprojects.com Mail to: ecwaytechnologies@gmail.com
REFERENCES:
[1] K. Kotwal and S. Chaudhuri, “A fast approach for fusion of hyperspectral images through
redundancy elimination,” in Proc. 7th ACM ICVGIP, 2010, pp. 506–511.
[2] K. Kotwal and S. Chaudhuri, “An optimization based approach to fusion of hyperspectral
images,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 5, no. 2, pp. 501–509, Apr.
2012.
[3] Y. Zhu, P. Varshney, and H. Chen, “Evaluation of ICA based fusion of hyperspectral images
for color display,” in Proc. 10th Int. Conf. Inf. Fusion, 2007, pp. 1–7.
[4] T. Stathaki, Image Fusion: Algorithms and Applications. Amsterdam, The Netherlands:
Elsevier, 2011.
[5 G. Piella, “Image fusion for enhanced visualization: A variational approach,” Int. J. Comput.
Vis., vol. 83, no. 1, pp. 1–11, Jun. 2009.
[6] J. P. Oliveira, J. M. Bioucas-Dias, and M. A. Figueiredo, “Adaptive total variation image
deblurring: A majorization minimization approach,” Signal Process., vol. 89, no. 9, pp. 1683–
1693, Sep. 2009.
[7] K. Kotwal and S. Chaudhuri, “A novel approach to quantitative evaluation of hyperspectral
image fusion techniques,” Inf. Fusion, vol. 14, no. 1, pp. 5–18, Jan. 2013.
[8] M. Cui, A. Razdan, J. Hu, and P. Wonka, “Interactive hyperspectral image visualization using
convex optimization,” IEEE Trans. Geosci. Remote Sens., vol. 47, no. 6, pp. 1673–1684, Jun.
2009.
[9] K. Kotwal and S. Chaudhuri, “Visualization of hyperspectral images using bilateral filtering,”
IEEE Trans. Geosci. Remote Sens., vol. 48, no. 5, pp. 2308–2316, May 2010.
[10] S. Li, H. Yin, and L. Fang, “Remote sensing image fusion via sparse representations over
learned dictionaries,” IEEE Trans. Geosci. Remote Sens., vol. 51, no. 9, pp. 4779–4789, Sep. 2013.