Fuzzy Type Image Fusion Using SPIHT Image Compression Technique
1 like201 views
This paper presents a fuzzy image fusion technique using the Set Partitioning in Hierarchical Trees (SPIHT) image compression method, which demonstrates improved fusion quality compared to previous techniques. The proposed algorithm is simple, easy to implement, and suitable for real-time applications, particularly in multi-modal image fusion. Performance evaluation metrics such as Peak Signal-to-Noise Ratio (PSNR) and Root Mean Square Error (RMSE) are utilized to validate the effectiveness of the proposed method.
![Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58
www.ijera.com 55 | P a g e
Fuzzy Type Image Fusion Using SPIHT Image Compression
Technique
Shyam Sundar, Mahendra Kumar, Gaurav Sharma
M.Tech Scholar with the Department of Electronics & Communication Engineering, Mewar University
Chittorgarh, India
Faculty with Department of Electronics & Communication Engineering, University College of Engineering,
RTU, Kota, India.
HOD with the Department of Electronics & Communication Engineering Mewar University Chittorgarh, India
Abstract
This paper presents a fuzzy type image fusion technique using Set Partitioning in Hierarchical Trees (SPIHT).
It is concluded that fusion with higher single levels provides better fusion quality. This technique can be used
for fusion of fuzzy images as well as multi model image fusion. The proposed algorithm is very simple, easy to
implement and could be used for real time applications. This is paper also provided comparatively studied
between proposed and previous existing technique and validation of the proposed algorithm as Peak Signal to
Noise Ratio (PSNR), Root Mean Square Error (RMSE).
Index Terms- Fuzzy type image, Set Partitioning in Hierarchical Trees (SPIHT), PSNR, RMSE.
I. INTRODUCTION
OFF late, multi sensor data fusion is found to
play a vital role in defence as well as in civilian
applications because diversity of sensors available
and these working in different spectral bands. Image
fusion, where multiple registered images are
combined together to increase the information
content, is a promising research area. Numerous
image fusion algorithms such as multi-resolution
[1, 2], multi scale [3] and statistical signal
processing [4,5,6] based techniques are presented and
evaluated.
The developments in the field of sensing
technologies multi-sensor systems have become a
reality in a various fields such as remote sensing,
medical imaging, machine vision and the military
applications for which they were developed.
The result of the use of these techniques is
a increase of the amount of data available. Image
fusion provides an effective way of reducing the
increasing volume of information while at the same
time extracting all the useful information from the
source images.
Multi-sensor data often presents complementary
information, so image fusion provides an effective
method to enable comparison and analysis of data.
The aim of image fusion, apart from reducing the
amount of data, is to create new images that are
more suitable for the purposes of human/machine
perception, and for further image- processing tasks
such as segmentation, object detection or target
recognition in applications such as remote sensing
and medical imaging. For example, visible-band and
infrared images may be fused to aid pilots landing
aircraft in poor visibility [3, 5,7].
Finally, the performance of the image fusion
scheme is evaluated as tradeoffs between true image
and fused image. In previous techniques when apply
fuzzy type images, the performance criterion is poor,
so this paper proposed a novel Set Partitioning in
Hierarchical Trees image compression techniques
which provide fused image with better quality. The
remainder of the paper is organized as follows: In
Section II, discussed Proposed Set Partitioning in
Hierarchical Trees image compression techniques.
Different Fusion Performance evaluation criterion
presented in section III. Results and comparatively
study of techniques is described in section IV and
conclusions are presented in Sections V.
II. Proposed Set Partitioning in
Hierarchical Trees (SPIHT) image
compression technique
SPIHT is computationally very fast and among
the best image compression algorithms known today.
According to statistic analysis of the output binary
stream of SPIHT encoding, propose a simple and
effective method combined with Huffman encode for
further compression. In this paper the results from the
SPHIT algorithm are compared with the existing
methods for compression like discrete cosine
transform (DCT) and discrete wavelet transform
(DWT).
The set partitioning in hierarchical tree algorithm
is proposed [11] and utilized for lossless image
compression nowadays. One of the most powerful
wavelet based image compression techniques is
RESEARCH ARTICLE OPEN ACCESS](https://image.slidesharecdn.com/h59035558-151001064221-lva1-app6891/85/Fuzzy-Type-Image-Fusion-Using-SPIHT-Image-Compression-Technique-1-320.jpg)
![Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58
www.ijera.com 56 | P a g e
SPIHT. The main advantages of SPIHT method are it
can provide Good Image quality with high PSNR and
low RMSE. First, the image is decomposed into four
sub-bands. The decomposition process is repeated
until reach the final scale. Each decomposition
consists of one low-frequency sub-band with three
high-frequency sub-bands. The extension and
efficient implementation of EZW-[Embedded Zero
Wavelet] algorithm [12, 13] is SPIHT algorithm. it is
represented by the equation as follows
Where ( )nS x ,is the importance of set of coordinate
x.
,i j
L is the coefficient value at each coordinate(i,j).
LL1 HL1
HL
LH1 HH1
LH HH
Fig.1: 2 level Discrete Wavelet Transform
The complete SPIHT algorithm does
compression in three steps such as sorting,
refinement and quantization. The SPIHT algorithm
encodes the image data using three lists such as LIP,
LIS and LSP. LIP contains the individual coefficients
having the magnitudes smaller than the threshold
values. LIS contains the overall wavelet coefficients
defined in tree structure having magnitudes smaller
than the threshold values. LSP is the set of pixels
having magnitude greater than the threshold value of
the important pixels.
SPIHT is computationally very fast and among
the best image compression algorithms known today.
The statistical analysis of the output binary
stream of encoding, propose a simple and effective
method combined with Huffman encode for further
compression transform as a branch of mathematics
developed rapidly, which has a good localization
property in the time domain and frequency domain,
can analyse the details of any scale and frequency.
so, it superior to Fourier and DCT. It has been widely
applied and developed in image processing and
compression.
III. PROPOSED IMAGE FUSION
SYSTEM
Fig.2: Proposed Image Fusion System
Consider image 1 as fuzzy type image 1 and
image 2 as fuzzy type image 2 and image
compression and feature extracted by SPIHT
technique and apply Image fusion algorithm as
averaging method for fused image and compare
actual/true image with fused image and calculate
PSNR and RMSE to check effectiveness of proposed
system.
Pseudo code:
Im1=DWT (image1)
En1=encoding (Im1)
De1=decoding (En1)
Im1’=IDWT (De1)
Id1=Im1-Im1’
Im2=DWT (image2)
En2=encoding (Im2)
De2=decoding (En2)
Im2’=IDWT (De2)
Id2=Im2-Im2’
Id=abs(Id1)-(Idf)>=0;
,( , ) 2max1,
0( )
n
i ji j x L
n othrewiseS x
Image1 Image2
SPIHT SPIHT
Decoding
IDWT
FUSION
ALGORITHM
Fused Image](https://image.slidesharecdn.com/h59035558-151001064221-lva1-app6891/85/Fuzzy-Type-Image-Fusion-Using-SPIHT-Image-Compression-Technique-2-320.jpg)
![Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58
www.ijera.com 57 | P a g e
Imf=Idf+Imf
Idfm=decoding(Imf)
Imf=IDWT (Idfm)
This pseudo code shows the implementation Process.
IV. FUSION PERFORMANCE
EVALUATION
The performance of image fusion algorithms
can be evaluated when the reference image is
available using the following metrics [3, 5, 8, 9]:
Root Mean Square Error:
It is computed as the root mean square error
(RMSE) of the corresponding pixels in the reference
image Ir and the fused image If. It will be nearly
zero when the reference and fused images are alike
and it will increase when the dissimilarity increases.
Peak Signal to Noise Ratio:
This value will be high when the fused and
reference images are alike and higher value implies
better fusion.
where, L in the number of gray levels in the image.
V. RESULTS AND COMPARATIVELY
STUDY
Fig 3: Reference fuzzy type Image
Fig 4: Fuzzy type Image 1
Fig 5: Fuzzy type Image 2
Reference Fuzzy type image mahi.jpg of size
420x342 is shown in Fig-3. The two images to be
fused are generated from the Fuzzy type truth image
using fuzzy type as shown in Fig-4&5. The fused
image is almost similar to reference image and the
error image is almost zero. It shows that the fused
image contains all information coming from the
complementary source images [10].
The fusion quality evaluation metrics are shown
in Table-1. The metrics showed in table with bold
font are better among others. Fusion with 3 level
pyramid or above are giving almost similar
performance.
Table 1 Proposed Hybrid Technique
Pyramid levels
Techniq
ues
1 3 5 7
RMSE
DCTPT 10.0311 9.3924 9.4921 12.7680
FFTPT 7.9812 7.8937 7.9885 8.0075
SPIHT 6.4560
PSNR
DCTPT 38.1513 38.4370 38.3912 37.1036
FFTPT 39.1441 39.1920 39.1402 39.1298
SPIHT 40.0651
Fig 6: Comparatively study for RMSE between DCT
based Pyramid Transform and Proposed Method](https://image.slidesharecdn.com/h59035558-151001064221-lva1-app6891/85/Fuzzy-Type-Image-Fusion-Using-SPIHT-Image-Compression-Technique-3-320.jpg)
![Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58
www.ijera.com 58 | P a g e
Fig 7: Comparatively study for PSNR between DCT
based Pyramid Transform and Proposed Method
VI. CONCLUSION
A novel image fusion technique using SPIHT
based has been presented and its performance
evaluated. It is concluded that fusion with three
level provides better fusion quality. This technique
can be used for fusion of fuzzy type images as well
as multi model image fusion. The proposed algorithm
is very simple, easy to implement and could be used
for real time applications. This paper is also
provided comparatively studied between
proposed and DCT based Pyramid transform
technique and validation of the proposed algorithm as
Peak Signal to Noise Ratio (PSNR) and Root Mean
Square Error (RMSE) in table 1 and plots also.
REFERENCE
[1] A. Toet, “A morphological pyramid image
decomposition”, Pattern Recogn. Lett. 9(4),
255–261 (1989).
[2] VPS Naidu and J.R. Raol, ”Pixel-Level
Image Fusion using Wavelets and Principal
Component Analysis – A Comparative
Analysis” Defence Science Journal, Vol.58,
No.3, pp.338-352, May 2008.
[3] VPS Naidu, “Discrete Cosine Transform-
based Image Fusion”, Special Issue on
Mobile Intelligent Autonomous System,
Defence Science Journal, Vol. 60, No.1,
pp.48-54, Jan. 2010.
[4] Mahendra Kumar et.al., “Digital Image
Watermarking using Fractional Fourier
transform via image compression”, In IEEE
International Conference on Computational
Intelligence and Computing Research 2013
(IEEE ICCIC-2013), 26-28 Dec., 2013.
[5] VPS Naidu, “A Novel Image Fusion
Technique using DCT based Laplacian
Pyramid”, International Journal of Inventive
Engineering and Sciences (IJIES) ISSN:
2319–9598, Volume-1, Issue-2, January,
2013.
[6] Rick S. Blum, “Robust image fusion using a
statistical signal processing approach”,
Image Fusion, 6, pp.119-128, 2005.
[7] Shutao Li, James T. Kwok and Yaonan
Wang, “Combination of images with diverse
focuses using the spatial frequency”,
Information fusion, 2(3), pp.167-176, 2001.
[8] V.P.S. Naidu, J.R. Rao. “Pixel-level Image
Fusion using Wavelets and Principal
Component Analysis”, Defence Science
Journal, pp. 338 -352, 2008.
[9] Seetha M, MuraliKrishna I.V &
Deekshatulu, B.L, (2005) “Data Fusion
Performance Analysis Based on
Conventional and Wavelet Transform
Techniques”, IEEE Proceedings on
Geoscience and Remote Sensing
Symposium, Vol 4, pp. 2842-2845.
[10] Yang, X.H., Huang, F.Z., Liu,G. 2009
“Urban Remote Image Fusion Using Fuzzy
Rule”. IEEE Proceedings of the Eighth
International Conference on Machine
Learning and Cybernetics, pp. 101-109,
(2009).](https://image.slidesharecdn.com/h59035558-151001064221-lva1-app6891/85/Fuzzy-Type-Image-Fusion-Using-SPIHT-Image-Compression-Technique-4-320.jpg)
Fuzzy Type Image Fusion Using SPIHT Image Compression Technique
- 1. Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58 www.ijera.com 55 | P a g e Fuzzy Type Image Fusion Using SPIHT Image Compression Technique Shyam Sundar, Mahendra Kumar, Gaurav Sharma M.Tech Scholar with the Department of Electronics & Communication Engineering, Mewar University Chittorgarh, India Faculty with Department of Electronics & Communication Engineering, University College of Engineering, RTU, Kota, India. HOD with the Department of Electronics & Communication Engineering Mewar University Chittorgarh, India Abstract This paper presents a fuzzy type image fusion technique using Set Partitioning in Hierarchical Trees (SPIHT). It is concluded that fusion with higher single levels provides better fusion quality. This technique can be used for fusion of fuzzy images as well as multi model image fusion. The proposed algorithm is very simple, easy to implement and could be used for real time applications. This is paper also provided comparatively studied between proposed and previous existing technique and validation of the proposed algorithm as Peak Signal to Noise Ratio (PSNR), Root Mean Square Error (RMSE). Index Terms- Fuzzy type image, Set Partitioning in Hierarchical Trees (SPIHT), PSNR, RMSE. I. INTRODUCTION OFF late, multi sensor data fusion is found to play a vital role in defence as well as in civilian applications because diversity of sensors available and these working in different spectral bands. Image fusion, where multiple registered images are combined together to increase the information content, is a promising research area. Numerous image fusion algorithms such as multi-resolution [1, 2], multi scale [3] and statistical signal processing [4,5,6] based techniques are presented and evaluated. The developments in the field of sensing technologies multi-sensor systems have become a reality in a various fields such as remote sensing, medical imaging, machine vision and the military applications for which they were developed. The result of the use of these techniques is a increase of the amount of data available. Image fusion provides an effective way of reducing the increasing volume of information while at the same time extracting all the useful information from the source images. Multi-sensor data often presents complementary information, so image fusion provides an effective method to enable comparison and analysis of data. The aim of image fusion, apart from reducing the amount of data, is to create new images that are more suitable for the purposes of human/machine perception, and for further image- processing tasks such as segmentation, object detection or target recognition in applications such as remote sensing and medical imaging. For example, visible-band and infrared images may be fused to aid pilots landing aircraft in poor visibility [3, 5,7]. Finally, the performance of the image fusion scheme is evaluated as tradeoffs between true image and fused image. In previous techniques when apply fuzzy type images, the performance criterion is poor, so this paper proposed a novel Set Partitioning in Hierarchical Trees image compression techniques which provide fused image with better quality. The remainder of the paper is organized as follows: In Section II, discussed Proposed Set Partitioning in Hierarchical Trees image compression techniques. Different Fusion Performance evaluation criterion presented in section III. Results and comparatively study of techniques is described in section IV and conclusions are presented in Sections V. II. Proposed Set Partitioning in Hierarchical Trees (SPIHT) image compression technique SPIHT is computationally very fast and among the best image compression algorithms known today. According to statistic analysis of the output binary stream of SPIHT encoding, propose a simple and effective method combined with Huffman encode for further compression. In this paper the results from the SPHIT algorithm are compared with the existing methods for compression like discrete cosine transform (DCT) and discrete wavelet transform (DWT). The set partitioning in hierarchical tree algorithm is proposed [11] and utilized for lossless image compression nowadays. One of the most powerful wavelet based image compression techniques is RESEARCH ARTICLE OPEN ACCESS
- 2. Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58 www.ijera.com 56 | P a g e SPIHT. The main advantages of SPIHT method are it can provide Good Image quality with high PSNR and low RMSE. First, the image is decomposed into four sub-bands. The decomposition process is repeated until reach the final scale. Each decomposition consists of one low-frequency sub-band with three high-frequency sub-bands. The extension and efficient implementation of EZW-[Embedded Zero Wavelet] algorithm [12, 13] is SPIHT algorithm. it is represented by the equation as follows Where ( )nS x ,is the importance of set of coordinate x. ,i j L is the coefficient value at each coordinate(i,j). LL1 HL1 HL LH1 HH1 LH HH Fig.1: 2 level Discrete Wavelet Transform The complete SPIHT algorithm does compression in three steps such as sorting, refinement and quantization. The SPIHT algorithm encodes the image data using three lists such as LIP, LIS and LSP. LIP contains the individual coefficients having the magnitudes smaller than the threshold values. LIS contains the overall wavelet coefficients defined in tree structure having magnitudes smaller than the threshold values. LSP is the set of pixels having magnitude greater than the threshold value of the important pixels. SPIHT is computationally very fast and among the best image compression algorithms known today. The statistical analysis of the output binary stream of encoding, propose a simple and effective method combined with Huffman encode for further compression transform as a branch of mathematics developed rapidly, which has a good localization property in the time domain and frequency domain, can analyse the details of any scale and frequency. so, it superior to Fourier and DCT. It has been widely applied and developed in image processing and compression. III. PROPOSED IMAGE FUSION SYSTEM Fig.2: Proposed Image Fusion System Consider image 1 as fuzzy type image 1 and image 2 as fuzzy type image 2 and image compression and feature extracted by SPIHT technique and apply Image fusion algorithm as averaging method for fused image and compare actual/true image with fused image and calculate PSNR and RMSE to check effectiveness of proposed system. Pseudo code: Im1=DWT (image1) En1=encoding (Im1) De1=decoding (En1) Im1’=IDWT (De1) Id1=Im1-Im1’ Im2=DWT (image2) En2=encoding (Im2) De2=decoding (En2) Im2’=IDWT (De2) Id2=Im2-Im2’ Id=abs(Id1)-(Idf)>=0; ,( , ) 2max1, 0( ) n i ji j x L n othrewiseS x Image1 Image2 SPIHT SPIHT Decoding IDWT FUSION ALGORITHM Fused Image
- 3. Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58 www.ijera.com 57 | P a g e Imf=Idf+Imf Idfm=decoding(Imf) Imf=IDWT (Idfm) This pseudo code shows the implementation Process. IV. FUSION PERFORMANCE EVALUATION The performance of image fusion algorithms can be evaluated when the reference image is available using the following metrics [3, 5, 8, 9]: Root Mean Square Error: It is computed as the root mean square error (RMSE) of the corresponding pixels in the reference image Ir and the fused image If. It will be nearly zero when the reference and fused images are alike and it will increase when the dissimilarity increases. Peak Signal to Noise Ratio: This value will be high when the fused and reference images are alike and higher value implies better fusion. where, L in the number of gray levels in the image. V. RESULTS AND COMPARATIVELY STUDY Fig 3: Reference fuzzy type Image Fig 4: Fuzzy type Image 1 Fig 5: Fuzzy type Image 2 Reference Fuzzy type image mahi.jpg of size 420x342 is shown in Fig-3. The two images to be fused are generated from the Fuzzy type truth image using fuzzy type as shown in Fig-4&5. The fused image is almost similar to reference image and the error image is almost zero. It shows that the fused image contains all information coming from the complementary source images [10]. The fusion quality evaluation metrics are shown in Table-1. The metrics showed in table with bold font are better among others. Fusion with 3 level pyramid or above are giving almost similar performance. Table 1 Proposed Hybrid Technique Pyramid levels Techniq ues 1 3 5 7 RMSE DCTPT 10.0311 9.3924 9.4921 12.7680 FFTPT 7.9812 7.8937 7.9885 8.0075 SPIHT 6.4560 PSNR DCTPT 38.1513 38.4370 38.3912 37.1036 FFTPT 39.1441 39.1920 39.1402 39.1298 SPIHT 40.0651 Fig 6: Comparatively study for RMSE between DCT based Pyramid Transform and Proposed Method
- 4. Shyam Sundar et al. Int. Journal of Engineering Research and Applications www.ijera.com ISSN: 2248-9622, Vol. 5, Issue 9, (Part - 3) September 2015, pp.55-58 www.ijera.com 58 | P a g e Fig 7: Comparatively study for PSNR between DCT based Pyramid Transform and Proposed Method VI. CONCLUSION A novel image fusion technique using SPIHT based has been presented and its performance evaluated. It is concluded that fusion with three level provides better fusion quality. This technique can be used for fusion of fuzzy type images as well as multi model image fusion. The proposed algorithm is very simple, easy to implement and could be used for real time applications. This paper is also provided comparatively studied between proposed and DCT based Pyramid transform technique and validation of the proposed algorithm as Peak Signal to Noise Ratio (PSNR) and Root Mean Square Error (RMSE) in table 1 and plots also. REFERENCE [1] A. Toet, “A morphological pyramid image decomposition”, Pattern Recogn. Lett. 9(4), 255–261 (1989). [2] VPS Naidu and J.R. Raol, ”Pixel-Level Image Fusion using Wavelets and Principal Component Analysis – A Comparative Analysis” Defence Science Journal, Vol.58, No.3, pp.338-352, May 2008. [3] VPS Naidu, “Discrete Cosine Transform- based Image Fusion”, Special Issue on Mobile Intelligent Autonomous System, Defence Science Journal, Vol. 60, No.1, pp.48-54, Jan. 2010. [4] Mahendra Kumar et.al., “Digital Image Watermarking using Fractional Fourier transform via image compression”, In IEEE International Conference on Computational Intelligence and Computing Research 2013 (IEEE ICCIC-2013), 26-28 Dec., 2013. [5] VPS Naidu, “A Novel Image Fusion Technique using DCT based Laplacian Pyramid”, International Journal of Inventive Engineering and Sciences (IJIES) ISSN: 2319–9598, Volume-1, Issue-2, January, 2013. [6] Rick S. Blum, “Robust image fusion using a statistical signal processing approach”, Image Fusion, 6, pp.119-128, 2005. [7] Shutao Li, James T. Kwok and Yaonan Wang, “Combination of images with diverse focuses using the spatial frequency”, Information fusion, 2(3), pp.167-176, 2001. [8] V.P.S. Naidu, J.R. Rao. “Pixel-level Image Fusion using Wavelets and Principal Component Analysis”, Defence Science Journal, pp. 338 -352, 2008. [9] Seetha M, MuraliKrishna I.V & Deekshatulu, B.L, (2005) “Data Fusion Performance Analysis Based on Conventional and Wavelet Transform Techniques”, IEEE Proceedings on Geoscience and Remote Sensing Symposium, Vol 4, pp. 2842-2845. [10] Yang, X.H., Huang, F.Z., Liu,G. 2009 “Urban Remote Image Fusion Using Fuzzy Rule”. IEEE Proceedings of the Eighth International Conference on Machine Learning and Cybernetics, pp. 101-109, (2009).