SlideShare a Scribd company logo
IOSR Journal of Electronics and Communication Engineering (IOSR-JECE)
e-ISSN: 2278-2834,p- ISSN: 2278-8735. Volume 5, Issue 3 (Mar. - Apr. 2013), PP 31-34
www.iosrjournals.org

  A New Method of Image Fusion Technique for Impulse Noise
Removal in Digital Images using the quality Assessment in Spatial
                            Domain
            Ajay kumar Beesetti1, Dr.Rajyalakshmi Valluri2, Minusha Songa3
   1
       Ajay Kumar Beesetti, B.E Student, Department of ECE, ANITS/Andhra University(A.U), Visakhapatnam.
          2
            Dr.Rajyalakshmi Valluri, Associate Professor, Department of ECE, ANITS/A.U, Visakhapatnam.
                       3
                         Minusha Songa, B.Tech Student, Department of ECE,SPMV, Tirupati.

Abstract: Due to advancement of technology, remote sensing plays very important role in satellite based
communication. Satellite gives images in digital format. The digital images are corrupted by impulse noise due
to errors generated in camera sensors, analog-to-digital conversion and communication channels. The noise
density varies depending on various factors namely reflective surfaces, atmospheric variations, noisy
communication channels etc. Impulse noise corruption is very common in digital images. Therefore it is
necessary to remove impulse noise in-order to provide further processing such as edge detection, segmentation,
pattern recognition etc. Filtering a noisy image, while preserving the image details is one of the most important
issues in image processing. The images captured by different sensors, producing different impulse noisy images
are considered and they undergoes iterative filtering algorithm, search for the noise-free pixels within a small
neighborhood. These filtered images are combined to a single image called image fusion, which retains the
important features of the images from individual sensors. In this paper, we introduce an image fusion technique
for impulse noise reduction, where the fused image will combine the uncorrupted pixels of the filtered noisy
images. The performance of the Image Fusion is evaluated by using a reference image quality metric, Structural
similarity Index (SSIM), to estimate how well the important information in the de-noised images is represented
by the fused image. Experimental results show that the fused image has more quality than other filtered images.
Keywords - Impulse Noise, Image fusion, Filtering, De-noised Image.

                                             I.      Introduction
       Digital images are often corrupted during acquisition, transmission or due to faulty memory locations in
hardware [1]. The impulse noise can be caused by a camera due to the faulty nature of the sensor or during
transmission of coded images in a noisy communication channel [2]. Noise is undesired information that
contaminates an image. Noise appears in image from various sources. The digital image acquisition process
converts an optical image into a continuous electrical signal. This electrical signal sampled, is primary process
by which noise appears in digital image. There are several ways through which noise can be introduced into an
image, depending on how the image is created. This is the main problem in remote sensing applications. In this
paper, the images captured by five different sensors are filtered using five different non-linear filtering
algorithms such as Standard Vector Median Filter (VMF), Rank Conditioned VMF, Rank conditioned and
Threshold VMF, Center Weighted VMF and Absolute Deviation VMF., producing five de-noised images. These
de-noised images are fused using our fusion technique, thus obtaining a high quality image.

                                  II.    Impulse Noise In Digital Images
          Impulse noise is independent and uncorrelated to the image pixels and is randomly distributed over the
image. For an impulse noise corrupted image all the image pixels are not noisy, a number of image pixels will
be noisy and the rest of pixels will be noise free. There are different types of impulse noise namely salt and
pepper type of noise and random valued impulse noise. In salt and pepper type of noise the noisy pixels takes
either salt value (gray level -225) or pepper value (grey level -0) and it appears as black and white spots on the
images. In case of random valued impulse noise, noise can take any gray level value from zero to 225. In this
case also noise is randomly distributed over the entire image and probability of occurrence of any gray level
value as noise will be same.

                         III.   Order Static Filters For Color Image Processing
     Order-static filters are nonlinear filters whose response is based on the ordering (ranking) the pixels
contained in the image area encompassed by the filter, and then replacing the value of the center pixel with the
value determined by the ranking result.


                                            www.iosrjournals.org                                        31 | Page
A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the

A. Vector Median Filter:
           In the vector median filter for the ordering of vectors in a particular kernel or mask a suitable distance
measure is chosen. The vector pixels in the window are ordered on the basis of the sum of distances between
each vector pixel and the other pixels in the window. The sum of the distances is arranged in ascending order
and then the same ordering is associated with vector pixels. The vector pixel with the smallest distances is the
vector median pixel. The vector median filter is represented as
                                                 XVMF = vector median (window)                                (1)
If ฮดi is the sum of the distances of the ith vector pixel with all the other vectors in the kernel, then
                                                   ๐›ฟ ๐‘– = ๐‘—๐‘›=1 โˆ†( ๐‘ฅ ๐‘– , ๐‘ฅ๐‘— )                                        (2)
 where (1โ‰ค i โ‰ค N) and Xi and Xj are the vectors, N=9. โˆ†(Xi, Xj) is the distance measure given by the L1 norm or
       the city block distance which is more suited to non correlated noise. The ordering may be illustrated as
                                                      ฮด1 โ‰ค ฮด2 โ‰ค ฮด3 โ‰ค โ€ฆ โ‰ค ฮด9                                    (3)
and this implies the same ordering to the corresponding vector pixels
                                              i.e., X(1) โ‰ค X(2) โ‰ค, ...,โ‰คX(9)                                   (4)
where the subscripts are the ranks. Since the vector pixel with the smallest sum of distances is the vector median
pixel, it will correspond to rank 1 of the ordered pixels,
                                                    i.e., XVMF = X (1)                                             (5)
B. Median Filter:
        The Median Filter [8] as the name implies, replaces the value of the pixel by the median of the intensity
values in the neighborhood of that pixel defined in (1). The pixel with the median magnitude is used to replace
the pixel in the signal studied.
                                 i.e.; Median Filter (x1, x2, โ€ฆxN) = Median(x1,x2,โ€ฆโ€ฆ, xN)                      (6)
 The median filter is more robust with respect to the presence of noise.
C. Center Weighted Vector Median Filter:
       In the case of center weighted median filter the kernel vector pixels are assigned some non negative values
called weights. The central vector pixel is assigned a non negative weight while the weight of the neighboring
pixels is kept unity. The weights denote the number of copies is obtained. The output Y (say), of a weighted
median filter of span N (where N generally denotes the kernel size, N=9) associated with N integer weights,
                                                         W=[W1.W2โ€ฆ..WN]                                       (7)
                                          Y = vector median [W1*X1, W2*X2, โ€ฆ., W9*X9]                          (8)
 Where the vector median [ ] denotes the vector median operation. The center weight is kept odd so that a
central element is always obtained. If the center weight has value of one then the present filter becomes a vector
median filter. If the center weight is given a value greater that the kernel size then the filter performs as an
identity filter. Larger central vector weights imply superior detail preservation and the inferior noise removal
capability as compared to smaller central weights.
D. Spatial Median Filter:
       The Spatial Median Filter (SMF) [5] is a uniform smoothing algorithm with the purpose of removing noise
and fine points of image data while maintaining edges around larger shapes. The SMF is based on the spatial
median quintile function which is a L1 norm metric that measures the difference between two vectors. The
spatial depth between a point and a set of points is defined by
                                                                              1       ๐‘    ๐‘‹โˆ’ ๐‘ฅ ๐‘–
                                     ๐‘บ ๐ท๐‘’๐‘๐‘ก ๐‘• ๐‘‹, ๐‘‹1 , ๐‘‹2 , ๐‘‹3,โ€ฆโ€ฆ. ๐‘‹ ๐‘ = 1 โˆ’          ๐‘– =1                          (9)
                                                                         ๐‘โˆ’1           ๐‘‹โˆ’ ๐‘ฅ ๐‘–


Let r1,r2,โ€ฆ.,rN represent X1,X2,โ€ฆ..XN in rank order such that
         โ‰ฅ SDepth(r1,X1,X2,โ€ฆโ€ฆXN)
         โ‰ฅ SDepth(r2,X1,X2,โ€ฆโ€ฆXN)
         โ‰ฅ SDepth(rN,X1,X2,โ€ฆโ€ฆXN)                                                                                 (10)

and let rc represent the center pixel under the mask.
Then SMF(x1, x2,โ€ฆโ€ฆxn )= r1                                                                                       (11)

                    IV.    Multi Sensor Image Fusion Using Qualitative Assessment
          Given five de-noised images, it is required to combine the images into a single one that has all objects
without producing details that are non-existent in the given images. Here R1 is VMF median filtered image, R2 is
median filtered image, R3 is the CMF filtered image, R4 is the SMF filtered image.
The fusion algorithm consists of the following steps:
 a. Input images Ri for i=1, 2, 3, 4 are divided into non-overlapping rectangular blocks with size of mxn (10x10
blocks). The jth image blocks of Ri are referred by Rij


                                             www.iosrjournals.org                                           32 | Page
A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the

b.Variance (VAR) of Rij are calculated for determining the sharpness values of the corresponding blocks and the
results of Rij are denoted by VARij. VAR is defined as:
                                              1
                                     ๐‘‰๐ด๐‘… =         ๐‘ฅ    ๐‘ฆ ๐‘“ ๐‘ฅ, ๐‘ฆ โˆ’ ๐‘“ )                                    (12)
                                                ๐‘šร—๐‘›
Where ๐‘“ is the average grey level over the image.
                                                 1
                                          ๐‘“ = ๐‘š ร—๐‘› ๐‘ฅ ๐‘ฆ ๐‘“( ๐‘ฅ, ๐‘ฆ )                                              (13)
c. In order to determine the sharper image block, the variances of image blocks from five images are sorted in
descending order and the same ordering is associated with image blocks. The block with the maximum variance
is kept in the fused image. The fusion mechanism is represented as follows:
If VAR (k) is the variance of block Rij, where k denotes the rank, the ordering of variances is given by
                                         VAR(1)>VAR(2) >VAR(3) >VAR(4) >VAR(5)                             (14)
   and this implies the same ordering to the corresponding blocks
                                          R(1) > R (2) > R (3) > R (4) > R (5)                             (15)
where the subscripts are the ranks of the image blocks. Since the block with the smallest variance is in the fused
image, it will correspond to rank 1 of the ordered blocks
                                           i.e. Fused Block = R(1)                                       (16)

                            V.       Qualitative Assessment Of Spatial Domain
          Quality assessment of images has been successfully employed in the authentication area, such as iris
and fingerprint verification, which is used to evaluate the quality of the captured images. Fingerprint quality is
usually defined as a measure of the clarity of the ridge and valley structures. In the multi-focus image fusion, it
is the first step to investigate whether the region of the image is clarity or not, so we incorporate the quality
assessment into multi-focus image fusion algorithm. Generally speaking, the quality assessment can be
classified into two ways, namely the frequency domain assessment and the spatial domain assessment. We
incorporate the latter one into this algorithm. In order to assess the image quality in a local region, we partition a
given image into a lattice of blocks of size bxb. For each block B, let gs= ( gsx,gsy) denote the gradient of the
gray level intensity at site s๐œ–B. The covariance matrix of the gradient vectors for all b2 sites in
                                          1                  ๐‘—    ๐‘—
                                     ๐ฝ = ๐‘ 2 ๐‘ โˆˆ๐‘ ๐‘” ๐‘ ๐‘‡ ๐‘” ๐‘  = 11 12                                                 (17)
                                                             ๐‘—21 ๐‘—22
The above symmetric matrix is positive semi definite with Eigen values
                                         1
                                    ๐›พ1 = 2 ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ + ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’ 4๐‘‘๐‘’๐‘ก ๐ฝ                                            (18)
                                     1
                                  ๐›พ2 = 2 ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’          ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’ 4๐‘‘๐‘’๐‘ก ๐ฝ                                      (19)
Where trace (J) =j11+j22, det (J) = j11j22-j12 and ๐›พ1โ‰ฅ ๐›พ2.
                                              2

Quality assessment defined as
                                                           ๐›พ= ๐›พ1-๐›พ2                                              (20)
This measure reflects the clarity of the local region. The clearer the local region is, the bigger the measure is.

                                         VI.      Experimental Results
          The performance evaluation of filtering using image fusion method is tested on the true color remote
sensing image with 290x290 pixels. The impulse noise is added into the image with different noise densities.
Here we are assuming, n=4, i.e.; 4 sensors. The noisy image is processed using a iterative filtering algorithm
based on the noise density in the image. These filtered images are fused into a single image using Quality
Assessment of Spatial Domain. The experimental results are shown in Figure 1. Table (1) shows the results of
PSNR values of fused image with different noise densities. We used the image quality metric, peak signal-to-
noise ratio (PSNR), to measure the quality of the restored image. The PSNR measure is defined as
                                                                (255 )2
                                        ๐‘ƒ๐‘†๐‘๐‘… = 10 log10 MSE                                                      (21)
 where MSE is the mean squared error between the original noise-free image and the restored image.

                                    SSIM              MSE            PSNR
Fusion                              0.99               50             29
VMF                                 0.93              200            25
Median Filter                       0.92              220            22
Center Weighted Median Filter       0.91              650            20
Spatial Median Filter               0.90              700            18

Table 1: Performance Evaluation of Fusion Method with filter responses at 30% of noise intensity


                                               www.iosrjournals.org                                         33 | Page
A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the




                          Fig 1: Fused Image and responses of different kinds of filters
a) Original image; b) Image with added Noise; c) Vector Median Filter; d) Center Weighted Median Filter;
e) Spatial Median Filter Image; f) Fused Image

                                                      VII.      Conclusion
         This paper presents a new method of image fusion technique for removal of impulse noise in images.
The images captured by sensors undergo filtering using VMF and SMF, and then the two individual de-noised
images are fused to obtain a high quality image. The Quality of the images is evaluated using Structural
Similarity Index (SSIM), Peak Signal-to-Noise ratio (PSNR) and Mean Square Error (MSE) with different noise
densities. The proposed techniques are algorithmically simple and can be used for real time imaging
applications.

                                                           References
[1]    Nikolaos Mitianoudis, Tania Stathaki, Pixel-based and Region-based Image Fusion schemes using ICA bases, Communications and
       Signal Processing group, Preprint submitted to Elsevier Science, 13th Dec 2007.
[2].   Reihard Berstein, โ€œAdaptive nonlinear filters for simultaneous removal of different kinds of noise in images,โ€ IEEE Trans on
       circuits and systems, Vol.cas-34, no11, pp.127-1291, Nivember1987.
[3]    A. Hyvร„arinen, P. O. Hoyer, and E. Oja. Image denoising by sparse code shrinkage. In S. Haykin and B. Kosko, editors, Intelligent
       Signal Processing. IEEE Press, 2001.
[4]    P. Hill, N. Canagarajah, and D. Bull. Image fusion using complex wavelets. In Proc.13th BritishMachineVisionConference,
       Cardiยฎ, UK, 2002.
[5]    James C. Church, Yixin Chen, and Stephen V. Rice , โ€•A Spatial Median Filter for Noise Removal in Digital Imagesโ€•, 2008
       IEEE.
[6]    M. Vett, erli and C. Herley, โ€œWavelets and filter banks: theory and design,โ€ IEEE Tram Signal Processing, vol. 40, pp. 2207-2232,
       September 1992.
[7]    P. Burt and R. Lolczynski, โ€œEnhanced image capture through fusion,โ€ in Proc. the Fourtt International Conference on Computer
       Vision, pp. 173- 182, 1993.
[8]    Tao Chen, Kai-Kaung Ma and Li-Hui Chen, โ€•โ€Tri-state median filter for image Denoisingโ€, IEEE Transactions on Image
       Processing, Vol 8, no.12, pp.1834-1838, December 1999.
[9]    S.Indu, Chaveli Ramesh, โ€• โ€œA noise fading technique for images corrupted with impulse noiseโ€, Proceedings of ICCTA07,IEEE.




                                                    www.iosrjournals.org                                                    34 | Page

More Related Content

What's hot

Comparative analysis of filters and wavelet based thresholding methods for im...
Comparative analysis of filters and wavelet based thresholding methods for im...Comparative analysis of filters and wavelet based thresholding methods for im...
Comparative analysis of filters and wavelet based thresholding methods for im...
csandit
ย 
Data structures
Data structuresData structures
Data structuresAshish Kumar
ย 
International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)
ijceronline
ย 
Digitized images and
Digitized images andDigitized images and
Digitized images andAshish Kumar
ย 
Bh044365368
Bh044365368Bh044365368
Bh044365368
IJERA Editor
ย 
J010245458
J010245458J010245458
J010245458
IOSR Journals
ย 
Image filtering in Digital image processing
Image filtering in Digital image processingImage filtering in Digital image processing
Image filtering in Digital image processing
Abinaya B
ย 
Image segmentation
Image segmentationImage segmentation
Image segmentation
Maneesha Krishnan
ย 
Lecture 6
Lecture 6Lecture 6
Lecture 6
Wael Sharba
ย 
It 603
It 603It 603
It 603
manjunathan_g
ย 
03 cie552 image_filtering_spatial
03 cie552 image_filtering_spatial03 cie552 image_filtering_spatial
03 cie552 image_filtering_spatial
Elsayed Hemayed
ย 
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
IDES Editor
ย 
Research Inventy : International Journal of Engineering and Science is publis...
Research Inventy : International Journal of Engineering and Science is publis...Research Inventy : International Journal of Engineering and Science is publis...
Research Inventy : International Journal of Engineering and Science is publis...researchinventy
ย 
Spatial Filters (Digital Image Processing)
Spatial Filters (Digital Image Processing)Spatial Filters (Digital Image Processing)
Spatial Filters (Digital Image Processing)
Kalyan Acharjya
ย 
DIP - Image Restoration
DIP - Image RestorationDIP - Image Restoration
DIP - Image Restoration
Eng. Dr. Dennis N. Mwighusa
ย 
Image Denoising Using WEAD
Image Denoising Using WEADImage Denoising Using WEAD
Image Denoising Using WEAD
QuEST Global (erstwhile NeST Software)
ย 
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
International Journal of Science and Research (IJSR)
ย 
Image restoration
Image restorationImage restoration
Image restoration
Upendra Pratap Singh
ย 
The International Journal of Engineering and Science (The IJES)
The International Journal of Engineering and Science (The IJES)The International Journal of Engineering and Science (The IJES)
The International Journal of Engineering and Science (The IJES)
theijes
ย 

What's hot (20)

Comparative analysis of filters and wavelet based thresholding methods for im...
Comparative analysis of filters and wavelet based thresholding methods for im...Comparative analysis of filters and wavelet based thresholding methods for im...
Comparative analysis of filters and wavelet based thresholding methods for im...
ย 
Data structures
Data structuresData structures
Data structures
ย 
International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)International Journal of Computational Engineering Research(IJCER)
International Journal of Computational Engineering Research(IJCER)
ย 
Digitized images and
Digitized images andDigitized images and
Digitized images and
ย 
Bh044365368
Bh044365368Bh044365368
Bh044365368
ย 
J010245458
J010245458J010245458
J010245458
ย 
Image filtering in Digital image processing
Image filtering in Digital image processingImage filtering in Digital image processing
Image filtering in Digital image processing
ย 
Image segmentation
Image segmentationImage segmentation
Image segmentation
ย 
Lecture 6
Lecture 6Lecture 6
Lecture 6
ย 
Im seg04
Im seg04Im seg04
Im seg04
ย 
It 603
It 603It 603
It 603
ย 
03 cie552 image_filtering_spatial
03 cie552 image_filtering_spatial03 cie552 image_filtering_spatial
03 cie552 image_filtering_spatial
ย 
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
Random Valued Impulse Noise Removal in Colour Images using Adaptive Threshold...
ย 
Research Inventy : International Journal of Engineering and Science is publis...
Research Inventy : International Journal of Engineering and Science is publis...Research Inventy : International Journal of Engineering and Science is publis...
Research Inventy : International Journal of Engineering and Science is publis...
ย 
Spatial Filters (Digital Image Processing)
Spatial Filters (Digital Image Processing)Spatial Filters (Digital Image Processing)
Spatial Filters (Digital Image Processing)
ย 
DIP - Image Restoration
DIP - Image RestorationDIP - Image Restoration
DIP - Image Restoration
ย 
Image Denoising Using WEAD
Image Denoising Using WEADImage Denoising Using WEAD
Image Denoising Using WEAD
ย 
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
Study and Analysis of Multiwavelet Transform with Threshold in Image Denoisin...
ย 
Image restoration
Image restorationImage restoration
Image restoration
ย 
The International Journal of Engineering and Science (The IJES)
The International Journal of Engineering and Science (The IJES)The International Journal of Engineering and Science (The IJES)
The International Journal of Engineering and Science (The IJES)
ย 

Viewers also liked

The Analysis of Plyometric Training Program on University Handball Players
The Analysis of Plyometric Training Program on University Handball PlayersThe Analysis of Plyometric Training Program on University Handball Players
The Analysis of Plyometric Training Program on University Handball Players
IOSR Journals
ย 
Multicloud Deployment of Computing Clusters for Loosely Coupled Multi Task C...
Multicloud Deployment of Computing Clusters for Loosely  Coupled Multi Task C...Multicloud Deployment of Computing Clusters for Loosely  Coupled Multi Task C...
Multicloud Deployment of Computing Clusters for Loosely Coupled Multi Task C...
IOSR Journals
ย 
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
IOSR Journals
ย 
D0331923
D0331923D0331923
D0331923
IOSR Journals
ย 
A Survey of Fuzzy Logic Based Congestion Estimation Techniques in Wireless S...
A Survey of Fuzzy Logic Based Congestion Estimation  Techniques in Wireless S...A Survey of Fuzzy Logic Based Congestion Estimation  Techniques in Wireless S...
A Survey of Fuzzy Logic Based Congestion Estimation Techniques in Wireless S...
IOSR Journals
ย 
B0560508
B0560508B0560508
B0560508
IOSR Journals
ย 
Notions via ฮฒ*-open sets in topological spaces
Notions via ฮฒ*-open sets in topological spacesNotions via ฮฒ*-open sets in topological spaces
Notions via ฮฒ*-open sets in topological spaces
IOSR Journals
ย 
D0321624
D0321624D0321624
D0321624
IOSR Journals
ย 
Ingoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
Ingoldian Fungi in Kigga Falls, Chikmagalur District, KarnatakaIngoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
Ingoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
IOSR Journals
ย 
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
IOSR Journals
ย 
Prevalence and morphological details of Nyctotherus periplanetae in the host ...
Prevalence and morphological details of Nyctotherus periplanetae in the host ...Prevalence and morphological details of Nyctotherus periplanetae in the host ...
Prevalence and morphological details of Nyctotherus periplanetae in the host ...
IOSR Journals
ย 
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced RatsHepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
IOSR Journals
ย 
A Quantified Approach for large Dataset Compression in Association Mining
A Quantified Approach for large Dataset Compression in Association MiningA Quantified Approach for large Dataset Compression in Association Mining
A Quantified Approach for large Dataset Compression in Association Mining
IOSR Journals
ย 
Estimating the Statistical Significance of Classifiers used in the Predictio...
Estimating the Statistical Significance of Classifiers used in the  Predictio...Estimating the Statistical Significance of Classifiers used in the  Predictio...
Estimating the Statistical Significance of Classifiers used in the Predictio...
IOSR Journals
ย 
F0522833
F0522833F0522833
F0522833
IOSR Journals
ย 
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
IOSR Journals
ย 
Optimal Estimating Sequence for a Hilbert Space Valued Parameter
Optimal Estimating Sequence for a Hilbert Space Valued ParameterOptimal Estimating Sequence for a Hilbert Space Valued Parameter
Optimal Estimating Sequence for a Hilbert Space Valued Parameter
IOSR Journals
ย 
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
IOSR Journals
ย 
A preliminary study on the toxic potentials of shea butter effluent using Cla...
A preliminary study on the toxic potentials of shea butter effluent using Cla...A preliminary study on the toxic potentials of shea butter effluent using Cla...
A preliminary study on the toxic potentials of shea butter effluent using Cla...
IOSR Journals
ย 

Viewers also liked (20)

F0733038
F0733038F0733038
F0733038
ย 
The Analysis of Plyometric Training Program on University Handball Players
The Analysis of Plyometric Training Program on University Handball PlayersThe Analysis of Plyometric Training Program on University Handball Players
The Analysis of Plyometric Training Program on University Handball Players
ย 
Multicloud Deployment of Computing Clusters for Loosely Coupled Multi Task C...
Multicloud Deployment of Computing Clusters for Loosely  Coupled Multi Task C...Multicloud Deployment of Computing Clusters for Loosely  Coupled Multi Task C...
Multicloud Deployment of Computing Clusters for Loosely Coupled Multi Task C...
ย 
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
Spatial Correlation Based Medium Access Control Protocol Using DSR & AODV Rou...
ย 
D0331923
D0331923D0331923
D0331923
ย 
A Survey of Fuzzy Logic Based Congestion Estimation Techniques in Wireless S...
A Survey of Fuzzy Logic Based Congestion Estimation  Techniques in Wireless S...A Survey of Fuzzy Logic Based Congestion Estimation  Techniques in Wireless S...
A Survey of Fuzzy Logic Based Congestion Estimation Techniques in Wireless S...
ย 
B0560508
B0560508B0560508
B0560508
ย 
Notions via ฮฒ*-open sets in topological spaces
Notions via ฮฒ*-open sets in topological spacesNotions via ฮฒ*-open sets in topological spaces
Notions via ฮฒ*-open sets in topological spaces
ย 
D0321624
D0321624D0321624
D0321624
ย 
Ingoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
Ingoldian Fungi in Kigga Falls, Chikmagalur District, KarnatakaIngoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
Ingoldian Fungi in Kigga Falls, Chikmagalur District, Karnataka
ย 
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
Detection Of Escherichia Coli, An Indicator Of Feacal Contamination, In Drink...
ย 
Prevalence and morphological details of Nyctotherus periplanetae in the host ...
Prevalence and morphological details of Nyctotherus periplanetae in the host ...Prevalence and morphological details of Nyctotherus periplanetae in the host ...
Prevalence and morphological details of Nyctotherus periplanetae in the host ...
ย 
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced RatsHepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
Hepatoprotective Activity of Chara Parpam in Ccl4 Induced Rats
ย 
A Quantified Approach for large Dataset Compression in Association Mining
A Quantified Approach for large Dataset Compression in Association MiningA Quantified Approach for large Dataset Compression in Association Mining
A Quantified Approach for large Dataset Compression in Association Mining
ย 
Estimating the Statistical Significance of Classifiers used in the Predictio...
Estimating the Statistical Significance of Classifiers used in the  Predictio...Estimating the Statistical Significance of Classifiers used in the  Predictio...
Estimating the Statistical Significance of Classifiers used in the Predictio...
ย 
F0522833
F0522833F0522833
F0522833
ย 
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
Protein structure determination of insulin of zebra fish (Danio rerio) by hom...
ย 
Optimal Estimating Sequence for a Hilbert Space Valued Parameter
Optimal Estimating Sequence for a Hilbert Space Valued ParameterOptimal Estimating Sequence for a Hilbert Space Valued Parameter
Optimal Estimating Sequence for a Hilbert Space Valued Parameter
ย 
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
Callus Induction and Plantlet Regeneration in Orthosiphon aristatus (Blume) M...
ย 
A preliminary study on the toxic potentials of shea butter effluent using Cla...
A preliminary study on the toxic potentials of shea butter effluent using Cla...A preliminary study on the toxic potentials of shea butter effluent using Cla...
A preliminary study on the toxic potentials of shea butter effluent using Cla...
ย 

Similar to F0533134

M017218088
M017218088M017218088
M017218088
IOSR Journals
ย 
Image Noise Removal by Dual Threshold Median Filter for RVIN
Image Noise Removal by Dual Threshold Median Filter for RVINImage Noise Removal by Dual Threshold Median Filter for RVIN
Image Noise Removal by Dual Threshold Median Filter for RVIN
IOSR Journals
ย 
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
IJEACS
ย 
Report medical image processing image slice interpolation and noise removal i...
Report medical image processing image slice interpolation and noise removal i...Report medical image processing image slice interpolation and noise removal i...
Report medical image processing image slice interpolation and noise removal i...
Shashank
ย 
[IJCT-V3I2P34] Authors: Palwinder Singh
[IJCT-V3I2P34] Authors: Palwinder Singh[IJCT-V3I2P34] Authors: Palwinder Singh
[IJCT-V3I2P34] Authors: Palwinder Singh
IJET - International Journal of Engineering and Techniques
ย 
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color ImagesReduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
IDES Editor
ย 
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
CSCJournals
ย 
PID3474431
PID3474431PID3474431
PID3474431pranay yadav
ย 
3 intensity transformations and spatial filtering slides
3 intensity transformations and spatial filtering slides3 intensity transformations and spatial filtering slides
3 intensity transformations and spatial filtering slides
BHAGYAPRASADBUGGE
ย 
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
IJERA Editor
ย 
DIP-Questions.pdf
DIP-Questions.pdfDIP-Questions.pdf
DIP-Questions.pdf
AratiKothari2
ย 
Image Fusion Based On Wavelet And Curvelet Transform
Image Fusion Based On Wavelet And Curvelet TransformImage Fusion Based On Wavelet And Curvelet Transform
Image Fusion Based On Wavelet And Curvelet TransformIOSR Journals
ย 
Nonlinear Transformation Based Detection And Directional Mean Filter to Remo...
Nonlinear Transformation Based Detection And Directional  Mean Filter to Remo...Nonlinear Transformation Based Detection And Directional  Mean Filter to Remo...
Nonlinear Transformation Based Detection And Directional Mean Filter to Remo...
IJMER
ย 
Comparative study of Salt & Pepper filters and Gaussian filters
Comparative study of Salt & Pepper filters and Gaussian filtersComparative study of Salt & Pepper filters and Gaussian filters
Comparative study of Salt & Pepper filters and Gaussian filtersAnkush Srivastava
ย 
Paper id 24201427
Paper id 24201427Paper id 24201427
Paper id 24201427IJRAT
ย 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...ijceronline
ย 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
ijceronline
ย 
www.ijerd.com
www.ijerd.comwww.ijerd.com
www.ijerd.comIJERD Editor
ย 
K010615562
K010615562K010615562
K010615562
IOSR Journals
ย 
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
iosrjce
ย 

Similar to F0533134 (20)

M017218088
M017218088M017218088
M017218088
ย 
Image Noise Removal by Dual Threshold Median Filter for RVIN
Image Noise Removal by Dual Threshold Median Filter for RVINImage Noise Removal by Dual Threshold Median Filter for RVIN
Image Noise Removal by Dual Threshold Median Filter for RVIN
ย 
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
Performance Assessment of Several Filters for Removing Salt and Pepper Noise,...
ย 
Report medical image processing image slice interpolation and noise removal i...
Report medical image processing image slice interpolation and noise removal i...Report medical image processing image slice interpolation and noise removal i...
Report medical image processing image slice interpolation and noise removal i...
ย 
[IJCT-V3I2P34] Authors: Palwinder Singh
[IJCT-V3I2P34] Authors: Palwinder Singh[IJCT-V3I2P34] Authors: Palwinder Singh
[IJCT-V3I2P34] Authors: Palwinder Singh
ย 
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color ImagesReduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
Reduced Ordering Based Approach to Impulsive Noise Suppression in Color Images
ย 
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
Robustness of Median Filter For Suppression of Salt and Pepper Noise (SPN) an...
ย 
PID3474431
PID3474431PID3474431
PID3474431
ย 
3 intensity transformations and spatial filtering slides
3 intensity transformations and spatial filtering slides3 intensity transformations and spatial filtering slides
3 intensity transformations and spatial filtering slides
ย 
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
Analysis of Non Linear Filters with Various Density of Impulse Noise for Diff...
ย 
DIP-Questions.pdf
DIP-Questions.pdfDIP-Questions.pdf
DIP-Questions.pdf
ย 
Image Fusion Based On Wavelet And Curvelet Transform
Image Fusion Based On Wavelet And Curvelet TransformImage Fusion Based On Wavelet And Curvelet Transform
Image Fusion Based On Wavelet And Curvelet Transform
ย 
Nonlinear Transformation Based Detection And Directional Mean Filter to Remo...
Nonlinear Transformation Based Detection And Directional  Mean Filter to Remo...Nonlinear Transformation Based Detection And Directional  Mean Filter to Remo...
Nonlinear Transformation Based Detection And Directional Mean Filter to Remo...
ย 
Comparative study of Salt & Pepper filters and Gaussian filters
Comparative study of Salt & Pepper filters and Gaussian filtersComparative study of Salt & Pepper filters and Gaussian filters
Comparative study of Salt & Pepper filters and Gaussian filters
ย 
Paper id 24201427
Paper id 24201427Paper id 24201427
Paper id 24201427
ย 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
ย 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...IJCER (www.ijceronline.com) International Journal of computational Engineerin...
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
ย 
www.ijerd.com
www.ijerd.comwww.ijerd.com
www.ijerd.com
ย 
K010615562
K010615562K010615562
K010615562
ย 
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
Edge Detection with Detail Preservation for RVIN Using Adaptive Threshold Fil...
ย 

More from IOSR Journals

A011140104
A011140104A011140104
A011140104
IOSR Journals
ย 
M0111397100
M0111397100M0111397100
M0111397100
IOSR Journals
ย 
L011138596
L011138596L011138596
L011138596
IOSR Journals
ย 
K011138084
K011138084K011138084
K011138084
IOSR Journals
ย 
J011137479
J011137479J011137479
J011137479
IOSR Journals
ย 
I011136673
I011136673I011136673
I011136673
IOSR Journals
ย 
G011134454
G011134454G011134454
G011134454
IOSR Journals
ย 
H011135565
H011135565H011135565
H011135565
IOSR Journals
ย 
F011134043
F011134043F011134043
F011134043
IOSR Journals
ย 
E011133639
E011133639E011133639
E011133639
IOSR Journals
ย 
D011132635
D011132635D011132635
D011132635
IOSR Journals
ย 
C011131925
C011131925C011131925
C011131925
IOSR Journals
ย 
B011130918
B011130918B011130918
B011130918
IOSR Journals
ย 
A011130108
A011130108A011130108
A011130108
IOSR Journals
ย 
I011125160
I011125160I011125160
I011125160
IOSR Journals
ย 
H011124050
H011124050H011124050
H011124050
IOSR Journals
ย 
G011123539
G011123539G011123539
G011123539
IOSR Journals
ย 
F011123134
F011123134F011123134
F011123134
IOSR Journals
ย 
E011122530
E011122530E011122530
E011122530
IOSR Journals
ย 
D011121524
D011121524D011121524
D011121524
IOSR Journals
ย 

More from IOSR Journals (20)

A011140104
A011140104A011140104
A011140104
ย 
M0111397100
M0111397100M0111397100
M0111397100
ย 
L011138596
L011138596L011138596
L011138596
ย 
K011138084
K011138084K011138084
K011138084
ย 
J011137479
J011137479J011137479
J011137479
ย 
I011136673
I011136673I011136673
I011136673
ย 
G011134454
G011134454G011134454
G011134454
ย 
H011135565
H011135565H011135565
H011135565
ย 
F011134043
F011134043F011134043
F011134043
ย 
E011133639
E011133639E011133639
E011133639
ย 
D011132635
D011132635D011132635
D011132635
ย 
C011131925
C011131925C011131925
C011131925
ย 
B011130918
B011130918B011130918
B011130918
ย 
A011130108
A011130108A011130108
A011130108
ย 
I011125160
I011125160I011125160
I011125160
ย 
H011124050
H011124050H011124050
H011124050
ย 
G011123539
G011123539G011123539
G011123539
ย 
F011123134
F011123134F011123134
F011123134
ย 
E011122530
E011122530E011122530
E011122530
ย 
D011121524
D011121524D011121524
D011121524
ย 

F0533134

  • 1. IOSR Journal of Electronics and Communication Engineering (IOSR-JECE) e-ISSN: 2278-2834,p- ISSN: 2278-8735. Volume 5, Issue 3 (Mar. - Apr. 2013), PP 31-34 www.iosrjournals.org A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the quality Assessment in Spatial Domain Ajay kumar Beesetti1, Dr.Rajyalakshmi Valluri2, Minusha Songa3 1 Ajay Kumar Beesetti, B.E Student, Department of ECE, ANITS/Andhra University(A.U), Visakhapatnam. 2 Dr.Rajyalakshmi Valluri, Associate Professor, Department of ECE, ANITS/A.U, Visakhapatnam. 3 Minusha Songa, B.Tech Student, Department of ECE,SPMV, Tirupati. Abstract: Due to advancement of technology, remote sensing plays very important role in satellite based communication. Satellite gives images in digital format. The digital images are corrupted by impulse noise due to errors generated in camera sensors, analog-to-digital conversion and communication channels. The noise density varies depending on various factors namely reflective surfaces, atmospheric variations, noisy communication channels etc. Impulse noise corruption is very common in digital images. Therefore it is necessary to remove impulse noise in-order to provide further processing such as edge detection, segmentation, pattern recognition etc. Filtering a noisy image, while preserving the image details is one of the most important issues in image processing. The images captured by different sensors, producing different impulse noisy images are considered and they undergoes iterative filtering algorithm, search for the noise-free pixels within a small neighborhood. These filtered images are combined to a single image called image fusion, which retains the important features of the images from individual sensors. In this paper, we introduce an image fusion technique for impulse noise reduction, where the fused image will combine the uncorrupted pixels of the filtered noisy images. The performance of the Image Fusion is evaluated by using a reference image quality metric, Structural similarity Index (SSIM), to estimate how well the important information in the de-noised images is represented by the fused image. Experimental results show that the fused image has more quality than other filtered images. Keywords - Impulse Noise, Image fusion, Filtering, De-noised Image. I. Introduction Digital images are often corrupted during acquisition, transmission or due to faulty memory locations in hardware [1]. The impulse noise can be caused by a camera due to the faulty nature of the sensor or during transmission of coded images in a noisy communication channel [2]. Noise is undesired information that contaminates an image. Noise appears in image from various sources. The digital image acquisition process converts an optical image into a continuous electrical signal. This electrical signal sampled, is primary process by which noise appears in digital image. There are several ways through which noise can be introduced into an image, depending on how the image is created. This is the main problem in remote sensing applications. In this paper, the images captured by five different sensors are filtered using five different non-linear filtering algorithms such as Standard Vector Median Filter (VMF), Rank Conditioned VMF, Rank conditioned and Threshold VMF, Center Weighted VMF and Absolute Deviation VMF., producing five de-noised images. These de-noised images are fused using our fusion technique, thus obtaining a high quality image. II. Impulse Noise In Digital Images Impulse noise is independent and uncorrelated to the image pixels and is randomly distributed over the image. For an impulse noise corrupted image all the image pixels are not noisy, a number of image pixels will be noisy and the rest of pixels will be noise free. There are different types of impulse noise namely salt and pepper type of noise and random valued impulse noise. In salt and pepper type of noise the noisy pixels takes either salt value (gray level -225) or pepper value (grey level -0) and it appears as black and white spots on the images. In case of random valued impulse noise, noise can take any gray level value from zero to 225. In this case also noise is randomly distributed over the entire image and probability of occurrence of any gray level value as noise will be same. III. Order Static Filters For Color Image Processing Order-static filters are nonlinear filters whose response is based on the ordering (ranking) the pixels contained in the image area encompassed by the filter, and then replacing the value of the center pixel with the value determined by the ranking result. www.iosrjournals.org 31 | Page
  • 2. A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the A. Vector Median Filter: In the vector median filter for the ordering of vectors in a particular kernel or mask a suitable distance measure is chosen. The vector pixels in the window are ordered on the basis of the sum of distances between each vector pixel and the other pixels in the window. The sum of the distances is arranged in ascending order and then the same ordering is associated with vector pixels. The vector pixel with the smallest distances is the vector median pixel. The vector median filter is represented as XVMF = vector median (window) (1) If ฮดi is the sum of the distances of the ith vector pixel with all the other vectors in the kernel, then ๐›ฟ ๐‘– = ๐‘—๐‘›=1 โˆ†( ๐‘ฅ ๐‘– , ๐‘ฅ๐‘— ) (2) where (1โ‰ค i โ‰ค N) and Xi and Xj are the vectors, N=9. โˆ†(Xi, Xj) is the distance measure given by the L1 norm or the city block distance which is more suited to non correlated noise. The ordering may be illustrated as ฮด1 โ‰ค ฮด2 โ‰ค ฮด3 โ‰ค โ€ฆ โ‰ค ฮด9 (3) and this implies the same ordering to the corresponding vector pixels i.e., X(1) โ‰ค X(2) โ‰ค, ...,โ‰คX(9) (4) where the subscripts are the ranks. Since the vector pixel with the smallest sum of distances is the vector median pixel, it will correspond to rank 1 of the ordered pixels, i.e., XVMF = X (1) (5) B. Median Filter: The Median Filter [8] as the name implies, replaces the value of the pixel by the median of the intensity values in the neighborhood of that pixel defined in (1). The pixel with the median magnitude is used to replace the pixel in the signal studied. i.e.; Median Filter (x1, x2, โ€ฆxN) = Median(x1,x2,โ€ฆโ€ฆ, xN) (6) The median filter is more robust with respect to the presence of noise. C. Center Weighted Vector Median Filter: In the case of center weighted median filter the kernel vector pixels are assigned some non negative values called weights. The central vector pixel is assigned a non negative weight while the weight of the neighboring pixels is kept unity. The weights denote the number of copies is obtained. The output Y (say), of a weighted median filter of span N (where N generally denotes the kernel size, N=9) associated with N integer weights, W=[W1.W2โ€ฆ..WN] (7) Y = vector median [W1*X1, W2*X2, โ€ฆ., W9*X9] (8) Where the vector median [ ] denotes the vector median operation. The center weight is kept odd so that a central element is always obtained. If the center weight has value of one then the present filter becomes a vector median filter. If the center weight is given a value greater that the kernel size then the filter performs as an identity filter. Larger central vector weights imply superior detail preservation and the inferior noise removal capability as compared to smaller central weights. D. Spatial Median Filter: The Spatial Median Filter (SMF) [5] is a uniform smoothing algorithm with the purpose of removing noise and fine points of image data while maintaining edges around larger shapes. The SMF is based on the spatial median quintile function which is a L1 norm metric that measures the difference between two vectors. The spatial depth between a point and a set of points is defined by 1 ๐‘ ๐‘‹โˆ’ ๐‘ฅ ๐‘– ๐‘บ ๐ท๐‘’๐‘๐‘ก ๐‘• ๐‘‹, ๐‘‹1 , ๐‘‹2 , ๐‘‹3,โ€ฆโ€ฆ. ๐‘‹ ๐‘ = 1 โˆ’ ๐‘– =1 (9) ๐‘โˆ’1 ๐‘‹โˆ’ ๐‘ฅ ๐‘– Let r1,r2,โ€ฆ.,rN represent X1,X2,โ€ฆ..XN in rank order such that โ‰ฅ SDepth(r1,X1,X2,โ€ฆโ€ฆXN) โ‰ฅ SDepth(r2,X1,X2,โ€ฆโ€ฆXN) โ‰ฅ SDepth(rN,X1,X2,โ€ฆโ€ฆXN) (10) and let rc represent the center pixel under the mask. Then SMF(x1, x2,โ€ฆโ€ฆxn )= r1 (11) IV. Multi Sensor Image Fusion Using Qualitative Assessment Given five de-noised images, it is required to combine the images into a single one that has all objects without producing details that are non-existent in the given images. Here R1 is VMF median filtered image, R2 is median filtered image, R3 is the CMF filtered image, R4 is the SMF filtered image. The fusion algorithm consists of the following steps: a. Input images Ri for i=1, 2, 3, 4 are divided into non-overlapping rectangular blocks with size of mxn (10x10 blocks). The jth image blocks of Ri are referred by Rij www.iosrjournals.org 32 | Page
  • 3. A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the b.Variance (VAR) of Rij are calculated for determining the sharpness values of the corresponding blocks and the results of Rij are denoted by VARij. VAR is defined as: 1 ๐‘‰๐ด๐‘… = ๐‘ฅ ๐‘ฆ ๐‘“ ๐‘ฅ, ๐‘ฆ โˆ’ ๐‘“ ) (12) ๐‘šร—๐‘› Where ๐‘“ is the average grey level over the image. 1 ๐‘“ = ๐‘š ร—๐‘› ๐‘ฅ ๐‘ฆ ๐‘“( ๐‘ฅ, ๐‘ฆ ) (13) c. In order to determine the sharper image block, the variances of image blocks from five images are sorted in descending order and the same ordering is associated with image blocks. The block with the maximum variance is kept in the fused image. The fusion mechanism is represented as follows: If VAR (k) is the variance of block Rij, where k denotes the rank, the ordering of variances is given by VAR(1)>VAR(2) >VAR(3) >VAR(4) >VAR(5) (14) and this implies the same ordering to the corresponding blocks R(1) > R (2) > R (3) > R (4) > R (5) (15) where the subscripts are the ranks of the image blocks. Since the block with the smallest variance is in the fused image, it will correspond to rank 1 of the ordered blocks i.e. Fused Block = R(1) (16) V. Qualitative Assessment Of Spatial Domain Quality assessment of images has been successfully employed in the authentication area, such as iris and fingerprint verification, which is used to evaluate the quality of the captured images. Fingerprint quality is usually defined as a measure of the clarity of the ridge and valley structures. In the multi-focus image fusion, it is the first step to investigate whether the region of the image is clarity or not, so we incorporate the quality assessment into multi-focus image fusion algorithm. Generally speaking, the quality assessment can be classified into two ways, namely the frequency domain assessment and the spatial domain assessment. We incorporate the latter one into this algorithm. In order to assess the image quality in a local region, we partition a given image into a lattice of blocks of size bxb. For each block B, let gs= ( gsx,gsy) denote the gradient of the gray level intensity at site s๐œ–B. The covariance matrix of the gradient vectors for all b2 sites in 1 ๐‘— ๐‘— ๐ฝ = ๐‘ 2 ๐‘ โˆˆ๐‘ ๐‘” ๐‘ ๐‘‡ ๐‘” ๐‘  = 11 12 (17) ๐‘—21 ๐‘—22 The above symmetric matrix is positive semi definite with Eigen values 1 ๐›พ1 = 2 ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ + ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’ 4๐‘‘๐‘’๐‘ก ๐ฝ (18) 1 ๐›พ2 = 2 ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’ ๐‘ก๐‘Ÿ๐‘Ž๐‘๐‘’ ๐ฝ โˆ’ 4๐‘‘๐‘’๐‘ก ๐ฝ (19) Where trace (J) =j11+j22, det (J) = j11j22-j12 and ๐›พ1โ‰ฅ ๐›พ2. 2 Quality assessment defined as ๐›พ= ๐›พ1-๐›พ2 (20) This measure reflects the clarity of the local region. The clearer the local region is, the bigger the measure is. VI. Experimental Results The performance evaluation of filtering using image fusion method is tested on the true color remote sensing image with 290x290 pixels. The impulse noise is added into the image with different noise densities. Here we are assuming, n=4, i.e.; 4 sensors. The noisy image is processed using a iterative filtering algorithm based on the noise density in the image. These filtered images are fused into a single image using Quality Assessment of Spatial Domain. The experimental results are shown in Figure 1. Table (1) shows the results of PSNR values of fused image with different noise densities. We used the image quality metric, peak signal-to- noise ratio (PSNR), to measure the quality of the restored image. The PSNR measure is defined as (255 )2 ๐‘ƒ๐‘†๐‘๐‘… = 10 log10 MSE (21) where MSE is the mean squared error between the original noise-free image and the restored image. SSIM MSE PSNR Fusion 0.99 50 29 VMF 0.93 200 25 Median Filter 0.92 220 22 Center Weighted Median Filter 0.91 650 20 Spatial Median Filter 0.90 700 18 Table 1: Performance Evaluation of Fusion Method with filter responses at 30% of noise intensity www.iosrjournals.org 33 | Page
  • 4. A New Method of Image Fusion Technique for Impulse Noise Removal in Digital Images using the Fig 1: Fused Image and responses of different kinds of filters a) Original image; b) Image with added Noise; c) Vector Median Filter; d) Center Weighted Median Filter; e) Spatial Median Filter Image; f) Fused Image VII. Conclusion This paper presents a new method of image fusion technique for removal of impulse noise in images. The images captured by sensors undergo filtering using VMF and SMF, and then the two individual de-noised images are fused to obtain a high quality image. The Quality of the images is evaluated using Structural Similarity Index (SSIM), Peak Signal-to-Noise ratio (PSNR) and Mean Square Error (MSE) with different noise densities. The proposed techniques are algorithmically simple and can be used for real time imaging applications. References [1] Nikolaos Mitianoudis, Tania Stathaki, Pixel-based and Region-based Image Fusion schemes using ICA bases, Communications and Signal Processing group, Preprint submitted to Elsevier Science, 13th Dec 2007. [2]. Reihard Berstein, โ€œAdaptive nonlinear filters for simultaneous removal of different kinds of noise in images,โ€ IEEE Trans on circuits and systems, Vol.cas-34, no11, pp.127-1291, Nivember1987. [3] A. Hyvร„arinen, P. O. Hoyer, and E. Oja. Image denoising by sparse code shrinkage. In S. Haykin and B. Kosko, editors, Intelligent Signal Processing. IEEE Press, 2001. [4] P. Hill, N. Canagarajah, and D. Bull. Image fusion using complex wavelets. In Proc.13th BritishMachineVisionConference, Cardiยฎ, UK, 2002. [5] James C. Church, Yixin Chen, and Stephen V. Rice , โ€•A Spatial Median Filter for Noise Removal in Digital Imagesโ€•, 2008 IEEE. [6] M. Vett, erli and C. Herley, โ€œWavelets and filter banks: theory and design,โ€ IEEE Tram Signal Processing, vol. 40, pp. 2207-2232, September 1992. [7] P. Burt and R. Lolczynski, โ€œEnhanced image capture through fusion,โ€ in Proc. the Fourtt International Conference on Computer Vision, pp. 173- 182, 1993. [8] Tao Chen, Kai-Kaung Ma and Li-Hui Chen, โ€•โ€Tri-state median filter for image Denoisingโ€, IEEE Transactions on Image Processing, Vol 8, no.12, pp.1834-1838, December 1999. [9] S.Indu, Chaveli Ramesh, โ€• โ€œA noise fading technique for images corrupted with impulse noiseโ€, Proceedings of ICCTA07,IEEE. www.iosrjournals.org 34 | Page