I.J. Image, Graphics and Signal Processing, 2014, 12, 65-69
Published Online November 2014 in MECS (s-press/)
DOI: 10.5815/ijigsp.2014.12.09
Performance Evaluation of Image Fusion Algorithms for Underwater Images-A study
bad on PCA and DWT
Ansar MK, Vimal Krishnan VR
Department of Computer Science, School of Mathematical and Physical Science
Central University of Kerala, Kasargod-671 314, India
(ansarmk07, vimallnair)@
Abstract—In this paper, a comparative study between two image fusion algorithm bad on PCA and DWT is carried out in underwater image domain. Underwater image fusion is emerged as one of the main image fusion area, here two or more images will be fud by retaining the most desirable characteristics of each underwater images. The DWT technique is ud to decompo the input image into four frequency sub bands and the low-low sub band images will be considered in fusion processing. In PCA method significant eigen values will be considered in fusion process to retain the important characteristics of the input images. The results acquired from both experiments are tabulated and compared by considering the statistical measures such as Peak Signal to Noi Ratio (PSNR), Mean Square Error (MSE) and Entropy. Results shows that underwater image fusion bad on DWT outperforms the PCA bad method. Index Terms—Image Fusion, Image Enhancement, PCA, DWT, MSE, PSNR.
I.I NTRODUCTION
友谊名言
Underwater image processing is emerged as one of the main rearch area of image processing. Especially it is widely ud in ocean exploration, defen, and fish detection [1]. However, the quality of the underwater images is reduced becau of the absorption and scattering effects of the underwater environment [2]. Also it may contain distortion and degradation in the form of noi, blur
etc. [3]. Rearchers come up with different techniques for improving the quality of underwater images. Image fusion is one such technique. This paper explain the performance evaluation of two algorithm bad on PCA and DWT.
The image fusion is a branch of data fusion and it is the process of combining two or more images to form a single image [4]. So the fud image gives much better information than the original images [5][15]. The Fusion process will reduce the volume of data by creating compatible images with perception capability of human operator by completing image processing tasks like: image gmentation, object detection or target recognition [5].
Image fusion is ud in the areas like defence [6], surveillance [7], target tracking [8], Medical Imaging [9][10], Biometrics [11], Robot vision, Aerial imaging and Satellite imaging [12][13][14] etc.
The Fusion process can be classified into three levels. They are pixel level image fusion, feature level image fusion; decision level image fusion [16][17]. The decision level and feature level fusions are high-level fusions that require more complex algorithms and more intensive computation. The pixel level fusion is the lowest level fusion that fus the images from different physical channels pixel by pixel to enhance the features not complete in either channel [18]. Therefore, it requires less
processing time and is found suitable for time critical image fusion applications such as underwater image processing specially for defence purpo.
The paper has been divided into five ctions. Section II describes the principal component analysis. Discrete wavelet transform bad fusion is discusd in Section III followed by experiment and comparative study in ction IV. Conclusions are summarized in ction V.
II.P RINCIPAL COMPONENT ANALYSIS
PCA is probably the most widespread multivariate statistical technique. Karl Pearson introduces it in 1901. Principal Component Analysis (PCA) is often ud to reduce multidimensional data ts to lower dimensions for analysis. It reveals the internal structure of data in an unbiad way [19]. The PCA image fusion method us the pixel values of all source images at each pixel location. Then adds a weight factor to each pixel value (it is known as standardized PCA). The average of the weighted pixel values will be ud to produce fud image [20]. The optimal weighted factors are determined by the PCA technique.
PCA is very uful for understanding the variability in underwater image data t. Sometimes especially in defen application underwater images may contain large amount of information. It can
be reduced by PCA without losing the information by compression. And also the PCA technique is uful for image encoding, image data compression and image enhancement [20].
PCA is implemented by using following mathematical procedure.
vps国外
Step1: Select two underwater images with same resolution.
Step2: Adjust the image matrix by subtracting the mean from both original image matrixes. The mean can be found using the formula
xi n
i =1n å
Step3: Calculate the covariance of the image matrix. It can be found by using the formula.
xi -x ()yi -y ()
i =1n
ån -1
Step4: Calculate the eigenvectors and eigenvalues from the covariance matrix. The eigenvector will give most important data‘s.
Step5: Form the feature vectors by ordering eigenvalues bad on their significance.
features vector
Step6: The fud image is formed by taking the transpo of the feature vector and multiplies it on the left of the original data t, transpod. That is
Im ageMatrix =RowFeatureVector *RowDataAdjust ()
III. D ESCRETE WAVELET TRANSFORM BASED FUSION
Wavelet is a famous technique ud for analyzing signals. It has the ability to prerve the time and frequency details of the images to be fud [22][23]. It provides a variety of channels reprenting the image feature by different frequency sub-bands. Li et al [22] and Chipman et al [24] introduced DWT into image fusion. The discrete 2-dimensional wavelet transform is computed by the recursive application of low pass and high pass filters in each direction of the input image followed by sub sampling [23][25]. The discrete wavelets transform (DWT) allows the image decomposition in different kinds of coefficients prerving the image information. When decomposition is performed, the approximation and detail component can be parated [15][16][26].
The DWT merges the coefficient to get the best result in the fud image. We can do it by considering the average of coefficient [19]. The average method and it is one of the basic methods to implement discrete wavelet fusion.
Here, two underwater images with same spatial resolution are ud. The decomposition is achieved by applying DWT on both images. Only the coefficients at
the same level and reprentation are fud. Final fud image is obtained by taking IDWT (Inver Discrete Wavelet transform).
The procedure given below shows different steps to perform DWT on underwater images.
努力作文600字Step1: Select two underwater images with same resolution.
Step2: Apply decomposition using DWT on both input images.
Step3: Fu each wavelet coefficient using average method.
Step4: perform IDWT to get the fud image.
IV. E XPERIMENT AND RESULT
In order to measure the performance of the PCA and DWT fusion techniques, two underwater images with same resolution are ud. The original images are in jpg format. The images of the scene1 and scene 2 are given in fig.1 and 2.
Fig 1. Underwater Image scene 1
Fig 2. Underwater Image scene 2
The performance measuring properties such as entropy, mean square error and peak signal to noi ratio shows the improvement in the fud image for both methods. The are the commonly ud statistical measures in asssing image fusion techniques. Mean Square Error and Peak Signal to Noi Ratio consider image as a special type of signal. Table 1 and 2 shows the measured values for both methods.
1eig ,2eig ,.........,n
eig我爱北京天门
(
)
A. Entropy
焖牛腩的做法大全
Entropy is a statistical measure of randomness. It can be ud to characterize the texture of the input image. Entropy is defined as
-sum(p´log2(p))(1) Where p contains the histogram counts returned from histogram of the image.
B. MSE (Mean Square Error)
The mean square error of an image can be finding out by using the following formulae.
MSE=
1
mn
I(i,j)-K(i,j)
[]2
j=0
n-1
å
i=0
m-1
å (2)
C. PSNR (Peak Signal to Noi Ratio)
The equation given below is ud to find the PSNR between input image and fud image.
PSNR=20log
10
MAX
I
()-10log10MSE
() (3)
The measures give only the global idea of the images. Also when asssing the performance of image fusion techniques using above measurements, we require the knowledge of both original image and fud image.
Fig 3. Histogram of Underwater image 1
Fig 4. Histogram of Underwater image 2
Fig 3 and 4 above shows the histogram of the underwater images of scene 1 and 2. The fud images of two scenes are given below.
Fig 5. PCA fud image
Fig 6. Wavelet fud image
PCA is a standard fusion technique bad on the spatial domain, so it has got lower processing spee
d becau of the prence of large amount of pixel level information. Where as in the ca of wavelet, fusion takes place in the transform domain by combining the wavelet coefficients. That speedup the process and also produce better fud image. The histogram of the both methods is given in the fig.7 and 8.
In the ca of under water images wavelet bad approach is very uful, becau we can fu the images with different resolution. But it is not possible in standard PCA. Decomposition and fusing of coefficient helps to collect the information appropriately in DWT. Higher
Fig 7. Histogram of PCA fud image
Fig 8. Histogram of wavelet-fud image
value of MSE value in PCA bad fusion indicates the perverance of spatial information. But it caus spectral degradation. That inverly affects the quality of the fud image. DWT out perform this problem by minimizing the spectral distortion. DWT produce higher PSNR value for fud image than PCA bad fusion. It shows the higher quality of fud image.
We have got a maximum of 17.8574 for PSNR while comparing figure 2 and fud image in DWT bad fusion. Where as in the ca of figure 1 and fud image it is 16.3618 only. But in the ca of PCA bad fusion it is about 9.2990 and 9.5698 for figure 1 and figure 2 while comparing with fud image.
Table 1. Performance measures of PCA
Table 2. Performance measures of DWT
V.C ONCLUSION
In this paper performance of the two fusion methods such as PCA and DWT were compared statistically in underwater domain. Image fusion is performed to create a single enhanced image more suitable for different application. PCA primarily works with spatial domain and it is very uful for image fusion, data classification and dimensionality reduction. It has been found that wavelet bad fusion techniques outperform the PCA fusion in spatial and spectral quality, especially in minimizing color distortion. Higher value of PSNR clearly shows it. So Wavelet bad fusion with higher level of decomposition showed better performance in underwater images. In order to get better spatial and spectral resolution it is recommended to u both PCA and Wavelet together.
英语口语顺口溜A CKNOWLEDGEMENT
The authors wish to thank Dr. Thasleema TM, Dr. Manjunath S, Dr. Abbas T.P, Mr. Kumar V, Mr. Sujith B, Ms. Daisy and Mr. Rajesh Kumar Assistant professors, Department of Computer Science, Central University of Kerala, India.
R EFERENCES
笔记本电脑风扇[1]Raimondo Schettini; Silvia Corchs, ―Underwater Image
Processing: State of the Art of Restoration and Image Enhancement Methods‖. EURASIP Journal on Advances in Signal Processing 2010, 2010:746052.
[2]Shuai Fang; Rong Deng; Yang Cao, ―Effective Single
Underwater Image Enhancement by Fusion‖. Journal of computers, Vol. 8, No. 4, April 2013.
[3]Ancuti; Haber; Bekaert, ―Enhancing underwater images
and videos by fusion‖, Computer Vision and Pattern Recognition -21 June 2012.
[4] A. Ardeshir Goshtasby; Stavri Nikolov,‖Image fusion:
洗衣液成分
Advances in the state of the art,‖ Volume 8, Issue 2, Pages 113-222 (April 2007).
[5]Petrică Turtoi; Niculae Guzulescu; Daniel Ţurcanu,
―Image Fusion Bad on Wavelet Transform‖, International Conference of Scientific Paper, AFASES 2013 Brasov, 23-25 May 2013.
[6]Toet A; Aguilare M; Waxman Am, ―Fusion of visible and
thermal imagery improves situation awareness‖, Volume 18, Issue 2, 30 December 1997, Pages 85–95.
[7]Snidaro, L; Luca Foresti, G; Niu, R; Varshney P.K,
― Sensor fusion
for video surveillance‖, Proceedings of the 7th International Conference on Information Fusion, Stockholm, Sweden, 2004, pp. 739–746.
[8]Sworder D, D; Boyd J.E; Clapp G.A, ―Image fusion for
tracking manoeuvring targets,‖ International Journal of Systems Science, Vol.28, Issue 1, 1997, pages 1–14. [9]Hill, D; Edwards,P; Hawkes, D; ―Fusing medical images,‖
Image Processing , Vol.6, Issue 2, 1994, pages 109–113.
[10]CYN Dwith; Vivek Angoth; Amarjot Singh, "Wavelet
Bad Image Fusion for Detection of Brain Tumor", IJIGSP, vol.5, no.1, pp.25-31, 2013.DOI:
10.5815/ijigsp.2013.01.04.
[11]Anouar Ben Khalifa; Najoua Essoukri BenAmara,
"Contribution to the Fusion of Biometric Modalities by the Choquet Integral", IJIGSP, vol.4, no.10, pp.1-7, 2012.
[12]Goshtasby A; ―2-D and 3-D Image Registration for
Medical, Remote Sensing, and In dustrial Applications,‖ Wiley Press, 2005.
[13]Pohl,C; Genderen J,L, ―Multinsor image fusion in
remote nsing: concepts, methods, and applications,‖ International Journal of Remote Sensing, Vol.19, Issue 5,1998, pages 823–854.
[14]Simone G; Farina A; Morabito F,C; Serpico S,B;
Bruzzone L, ―Image fusion techniques for remote nsing applications,‖ Information Fusion, Vol. 3,2002, pag es 3–
15.
[15]Varsha patil; Deepali sale; Joshi M.A, ―Image Fusion
Methods and Quality Asssment Parameters‖.
[16]Bedi S.S; Khandelwal Rati, ―Comprehensive and
Comparative Study of Image Fusion Techniques‖, International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Vol.3, Issue-1, March 2013. [17]Mrinal Kanti Bhowmik, Kankan Saha, Sharmistha
Majumder, ―Thermal Infrared Face Recognition–a Biometric Identification Technique for Robust Security System‖.
[18]Tumpa Dey, ―A Survey on Different Fusion Techniques
of Visual and Thermal Images for Human Face Recognition‖, International Journal of Electronics Communication and Computer Engineering Vol. 4, Issue 6, NCRTCST -2013, ISSN 2249–071X.
[19]Zhijun Wang, Djemel Ziou, Costas Armenakis, Deren Li,
and Qingq uan Li in ―A Comparative Analysis of Image Fusion Methods‖. IEEE transactions on geoscience and remote nsing, vol.43, no.6, June 2005.
[20]Lanir J, ―Comparing multispectral image fusion methods
for a target detection task,‖ Optical Engineering, vol. 46, pp. 66402-1, 2007.
[21]Nirosha Joshitha J; Medona Selin R,―Image Fusion using
PCA in Multifeature Bad Palmprint Recognition,‖ International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Vol.2, Issue-2, and May 2012.
[22]Hui Li; Manjunath, B.S.; Mitra, S.K, "Multi-nsor image
fusion using the wavelet transform," Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference, vol.1, pp.51, 55 vol.1, 13-16 Nov 1994. [23]Gonzalo Pajares; Jesus Manuel de la cruz, ―A wavelet
bad image fusion tutorial,‖ Pattern Recognition, Vol. 37, 2004, pages 1855–1872.
[24]Chipman L.J; Orr Y.M; Graham L.N, ―Wavelets and
image fusion,‖ Proceedings of the International
Conference on Image Processing, Washington, USA, 1995, pp. 248–251.
[25]Shefer r; Malhi M; Shenhar A, ―Wave distortion
correction using cross correlation,‖ Technion, Israel Institute of Technology.
[26]Amarjot Kaur; Sunil Khullar ―Image Fusio n using HIS,
PCA and Wavelet Technique,‖ International Journal of Computer Science and Communication Engineering Vol.
2 Issue 2 (May 201
3 Issue).
Ansar MK is a Master of Science
student with Specialization in
Computational Intelligence at Central
University of Kerala, India. He
completed Bachelors in Computer
Science from Kannur University Kerala.
He is the founder and chairman of
‗innov DREAMZ’, a potential rearch
group of students who works together to promote science rearch. His rearch interests include Image fusion, Computer Vision, Pattern Recognition, Data Mining etc.
Vimal Krishnan VR is Assistant
professor at Central University of Kerala,
India. He completed his Masters in
Software Science from Periyar
University and doing PhD in Speech
Processing at Kannur University. He is a
Free Software and Knowledge Freedom
Activist. His rearch interests include
Speech Recognition, Artificial Neural
Network, Wavelet, Signal Processing, and Emotion Recognition etc.
How
to
cite this paper: Ansar MK, Vimal Krishnan VR,"Performance Evaluation of Image Fusion Algorithms for
Underwater Images-A study bad on PCA and DWT", IJIGSP, vol.6, no.12, pp. 65-69, 2014.DOI: 10.5815/ijigsp.2014.12.09
