Multimodality Image Fusion of CT and MRI Images using Discrete Wavelet Transform (DWT)

DOI : 10.17577/IJERTCONV6IS08035

Download Full-Text PDF Cite this Publication

Text Only Version

Multimodality Image Fusion of CT and MRI Images using Discrete Wavelet Transform (DWT)

P Poornima (Guide)

Electronics and Communication Engineering,

V.S.B. Engineering College, Karur

S Govendhan (Author) Electronics and Communication Engineering,

V.S.B. Engineering College, Karur

S Crane (Author)

Electronics and Communication Engineering,

V.S.B. Engineering College, Karur.

M Balaaji (Author)

Electronics and Communication Engineering,

V.S.B. Engineering College, Karur

Abstract:- The objective is fusion of the image can be combine multiple images of the same scene into a single image obtain the important and required features from each of the original image. Medical image fusion is used to derive useful information from multimodality medical images which provides more information to the doctor. Nowadays, with the rapid development in high technology and modern instrumentation, medical imaging has become a vital component of a large number of applications, including diagnosis, research and treatment. Medical image fusion is the idea to improve the image content by fusing images taken from different imaging tools like computed tomography(CT), magnetic resonance imaging(MRI) for medical diagnosis , computed tomography(CT) provides the best information on denser tissue with less distortion. MRI provides better information on soft tissue with more distortion. In this case, only one kind of image may not be sufficient to provide accurate clinical requirements for the physician. Therefore, the fusion of the multimodal medical images is necessary. This paper presents a method of image fusion based on discrete wavelet transform. Two dimensional DWT is used to decompose the image. The fusion performance is evaluated on the passage of the root mean square error (RMSE) and peak signal to noise ratio (PSNR) and mean square error (MSE).

Keywords:- Medical image fusion, Computer tomography, magnetic resonance image, spatial filter, root mean square error(RMSE) and peak signal to noise ratio(PSNR), discrete wavelet transform (DWT), inverse discrete wavelet transform (IDWT)

I INTRODUCTION:

The term fusion means in general an approach to extraction of information acquired in several domains. The objective image fusion is to combine information from multiple images of the same scene into a single image retain the important and required feature from each of the original image. The main task of image fusion integrating complementary information from multiple images into

good easily confirm the position of illness. Medical imaging provides a variety of modes of image information for clinical diagnosis such as CT, X-ray, MRI,PET and etc. different medical images have different characteristics , which can provides structural information of different organs. For example CT (computer tomography) and MRI (Magnetic resonance imaging) with high spatial filter resolution can provide anatomical structure information of organs. Thus, a variety of imaging for the same organ, they are contradictory complementary and interconnected

In this paper, a novel approach for the fusion of computer tomography (CT), magnetic resonance imaging (MRI) images faced on wavelet transform has been presented. Different fusion rules are then performed on the wavelet coefficients of low and high frequency portions. The registered computer tomography (CT) and magnetic resonance imaging (MRI) images of the same people and same spatial parts have been used for analysis. Registration is a fundamental task in image processing. It is the process of spatially aligning two or more images of a scene. The processing brings into correspondence individual pixels in the images. Therefore, given a point in one image, the registration processing will determine the positions of the same point in other image. In this paper, input images taken are registered images.

II.METHODOLOGY

Image fusions based on wavelet transform

Wavelet transform

Wavelet means small waves so wavelet analysis is about analyzing signal with short duration finite energy function.

They transform the signal under in investigation into another representation which presents the signal in a more useful form.mathematically,we donate the DWT,

single image. The resultant fused will be more informative

and complete then any of input images and is more suitable

,

() = 1

||

(( )/)

for human visual and machine perception. Medical image fusion is the technology that good compound to mutual images into one according to certain rules to achieve clear visual effect. By observing medical fusion image, doctor

Where

b=is location parameter a=is scaling parameter

It is performed with the two medical images are mathematically represented as,

(, ) = 1(((1(, ), (2(, ))

  1. Pixel Averaging

    This is one of the simplest image fusion methods. In this method, average of the two input images is taken. However, when this direct method is applied, contrast of the image is reduced.

  2. Fusion with Discrete Wavelet Transform

    The basic idea of image fusion based on wavelet transform is that the coefficients of both the low frequency band and high frequency bands are performed with a certain fusion rule as shown in Fig 1. The widely used fusion rule is maximum selection scheme. This simple scheme just picks the coefficients in each sub band with the largest magnitude. After that, the fused image is obtained by perform the inverse DWT (IDWT) for the corresponding combined wavelet coefficients. The general fusion procedure based on wavelet transform can be described as follows.

    Step 1:

  3. Proposed Fusion Method

    As shown in the fusion block Fig 1, it is easy to find that the key step in image fusion based on wavelet is that of coefficient combination, namely, the fusion rules because it will decide how to merge the coefficients in an appropriate way so that a high-quality fused image can be obtained. Therefore, for this kind of image fusion method the key issue is its fusion rule design, and it should be paid more attention.

    Read the set of multifocal in images. Two images are same size and registered images

    Step 2:

    Apply decomposition by wavelet on both the image with the fused spatial filter.

    Step3:

    i) CT image ii) MRI image Fig2: Two Pair of Multimodality input medical images.

    Extracts from the wavelet decomposition structure of horizontal, vertical or diagonal in detail.

    Step 4:

    Perform the average of approximation coefficient of both the images

    Step 5:

    Compare horizontal, vertical and diagonal Coefficient of both the images. Apply maximum selection scheme to select the maximum coefficient value by

    1. Average Method

      Fig 3: Fused image

      comparing the coefficient of two images. Perform this for all the pixel values of images i.e.) m×n.

      Step 6:

      Apply inverse wavelet decomposition on both the images using different wavelet filters.

      Step7;

      Display the final fused images of averaging, maximum, minimum of the images.

      In this method the resultant fused image is obtained by taking the average intensity of corresponding pixels from both the input image.

      (, ) = (((, ) + (, ))/2

      Where

      A (x, y), B (x, y) are input image F (x, y) is fused image.

      Point (x, y) is the pixel value. For weighted average method-

      (, ) = ((, ) + (1 )(, )

      =0 =0

      Where W is wight factor and point (x, y) is the pixel value

    2. Select Maximum

      In this method, the resultant fused image is obtained by selecting the maximum intensity of corresponding pixels from both the input image

      Where

      (, ) = ((, ) + (, ))

      =0 =0

      Fig 1: Image fusion procedure based on DWT

      A (x, y), B (x, y) are input image F (x, y) is fused image, and point

      (x, y) is the pixel value.

    3. Select Minimum

In this method the resultant fused image is obtained by selecting the minimum intensity of corresponding pixels from both the input images

Entropy

It is measure of information content in an image. The maximum value of entropy can be produced when each gray level of whole range has the same frequency. If entropy of fused of images is higher than parent image then it indicates that the fused images contain more in information.

(, ) = ((, ) + (, ))

=0 =0

Where

A (x, y), B (x, y) are input image

Mean

=

=1

log

F (x, y) is fused image, and Point (x, y) is the pixel value.

If measures the mean value of the pixel in the image.

1

EXPERIMENTAL RESULTS:

Where

=

×

(, )

=1 =1

f(x, y)= denotes the gray level of a pixel with coordinate. (x, y)= The mean value represents the average intensity of an image.

Standard Deviation

The standard deviation of an image with size of m

× n is defined as

1

Fig: Averaging fusion image

Where

=

×

(, )

=1 =1

µ is the mean value of the image.

The standard deviation is the most common measure of statistical dispersion, which can be used to evaluate how widely spread the gray values in an image. So, the larger the standard deviation, the better the result.

RMSE (Root mean square error):

This is evaluate the fused image. It presents the amount of deviation present in the fused image compared to reference image.

Fig: Maximum fusion image

[(, ) (, )]2

=

=1 =1

R=input image F=fused image

×

Fig: Minimum fusion image

(x, y)=spatial position of pixels on m and n dimension of the image.

MSE (Mean square error)

It is one of the frequently measures the capture the deviation between the original and fused image. It is computed by finding the squared error divided by total number of pixels in the image. The spatial distortion introduced by the fusion process of are measured by MSE

III MEASUREMENT OF PERFORMANCE IN IMAGE

1

FUSION

In this section, the result of the algorithm is displayed. The

=

[(, ) (, )]2

=1 =1

algorithm has been implemented using MATLAB 7.0.4.The proposed method has been applied to CT and MRI images and compared with pixel averaging and DWT with MS rule method.

Peak signal to noise ratio: It is ratio between the maximum the value of an image and then magnitude of backround noise. It is commonly measured the qualified of

reconstruction in image fusion. The higher value PSNR the better is the fused image is

  1. REFERENCES

    PSNR = log10

    (2 1)2

    1. H.Hariharan, A.Koschan and M.Abidi, Multifocus image fusion by establishing focal connectivity, Proceedings of IEEE

      Computer Society, Conference on Computer and Pattern

      TABLE I: QUANTITATIVE RESULTS OF THREE DIFFERENT FUSION TECHNIQUES FOR CT & MRI IMAGES.

      Fusion Methods

      Mean Value

      Standard Deviation

      PSNR

      Pixel

      Averaging

      31.7994

      32.9336

      17.6900

      DWT

      32.0819

      34.9640

      30.5026

      Proposed

      Method

      59.7249

      61.5909

      41.4506

      TABLE II: QUANTITATIVE RESULTS OF THREE DIFFERENT FUSION TECHNIQUES FOR MRI IMAGES

      Fusion

      Methods

      Mean

      Value

      Standard

      Deviation

      PSNR

      Pixel

      Averaging

      39.6132

      47.7422

      29.6285

      DWT

      45.4690

      62.0304

      58.0040

      Proposed

      Method

      49.8099

      65.6692

      59.0000

      From Table I and II, the proposed method is better than the existing methods

      CONCLUSION

      This paper presents a new wavelet based approach for fusion of CT and MRI images. Firstly, the images to be fused are decomposed by wavelet transform. Then, by considering the characteristics of the two images, the coefficients of low frequency and coefficients of high frequency are selected by different fusion rule. Finally, the fused image is constructed with the inverse wavelet transform by the combined coefficients from the low frequency and high frequency bands. Experimental results on CT and MRI images indicate that the proposed method is better than existing methods.

      Recognition (ICIP-2007) vol 1-4244-1437.

    2. Shutao Li, James T Kwok,YaonanWang ,Combination of images with diverse focuses using spatial frequency,Science Direct, Information Fusion ,vol2,page no:169-176, May-2001.

    3. Yufeng Zeheng, Edward A Essock,Bruce Hansen, Andrew M Haun, A new metric based on spatial frequency and its application to dwt based fusion algorithm Science Direct: Information fusion vol:8 page no:177-192, 2007.

    4. J. Yang, Y. Ma, W. Yao, W. T. Lu ,A spatial domain and frequency domain Integrated approach to fusion of multifocus images.The International Archieves Of ThePhotogrammetry, Remote Sensing and Spatial Iinformation Sciences VolXXXVII. Part B7,Benjing.

    5. Hui Li, B.S Manjunath, Sanjit .K.Mitra, Multi sensor image fusion using wavelet transform. IEEE 0-8186-6950-1994

    6. Wei-wei wang,Peng-lang shui,Guo-xiang song,Multifocus image fusion in wavelet domain,Proceedings of the second International Conference on Machine Learning and Cybernetics,Xian,2-5,November 2003.

    7. Z. Zhang and R. S. Blum, A categorization of multiscale decomposition- based image fusion schemes with a performance study for a digital camera application Proceedings of the IEEE, vol. 87, no. 8, pp. 1315-1326, 1999.

    8. Yong Yang, Dong Sun Park, Shuying Huang and Nini Rao, Medical Image Fusion via an Effective Wavelet-Based Approach, EURASIP journal on Advances in Signal Processing, Volume 2010.

    9. S.Jayaraman, S.Esakkirajan, T.Veerakumar, Digital Image Processing,TMH, New Delhi,2009.

    10. A. Soma Sekhar, Dr.M.N.Giri Prasad, A Novel Approach of Image Fusion on MR and CT Images Using Wavelet Transforms IEEE 2011

    11. Sruthy S, Dr. Latha Parameswaran Ajeesh P Sasi, Image Fusion Technique using DT-CWT IEEE 978-1-4673-5090- 7/13/2016

Leave a Reply