You are on page 1of 7

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

Restoration of MRI Images with various types of techniques and Compared with Wavelet transform
Amanpreet Kaur* Hitesh Sharma**

ABSTRACT
In this paper, Wavelet is implemented on various noise images for the removing various types of noise from the MRI Images. Also discussed about the various types of other techniques of Image restoration like median filter, Weiner filters etc and find out the results. To validate the effectiveness of the wavelet restoration technique, both conventional and wavelet restoration methods will be applied to images and performance is compared based on image quality parameters like contour plots, MSE(Mean Square Error),PSNR(Point Signal to Noise Ratio).with the help of these parameters to find the best results between the techniques. Keywords: Wavelet transform, conventional techniques, median, weiner filtering, adaptive filtering.

I. INTRODUCTION
Image processing deals with analysis of various images by using different types of techniques. Image processing means to deals with various actions to change an image .In other words, Image is defined as two dimensional function such as f(x,y) where x and y are plan coordinates. When x,y and the amplitude values of f are all finite, discrete quantities, we call the image a digital

image. [1]A digital image is composed of a finite number of elements, each elements has a particular location and value. These elements are called to as picture elements or pixels. Digital Image Processing has numerous of applications in various studies and researches of science and technology. Some of fields that use Digital Image Processing include finger print, medical fields, photography, films and games industriesDigital image processing (DIP) is a part of signal processing where processing of digital images using various types of computer algorithm. This algorithm can be modified so that we can also change the appearance (color, size) of the digital image easily and quickly. Magnetic Resonance Imaging is a powerful clinic diagnosis technique.MRI is a test that uses magnetic fields and pulses of radio wave energy to make pictures of organs and structures include the body. In many of cases MRI gives different types of information about structures in the body that can be seen with X-rays, Ultra sound or computed tomography (CT) scan.MRI also may show problems that cannot be seen with other imaging methods. For the MRI test, the area of the body being studies is placed inside a special machine that contains large magnet. Picture from MRI scan are digital images that can be saved and stored on a computer for more study. The images also can be reviewed remotely, such as in clinic or an operating room. An MRI scan be done for head, chest, Blood vessel, Bones and joints.

*Department of Computer Science, Baba Farid College, Bathinda, Punjab (India) **Department of Computer Science, Lovely Professional University, Jalandhar, Punjab (India)

www.cpmr.org.in

CPMR-IJT: International Journal of Technology

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

As the field of MRI continues to grow, so do the opportunities in MRI.Magnetic Resonance Imaging is a powerful clinic diagnosis technique. As the field of MRI continues to grow, so do the opportunities in MRI. However, MR images acquired with fast scanning or desired for high temporal resolution often exhibit poor signal-to-noise ratio(SNR), which may lead to misinterpretation and bring difficulties to subsequent diagnosis and cure. The Indian health care industry is one of the biggest industries in the world[9]. Where, the telemedicine is the method, by which the patients can be examined, investigated, monitored and treated, with the patient and the doctor physically located in different places. A major goal of telemedicine is to eliminate unnecessary travelling of the patients and their escorts. Medical image acquisition, storage, display and processing and image transfer forms the telemedicine. This acquisition, storage requires the enormous amount of data. Digital Image Processing has fundamental classes which are grouped depending on various types of operations[5]:

1.4 Image Compression


Image Compression deals with the compression of the size of the image. so with the help of compression image can easily to be stored electronically. The compressed images are then decomposessed to their original forms. Images are the decomposed to their original forms. Image compression and decompression can either lose their size by maintaining high quality or preserve the original image without losing size.

1.5 Image Synthesis


In this class digital image processing is well known now days in the film and game industry is advanced in 2 dimensional and 3 dimensional productions. In these cases the images and videos scenes are constructed using certain techniques. Image restoration is one of the important techniques in the image processing field. Image restoration is an area that deals with improving the appearance of an image. The ultimate goal of restoration technique is to improve an image in some predefined sense. The purpose of image restoration is store the image without the noise. Degradation model is that defines the imfocus camera, motion, dust particles etc. The image degradation process can be modeled by the following equation [2]: g(x,y)=H(x,y).f(x,y)+n(x,y) Where, H(x,y)degradation function represents as camera defocus.For example, motion blur, imperfections of the lenses all can be modeled by H. The values g(x,y), f(x,y), and K(x,y) represent the observed values or degraded image, the original image or input image and the additive noise respectively.

1.1 Image Enhancement


Image Enhancement is among the simplest and most appealing areas of Digital image Processing. Image Enhancement deals with constract enhancement, spatial filtering, edge enhancement and noise reduction.

1.2 Image Restoration


It deals with improving the appearance of an image. The Image is corrected using different with improving the appearance of an image. The Image is corrected using different correction methods like Median filtering, Linear Filtering, Adaptive Filtering etc in order to restore an image to its original forms.

II. IMAGE ENHANCEMENT


Image Enhancement is among the simplest and most appealing areas of digital image processing. Basically, the idea behind enhancement techniques is to bring out detail that is obscured or simply to highlight certain features of interest in an image. A familiar example of enhancement is when we increase the contrast of an image because it looks better. It is important to keep in mind that enhancement is a very subjective area of
CPMR-IJT: International Journal of Technology

1.3 Image Analysis


Image Analysis deals with statistical details of an image. Here it is possible to examine the information of an image in details. This information helps in image restoration and enhancement. One of the representations of information is the Image Histogram to show brightness and darkness.
www.cpmr.org.in

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

image processing. Wavelets are the foundation for representation images in various degrees of resolution. Fourier transform has been the mainstay of transformbased image processing but now a more recent transformation called wavelet transformation is making it even easier to compress, transmit, restore, and analyze many images. Unlike the Fourier transform, whose basis function are sinusoids, wavelet transform are based on the small waves, called wavelets, of varying frequency and limited duration. Multiresolution theory is the main driving forces behind a lot of interest in the wavelet transform. Multiresolution theory incorporates and unifies variety of techniques including subband coding and pyramid image processing. A wavelet is a waveform of effectively limited duration that has an average value of zero. where sinusoids are smooth and predictable, wavelets tend to be irregular and asymmetric.Fourier analysis consists of breaking up a signal into sine waves of various frequencies. Similarly, wavelet analysis is the breaking up of a signal into shifted and scaled versions of the original (or mother) wavelet.signals with sharp changes might be better analyzed with an irregular wavelet than with a smooth sinusoid.

j in four components: the approximation at level j + 1, and the details in three orientations like horizontal, vertical and diagonal. [6]

3.1 Median Filtering


As the name clear the median filter is stastistics method[1].In this method we find the median of the pixel the replace the pixel by median of the gray levels in their neighbourhood of that pixels: f^ (x,y)=median{g(s,t)} The median filter is used to remove the noise like sait and pepper.

3.2 Adaptive Filtering


A adaptive filter that uses the gray and color space for removal impulsive noise in images. All processing is based on the gray and color space. This can provide the best noise suppression results and better preserve thin lines, edges and image details and yield better image quality compared to other filters.

3.3 Weiner Filtering


Wiener filter [2] incorporates both the degradation function and statistical characteristics of noise into the restoration process. The method is founded on considering images and noise as random processes, and the objective is to find an estimate f^ of the uncorrupted image f such that the mean square error between them is minimized. This error measure is given by e2 = E {(f- f^) 2} Where E {.} is the expected value of the argument. It is assumed that the noise and the images are uncorrelated; that one or the other has zero mean; and that the gray levels in the estimate are a linear function of the levels in the degraded image.

III. PROBLEM FORMULATION


In this paper,we have dicussed about the various tyes of images restoration techniques like Median filter,adaptive Filter,CLAHE,Wineier Filter and Wavelet Transform.So, these are used to remove the noise from the image.With the help of these techniques we will restored the MRI images and find the best techniques between these techiniques.The result find with the help of image restoration parameters like Contour Plots,MSE,PSNR.These parameters are implemented on noise MRI Images. We have used 2-D discrete wavelet transformation to decompose image into approximations and details. The approximations are the high-scale, low-frequency components of the signal. The details are the low-scale, high-frequency components. Since we have used 2-dimensional discrete wavelet transformation, this kind of two-dimensional DWT leads to a decomposition of approximation coefficients at level
www.cpmr.org.in

3.4 Wavelet Transform


The concept of wavelet was hidden in the works of mathematicians. During the year 1873, Karl Weirstrass mathematically described wavelet how a family of functions can be constructed by superimposing scaled versions of a given basis function. The definition of
CPMR-IJT: International Journal of Technology

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

Wavelet means a small wave. The smallness refers to the condition that the window function is of finite length. A wave is an oscillating function of time or space and is periodic. In contrast, wavelets are localized waves. They have their energy concentrated in time and are suited to analysis of transient signals. While Fourier Transform and STFT use waves to analyze signals, the Wavelet Transform uses wavelets of finite energy.The wavelet transform has become a useful computational tool for a variety of signal and image processing applications. For example, the wavelet transform is useful for the compression of digital image _les; smaller _les are important for storing images using less memory and for transmitting images faster and more reliably. NASAs Mars Rovers used wavelet transforms for compressing images acquired by their 18 cameras. [2]The waveletbased algorithm implemented in software onboard the Mars Rovers is designed to meet the special requirements of deep-space communication. In addition, JPEG2K is based on wavelet transforms. Wavelet transforms are also useful for `cleaning signals and images (reducing unwanted noise and blurring). Some algorithms for processing astronomical images, for example, are based on wavelet and wavelet-like transforms. The wavelet transform is a method for complete frequency localization of a time-varying, aperiodic signal. The Fourier transform provides information on the frequency components of the signal. But the Fourier transform does not provide any information about frequency localization, that is, it does not provide information about a specific frequency occurring in the signal. For improving the localization of the Fourier transform, a short-term Fourier transform can be used. In the short-term Fourier transform method, the whole signal is split into small windows and the Fourier transform is separately computed over each windowed signal. Although the short-term Fourier transform provides some localization information depending on the size of the window, it does not provide whole time-frequency localization.[3] For application in image processing and analysis, frequency localization will need to determine the spatial position of specific
www.cpmr.org.in

frequency component in the image. The wavelet transform provides a series expansion of a signal using a set of orthonormal basis functions that are generated by scaling and translation.

3.5 Types of Wavelet Transforms


There are mainly two types of Wavelet Transforms Continuous Wavelet Transformation (CWT) Discrete Wavelet Transformation (DWT) Since our algorithm is to be based on discrete wavelet transform, so we will discuss about the concepts of DWT .Two commonly used two types of techniques in Wavelet i.e DWT and IDWT DWT stands for Discrete Wavelet Transformation. It is the Transformation of sampled data, e.g. transformation of values in an array, into wavelet coefficients. IDWT is Inverse Discrete Wavelet Transformation: It means the procedure converts wavelet coefficients into the original data.

IV. EXPERIMENTATION AND RESULTS


These following figures restored with the help of various filters and results are calculated with the help of contour plots,PSR,PSNR.

Figure 1: Original Image

CPMR-IJT: International Journal of Technology

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

Figure 2: Median Filter Figure 6: Wiener Filter

Figure 3: Contour Plot of Image Figure 7: Wavelet Transform

Figure 4: Adaptive Filter Figure 8: Contour Plot of Image

4.1 Contour Plots


A contour plot is a path in an image which the image intensity value equal to a constant. It automatically sets up the axes and aspect ratio. Better restored image will have less congested contour plot and display each featured intensity value. If contour plot of the image is less dense (congested) and it is featuring each detail, then we can say that image is of the better quality. By restored images more regular edges I will use above feature of contour plots to evaluate the performance of
CPMR-IJT: International Journal of Technology

Figure 5: Contour Plot of Image


www.cpmr.org.in

10

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

mentioned restoration techniques.In the above diagrams of contour plots of image is shown.

4.2 MSE (Mean Square Error)


Mean Square Error is used to measure the pixel difference based measures. In the image coding and computer vision literature, the most frequently used measures and deviation between the original and noise image[5]. The reasons for these metrics widespread popularity are their mathematical tractability and the facts that it is often straight forward to design systems that minimize the MSE. Table 1 Shows MSE
S.No. Name of Image 1. 2. BRAIN MRI MRI BRAIN Median Adaptive 42.63 65.5 84.09 276.35 Weiner 21604.9 26122.32 Wavelet 0.00 75.92

standard Fourier transform is only localized in frequency. The Short-time Fourier transform (STFT) is also time and frequency localized but there are issues with the frequency time resolution and wavelets often give a better signal representation using Multiresolution analysis. The discrete wavelet transform is also less computationally complex, taking O(N) time as compared to O(N log N) for the fast Fourier transform. This computational advantage is not inherent to the transform, but reflects the choice of a logarithmic division of frequency, in contrast to the equally spaced frequency divisions of the FFT. Fourier analysis consists of breaking up a signal into sine waves of various frequencies. Similarly, wavelet analysis is the breaking up of a signal into shifted and scaled versions of the original (or mother) wavelet. Signals with sharp changes might be better analyzed with an irregular wavelet than with a smooth sinusoid.

4.3 PSNR (Peak Signal to Noise ratio)


Large PSNR indicates a smaller difference between the original (without noise) and reconstructed image.This is the mostly used objective image quality measure[5]. The main advantage of this measure is ease of computation but it is does not reflect perceptual quality. An important property of PSNR is that a slight spatial shift of an image can cause a large numerical distortion but no visual distortion and conversely a small average distortion can result in a damaging visual artifact,if all the error is concentration in a small important region. This metric neglects global and composite errors. Table 2 Shows PSNR
S.No. Name of Image 1. 2. BRAIN MRI MRI BRAIN Median Adaptive 31.87 30.92 28.92 23.75 Weiner 4.82 3.99 Wavelet 314.86 29.36

VI. CONCLUSION
In MRI images, noise particles are a particularly delicate and very difficult task. A tradeoff between noise reduction and the preservation of actual image features (without noise) has to be made in a way that enhances the diagnostically relevant image content. The idea is to transform the data (various form) into the wavelet basis, in which the large coefficients are mainly the signal and the smaller ones represent the noise. By suitably modifying these coefficients, the noise can be removed from the data. An important application domain of the wavelet theory is restoration the data. In this thesis work, use of wavelet transforms for the restoration of MRI images.In this paper, wavelet tranform techniques has been implemented and compared with Various conventional techniques like Median Filter,Wenier Filter,Adaptive Filter by using Various Parameters like Contour plots,MSE,PSNR as performance evaluation criteria: From the evaluation results, wavelet based image restoration technique is the best among these various techniques. This implementation of wavelet restoration technique has been done by using 2- dimensional discrete wavelet transform, with the single level decomposition. This
CPMR-IJT: International Journal of Technology

V. COMPARISONS WITH FOURIER TRANSFORM


The wavelet transform is often compared with the Fourier transform, in which signals are represented as a sum of sinusoids. The main difference is that wavelets are localized in both time and frequency whereas the
www.cpmr.org.in

11

ISSN: 2277-4629 (Online) | ISSN: 2250-1827 (Print)

CPMR-IJT Vol. 2, No.1, June 2012

can be done using Wavelet packet transform, with the multilevel decomposition. Which can further improve results i.e. better restoration of the image. Here wavelet restoration is implemented for MRI images. We can implement wavelet restoration technique for the restoration for all other kind of images (art, engineering, medical etc).

[4] Juang Puu-An,Ming-Ni Wu MRI Brain Phase Image Unwrapping Process by Wavelet Zero-Padding Technique 2007 pp. 229-232 [5] MATLAB 7.8.0 Documentation. (R2009b), Help

VII. REFERENCES
[1] Gonzalez,R.C., and Woods, R.E. Digital Image processing , Prentice Hall of India cliffs,2002 [2] Ivan W. Selesnick, Wavelet Transforms A Quick Study, Sep 2007. [3] Jeba Akewak, Digital image processing and image restoration,May 2011

[6] Poobal Sumathi,Ravindrang G, the ]performance of fractal Image compression on Different imaging modalitiesUsing objective qualityJan 2011. [7] Prinosil Jiri, Smekal Zdenek, Bartusek Karel, Wavelet Thresholding Techniques in MRI Domain,March 2010, pp 58-63 [8] Sifuzzaman M, Islam M.R. and Ali M.Z, Application of Wavelet Transform and its Advantages Compared to Fourier Transform,2009

www.cpmr.org.in

CPMR-IJT: International Journal of Technology

12

You might also like