Wavelet Based Image Fusion for Detection of Brain Tumor

visibility

…

description

7 pages

link

1 file

Brain tumor, is one of the major causes for the increase in mortality among children and adults. Detecting the regions of brain is the major challenge in tumor detection. In the field of medical image processing, multi sensor images are widely being used as potential sources to detect brain tumor. In this paper, a wavelet based image fusion algorithm is applied on the Magnetic Resonance (MR) images and Computed Tomography (CT) images which are used as primary sources to extract the redundant and complementary information in order to enhance the tumor detection in the resultant fused image. The main features taken into account for detection of brain tumor are location of tumor and size of the tumor, which is further optimized through fusion of images using various wavelet transforms parameters. We discuss and enforce the principle of evaluating and comparing the performance of the algorithm applied to the images with respect to various wavelets type used for the wavelet analysis. The performance efficiency of the algorithm is evaluated on the basis of PSNR values. The obtained results are compared on the basis of PSNR with gradient vector field and big bang optimization. The algorithms are analyzed in terms of performance with respect to accuracy in estimation of tumor region and computational efficiency of the algorithms.

Figures (6)

The paper uses wavelet based image fusion technique for brain tumor detection. Wavelet analysis is widely used method for solving difficult problems in mathematics, physics and engineering in the modem times. It has diverse applications in the fields of wave propagation, data compression, signal processing, image processing, pattem recognition and medical imaging technologies. Wavelet analysis decomposes complex information to elementary forms at different positions and scales, which can be easily reconstructed with high precision. Wavelet transform is a powerful tool in the analysis of signals compared to Fourier transform, because the later method fails in the analysis of non stationary signals. The first recorded mention of what we now call a “wavelet” seems to be in 1909, in a thesis by Alfred Haar [10]. Wavelet transforms is a new area 0 technology, replacing the Fourier transform in various fields of application like image processing, heart-rate and ECG analyses, DNA analysis, protein analysis, climatology, general signal processing, speech recognition, computer graphics and multi fractal analysis [11]. In this paper, our main goal is to exploit the advantages of wavelet transform compared to most widely used Fourier transform, in case of image fusion algorithm for the detection of brain tumor. Image Fusion technique is effectively used to the detect tumor in various complex backgrounds. Image Fusion is applied by merging multiple images resulting into precise information about the size, shape and location of the tumor. In this paper, the MRI and CT image (Fig. 5(a) and (b)) are processed using wavelet analysis. Before applying image fusion, the source images must be pre- processed through various Image enhancement techniques. The Image enhancement techniques involve point operations, mask operations, and global operations which sharpen image features for efficient analysis. In the proposed algorithm mask operations were used for the enhancement of brain tumor detection efficiency. These techniques improve the efficiency of the image fusion algorithm through elimination of ambient elements from the source images. In this paper various results are obtained by changing the wavelet used for decomposition. The paper further discusses the variation in result on changing the parameters of wavelet transforms used for the decomposition.

The MR image fusion approach first proposed by Burt and Adelson [15] was implemented using pyramid structures called the Laplacian pyramid. The basic idea was to use multi scale transform (MST) on the source images followed by inverse multi scale transform (IMST) resulting into composite of images. In the first step multi scale transforms or multi resolution analysis involves decomposition of the original image into different levels. The image signal consists of various bands like low-low, low-high, high-low and high-high on decomposition by wavelet transform. Higher number of decomposition levels [16] does not necessarily produce better result because by increasing the analysis depth the neighboring features of lower band may overlap. This leads to discontinuities in the composite representation and thus introduces distortions, such as blocking effect or ringing artifacts into the fused image. Thus in the next step, selection of the appropriate decomposition level allows the combination of salient features of each image. The block diagram of the basic image fusion is shown in the Fig. 3. Fig. 3. Block diagram depicting basic image fusion using multi resolution analysis. The principle of image fusion [14] using wavelets was first introduced by H. Li. in which image fusion was used for the fusion of synthetic aperture radar and multispectral image data. Image fusion can be defined as combining various signal sources from different sensors and multi-modality images. Simple image fusion can be achieved through taking pixel by pixel average of the source images. This process may lead to undesired side effects or distortions in the fused image. In order to avoid unnecessary distortions, the principle of pyramid transform is used to construct the pyramid transform o the source images. Pyramid transforms of the source images are combined and on applying inverse pyramid transform fused image is obtained. The basic idea used is to form multi scale transforms on the input images, and to form a multi scale composite representation from these and form the required image by applying inverse transforms. In case of Wavelet based image fusion, Wavelet decomposition is applied to the original images by passing the input images through a wavelet filter which gives the Approximation coefficients and Detail coefficients of the images. The Approximation and Detail coefficients of the two images are combined using average of the coefficients or the maximum or minimum of the coefficients. The resultant image is

Fig. 4 Flow Chart of the Proposed Algorithm image. This facilitates better detection of the tumor with minimum error. The output of the proposed image fusion algorithm is dependent on three maisan parameters: (a) the wavelet used for the image decomposition i.e. Haar(haar), Daubechies(db), Coiflets(coif ) or symlets(sym) (b) the approximation or detail coefficients obtained can be fused with different approaches, which involve taking the maximum, the minimum orthe mean of the coefficients from the two input images (c) the total number of decompositions of the image using the wavelets analysis. As number of decompositions increase we may lose some minute details of image in the image as these small regions can be of significant importance in the medical analysis. Hence, the first level decomposition is used in the proposed algorithm. The results are analyzed by varying the other two parameters in order to obtain optimum results.

Fig. 6. The Segmentation results after the algorithm is applied on the input images using coif2 wavelet.

Fig. 7. Variation of PSNR when the algorithm is applied on different wavelets at 4th decomposition

Fig. 5. Input MR Image and CT scan Image.