Table of ContentsSummaryIntroductionIi. Fcm (Fuzzy C Mean) segmentationiii. Dwt (Discrete Wavelet Transform)Iv. Median filterV. Svm (Support Vector Machine) classifier VI. Proposed algorithmVii. Simulation resultsViii. ConclusionSummaryA brain tumor can be detected using a computerized image processing algorithm. An MRI was done to detect the brain tumor. MRI images are not enough to thoroughly diagnose the tumor. The c-mean fuzzy algorithm is a very popular image segmentation. The output of the Fuzzy C Mean algorithm also contains unwanted parts. In the proposed work, these unwanted parts can be removed using a median filter. In the proposed work, DWTs with SVM are used to identify tumor types, whether benign or malignant. The image filtered by median filter also contributes to better detection by the SVM classifier. Say no to plagiarism. Get a tailor-made essay on “Why Violent Video Games Should Not Be Banned”? Get the original essayTagsFCM (Fuzzy C Mean), MRI (Magnetic Resonance Imaging), DWT (Discrete Wavelet Transform), SVM (Support Vector Machines), Image Segmentation, Grayscale Image, MRI (Support Vector Machines), magnetic resonance), computed tomography (CT), image preprocessing, image filtering.IntroductionBrain tumor can be detected by various brain exploration techniques. CT scan provides a detailed image of the brain and an MRI test where the computer is linked to a strong magnetic field which provides a clear 2D image of the brain. MRI (magnetic resonance imaging) eliminates radiation unlike CT [2,4]. The MRI image gives a complete view of the brain and proper inspection must be done by an expert to find the tumor, which makes the process slower and more expensive. To solve this problem, computerized segmentation algorithms were created. These algorithms provide the tumor as the output image. The most commonly used segmentation is the FCM (Fuzzy C Mean) segmentation algorithm. The FCM algorithm provides accurate results for overlapping datasets and it is much more efficient than the k-means algorithm [1]. Brain tumors can be classified as benign and malignant. A benign tumor is one that does not grow suddenly. It never affects neighboring tissues and does not spread to other parts at all. A malignant tumor worsens over time and ultimately proves fatal. We can say that malignancy is a tumor at a descriptive or advanced stage from which it is almost impossible to go back [4]. To extract features from an MRI brain image, the wavelet transform is effective because it allows image analysis at different motion levels adapted to its multi-resolution diagnostic property [1]. In order to differentiate the type of brain tumor, a Support Vector Machines (SVM) classifier is commonly used. The SVM model represents points in space that are mapped such that examples of distinct categories are divided by as wide a clear space as possible. In our proposed work, FCM algorithm is used for brain MRI image segmentation. The segmented image is further enhanced through the use of a median filter. Here, the median filter removes the unwanted segmented part considering it as noise. The segmentation result is then fed to the DWT and SVM classifier to correctly identify the tumor type.Ii. Fcm (Fuzzy C Mean) SegmentationFuzzy c-mean can be called as a sub-optimal segmentation method which abandons optimalityoverall for improved statistical performance and adaptability of the segmentation process. Computational evaluation on FCM is determined by the amount of image points that need to be highly processed in each iteration [5]. FCM is a clustering technique which allows information belonging to two or more clusters to be obtained [6]. The main aspect of this algorithm works by assigning membership values ​​to each data point resulting from each cluster center based on the distances between the cluster and the data point. The higher the membership value, the closer the data is to the center of the cluster. Clearly, the sum of membership at each data point must equal one [10]. The FCM algorithm is an iterative clustering method that produces an optimal partition c by minimizing the weight in the group sum of the squared error objective function (JFCM) [8].( 1) Where,X = {x1, x2, ..., xn} ≤ R,n = number of data elements,c = number of clusters with 2 ≤ c < n,uik = degree of belonging of xk to the ith cluster,q = weighting exponent on each fuzzy membership,vi = prototype of the center of cluster i,d2(xk,vi) is a measure of distance between the object xk and the center of cluster vi. An object function solution (JFCM) can be calculated by an iterative process, which is as follows: First set the values ​​for q, c and e, Second, the fuzzy partition matrix must be initialized, Third, you must set the loop counter such that b = 0, Calculate the cluster centers c { vi(b)} with U(b)(2) Calculate the membership U(b+1), For k = 1 to n, calculate the following:Ik={i|1<=i<=cdik=||xk -vi||=0},~Ik={1,2,……c}-Ik, for the kth column of the matrix, calculates the new membership values, and if Ik=Ø , then(3) otherwise uik(b +1) = 0 for all iє~Ik and ƩiєIk uik(b+1) =1, next k [9] ,if ||Ub-U(b+1)|| < Ɛ, stop; otherwise, set b=b+1 and go to step 4. For medical image segmentation, the appropriate clustering type is fuzzy clustering. Fuzzy c-means (FCM) can be considered as the fuzzified version of the k-means algorithm. It is a kind of clustering algorithm that allows data elements to have a degree of membership in each cluster by degree of membership [6].iii. Dwt (Discrete Wavelet Transform) The wavelet gives an idea of ​​the different frequencies in an image using different scales. DWT provides wavelet coefficient from brain MRI images. The two-dimensional DWT gives four sub-bands, which are LL (low-low), HL (high-low), LH (low-high), HH (high-high) with the two-level wavelet decomposition of the region d interest (ROI). The wavelet approximations at the first and second level are represented by LL1, LL2, respectively; which represents the low frequency part. The high frequency part of the images is represented by LH1, HL1, HH1, LH2, HL2 and HH2 which give the details of horizontal, vertical and diagonal directions at the first and second level, respectively, as shown in fig. 1 below [2].IV. Median FilterThe median filter is very popular in image filtering. It behaves as a low-pass filter that blocks all high-frequency components of images such as noise and edges, thus making the image blurry [11]. For filtering corrupted images at high density, a large window size is required so that a sufficient number of noise-free pixels are present in the window. Thus, the size of the sliding window in the median filter varies depending on the noise density. The window size median filters 3×3, 5×5, 7×7 and 9×9 are mainly applicable. The output of the median filter is given by y(i,j)=median{x(is,jt),x(i,j)/(s,t)∈W,(s,t)≠(0,0) } (4) where {x} is the noisy image and y(i,j) is the recovered image.