Brain Tumor detection of MRI images using Segmentation techniques, SVM and Edge detection Aakriti Singla1, Nikita Bajaj2, Nidhi Pandey3, Rashmi Saini4, Suneet Gupta51, 2,3,4,5 CSE Department, School of Engineering and Technology, MUST, Lakshmangarh [email protected], [email protected], [email protected], [email protected] detection and removal is one medical issue that still remains challenging in the field of biomedicine. Early imaging techniques such as pneumoencephalography and cerebral angiography had the drawback of being invasive and hence the CT and MRI imaging techniques help the surgeons in providing a better vision.
In this paper, we will focus on segmentation of Magnetic Resonance brain Images (MRI). Our idea is to consider this problem as a classification problem where the aim is to distinguish between normal and abnormal pixels on the basis of several features, namely intensities and texture. More precisely, we propose to use Support Vector Machine (SVM) which is within popular and well motivating classification methods followed by Edge detection technique of the image for more clarity.
The experimental study will be carried on different tumor shapes, locations, sizes and image intensities.Keywords: Support Vector Machine, Segmentation method, Enhancement techniques, Watershed algorithm, MRI images, Edge detectionINTRODUCTIONDigital Image Processing is the use of computer algorithms to perform image processing on digital images. As a subcategory or field of digital signal processing it has the capability to focus particularly on images apart from digital signals and systems. Filtering is also include in digital image processing which is helps to blur or sharpen the image. Digital image processing deals with manipulation of digital images through a digital computer. DIP focuses on developing a computer system that is able to perform processing on an image. The input of that system is a digital image and the system process that image using efficient algorithms, and gives an image as an output. The most common example is Adobe Photoshop. It is one of the widely used application for processing digital images.A brain tumor is a collection, or mass, of abnormal cells in your brain. Your skull, which encloses your brain, is very rigid. Any growth inside such a restricted space can cause problems. Brain tumors can be cancerous (malignant) or noncancerous (benign). When benign or malignant tumors grow, they can cause the pressure inside your skull to increase. This can cause brain damage, and it can be life-threatening . Image Processing is a technique useful in enhancing raw images received from cameras/sensors placed for various applications including enhancing images obtained from unmanned spacecrafts, space probes and military reconnaissance flights . LITERATURE REVIEWImages with contrast enhancement have been analyzed. In order to extract significant feature points in the image, applied a feature point extraction algorithm based on a fusion of edge maps using morphological and wavelet methods. Evaluation of feature points thus obtained has been done for geometric transformations and image scaling. A region growing algorithm was then employed to isolate the tumor region. Preliminary results show that our approach has achieved good segmentation results. Also this approach was reduces a large amount of calculation. Future work will involve an investigation of the method in automatic 3D tumor segmentation, segmentation of ROI’s in other medical images, as well as the importance of implemented technique in medical image retrieval applications.STEPS IN IMAGE PROCESSING There are some common steps in image processing as listed below, though all image processing systems would not require all steps together. Image acquisition: An image is captured by a sensor such as a TV camera and it is digitized. Image Preprocessing: The first step in preparing the picture for higher-level processing is called pre-processing and the purpose of pre-processing is two-fold: to eliminate undesirable features that will hinder further processing and to extract the desirable features that represent useful information in the image. Image segmentation: The main aim of segmentation is to reduce the information to enable easy analysis. Segmentation is also useful in Image Analysis and Image Compression .Representation and description: Representation is transforming raw data into a form suitable for computer processing. Description, which is also called feature extraction, deals with extracting features that result in some quantitative information of interest or features which are basic for differentiating one class of objects from another.Recognition & Interpretation: Recognition is the process which assigns a label to an object based on the information provided by its descriptors. Interpretation is the process of assigning meaning to an ensemble of recognized objects.PROPOSED METHODOLOGYThe basic purpose of this paper is to show the tumor region. In this paper, we are implementing the system for brain tumor detection from MRI images, the malignant or benign tumor region we will find by this system. The complete system includes preprocessing of MRI by using Median filtering, skull removing by morphological filtering, and segmentation by using watershed segmentation technique, and linear SVM implementation by using extracted feature of the MRI followed by edge detection for sharpening of edges to get a more clear and enhanced image. In the testing part we are passing the parameter to the SVM that is the previously stored feature with class name and the extracted feature of new MRI image. Preprocessing Preprocessing of MRI images includes the de-noising the MRI image and also skull masking. The Median filter is used to de-noising the MRI images by converting first the RGB image to grayscale image so we can get one intensity scale. The skull masking is the process which will perform on the de-noised image. The purpose of this process is to remove fatty tissue, skull part and hair part in the MRI so we can process with purely brain tissues only and the ambiguity in identification of tumor get reduce. For morphological filtering we are using different masks on MRI image horizontal, diagonal, anti-diagonal and vertical masks are used to process the skull masking. Steps for preprocessing are as follows: 1) Image is converted to gray scale. 2) A 3×3 median filter is applied on brain MR image in order to remove the noise.3) The obtained image is then passed through a high pass filter to detect edges. The high pass filter mask is used. The edge detected image is added to the original image in order to obtain the enhanced image. In order to preserve the local details of the image, median filter should only modify the intensity of ruined pixels on the damaged image. However, it is very difficult to detect the ruined pixels from this image correctly. Even for fixed-valued impulse noise (i.e. salt-and-pepper noise)Segmentation Image Segmentation is the process of partitioning a digital image into multiple regions or sets of pixels. Essentially, in image partitions are different objects which have the same texture or color. The image segmentation results are a set of regions that cover the entire image together and a set of contours extracted from the image. All of the pixels in a region are similar with respect to some characteristics such as color, intensity, or texture. Adjacent regions are considerably different with respect to the same individuality.Watershed segmentationWatershed Segmentation: Image segmentation is stated as the process of assigning a label to every pixel in an image such that Pixels with same label share certain identical visual characteristics. With respect to our paper we are using watershed segmentation.Any grayscale image can be viewed as a topographic surface where high intensity denotes peaks and hills while low intensity denotes valleys. You start filling every isolated valleys (local minima) with different colored water (labels). As the water rises, depending on the peaks (gradients) nearby, water from different valleys, obviously with different colors will start to merge. To avoid that, you build barriers in the locations where water merges. You continue the work of filling water and building barriers until all the peaks are under water. Then the barriers you created gives you the segmentation result. Support Vector Machine (SVM)SVM Here we are using texture feature and color feature to train SVM. For texture feature we have taken the edge width as a parameter and for color feature we have taken RGB as a parameter.Fig: the classification process of SVMSVM are based on optimal hyper plane for linearly pair able patterns but can be extended to patterns that are not linearly separable by transformations of original data to map into new space. They are explicitly based on a theoretical model of Learning and come with theoretical guarantees about their performance. They also have a modular design that allows one to separately implement and design their components and are not affected by local minima. Support vectors are the elements of the training set that would change the position of the dividing hyper plane if removed. Support vectors are the critical elements of the training set. The problem of finding the optimal hyper plane is an optimization problem and can be solved by optimization techniques.Edge DetectionIn image processing especially in computer vision, the edge detection treats the localization of important variations of a gray level image and the detection of the physical and geometrical properties of objects of the scene. It is a fundamental process detects and outlines of an object and boundaries among objects and the background in the image. Edge detection is the most familiar approach for detecting significant discontinuities in intensity values . Edges are local changes in the image intensity. Edges typically occur on the boundary between two regions. The main features can be extracted from the edges of an image and this provides major features for image analysis. These features are used by advanced computer vision algorithms. Edge detection is used for object detection which serves various applications such as medical image processing, biometrics etc. Edge detection is an active area of research as it facilitates higher level image analysis. There are three different types of discontinuities in the grey level like point, line and edges . Spatial masks can be used to detect all the three types of discontinuities in an image.MRI/CT images of the brain are processed for the detection of tumor using MATLAB. The block diagram in Figure shows the overall processing technique. The Methodologies employed here are Algorithm Used for Detection for Tumor:Step 1: Input the imageStep 2: Convert the RGB image to greyStep 3: Use the filters and antistrophic diffusion for removing noise and perform subtraction operation and form historgramic equivalent for Contrast enhancement Step 4: Apply Watershed Segmentation for tumor detectionStep5: Use SVM classifier for detecting exact pixels of the tumor regionStep6: Final segmented tumor detected area in the imageEXPERIMENTAL ANALYSIS AND RESULTSProposed Brain tumor detection system is improve with segmentation of preprocessed image then resulted image goes with object labeling and feature extraction. Extracted features used to train SVM and the database of feature is use for pattern matching and test the system.CONCLUSIONThe proposed system is the combinations of some technologies like k-means for segmentation, HOG for object labeling, median filter, morphological filter and wavelet transform for the preprocessing and skull masking. So the result of this all combination is very faire than the individual of them or the some other combinations. The linear SVM and HOG are work with coordination because the HOG extracts the feature and SVM use that data for learning the SVM, so the SVM will able to make the patterns and after training in testing it will work for the test the pattern and gives the conclusion. Here we are dividing the tumor images in Malignant or Benign classes. Also after identification the image and the feature of it are added into the database of the SVM so we can increase the accuracy of the proposed system. The proposed methodology aims to detect the brain Tumor from CT/MRI brain images. The detected tumorous lesion is then segmented using image processing algorithms and the morphological operations are performed to obtain the vital parameters like Mean, Standard deviation, Third moment, Area, Entropy of the image. The results are depicted in two tabulations, one for CT and the other for MRI. From the obtained numerical results we interpret that the values for abnormal condition is always high. Thus through this thesis an attempt hasReferences Karuna Ankita Joshi, M., Automatic detection and severity analysis of brain tumors using GUI in MATLAB, IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163, PISSN: 2321-7308. R. Yogamangalam and B. Karthikeyan, Segmentation techniques comparison in image processing, International Journal of Engineering and Technology (IJET), 2013, 5(1), 307-313. A. Chaudhary and T. Gulati, Segmenting digital images using edge detection, International Journal of Application or Innovation in Engineering & Management (IJAIEM), 2013, 2(5). Ashraf Anwar and Arsalan Iqbal, 2013. Image Processing Technique for Brain Abnormality Detection. International Journal of Image Processing (IJIP), (7). R.Muthukrishnan and M.Radha, Edge Detection Techniques for Image Segmentation, in International Journal of Computer Science & Information Technology, Vol 3, No 6, Dec 2011. R. Yogamangalam and B. Karthikeyan, Segmentation techniques comparison in image processing, International Journal of Engineering and Technology (IJET), 2013, 5(1), 307-313.