Effective MKSVM Classifier with LDA Methods for Brain Tumor Detection in MR Images
K. Shankar1, M. Ilayaraja2, P. Deepa Lakshmi3, S. Ram Kumar4, K. Sathesh Kumar5
1K. Shankar, Department of Computing, Kalasalingam Academy of Research and Education College, Krishnankoil (Tamil Nadu), India.
2M. Ilayaraja, Department of Computing, Kalasalingam Academy of Research and Education College, Krishnankoil (Tamil Nadu), India.
3P. Deepa Lakshmi, Department of Computing, Kalasalingam Academy of Research and Education College, Krishnankoil (Tamil Nadu), India.
4S. Ram Kumar, Department of Computing, Kalasalingam Academy of Research and Education College, Krishnankoil (Tamil Nadu), India.
5K. Sathesh Kumar, Department of Computing, Kalasalingam Academy of Research and Education College, Krishnankoil (Tamil Nadu), India.
Manuscript received on 05 December 2019 | Revised Manuscript received on 24 December 2019 | Manuscript Published on 31 December 2019 | PP: 987-992 | Volume-8 Issue-4S2 December 2019 | Retrieval Number: D10971284S219/2019©BEIESP | DOI: 10.35940/ijrte.D1097.1284S219
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In recent times, one of the perilous diseases which cause to increase the patient’s death rate is a brain tumor. For diagnosing the tumor diseases from Magnetic Resonance Images (MRI), different classification methods have been analyzed. This paper presented an innovative method to diagnose brain tumor disease by using the classification process in MRI. From the input MR images, brain tumor image is classified by the supervised learning classifier i.e. Multi Kernel Support Vector machine (MKSVM). This model incorporates the extraction of feature and reduction process. All the MRI brain images are considered to extract some standard features and reduction reason dimensionality reduction that is Linear Discriminant Analysis (LDA) applied. Reason for this technique to the removal of multi-collinearity enhances the execution of the proposed model. Utilizing the feature vector attained out of the MRI images, SVM classifiers are utilized to image classification. The procedure comprises of two parts that are training stage as well as a testing stage. Parameters used to analyze the classified images as sensitivity, specificity, accuracy and so on. A cross-validation plot is adopted to enhance the generalization capability of the framework.
Keywords: Image Processing, MRI, Brian, Tumor Detection, Classification, LDA, MKSVM.
Scope of the Article: Image Security