• Dr.S. Prabaharan
• Dr. R. Jegadeesan
• B.Madhav
• K.Laxmi priyanka
• T.Aishwarya

In this paper, we propose a comparative hybrid model for the detection of brain tumors based on the image processing, k-means,EM models . The K-Means algorithm is used to divide the image into K groups based on the distance between the pixels. Each group is grouped by the group center point,I.e., a centroid.In order to avoid inefficiency we use image processing. To remove the unwanted information (noise, film artifacts in MRI) in the original MRI image we use the pre-processing technique. We use the segmentation to extract a certain area of the image. We distribute an image to small parts and analyze information in a digital format.We use the extraction feature where some parameters are considered to extract features such as configuration, shape (shape), size and location of the image.Quality metrics such as PSNR is calculated to extract the features. With regard to the results obtained from extraction properties we perform the tumor classification process.We EM algorithm which is a general technique for maximum likelihood estimation or maximum a posteriori when there is a missing or unknown data.. Finally we compare the results obtained fron K-Means and EM algorithms and propose a best algorithm to detect the brain tumor .These techniques can then be applied to predict brain tumor detection at an early stage.

Pre processing, brain tumor,feature extraction, classification, MRI,support vector machine.

"A COMPARATIVE STUDY ON BRAIN TUMOR DETECTION USING IMAGE PROCESSING K-MEANS AND EM ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 3, page no.60-64, March-2019, Available :http://www.jetir.org/papers/JETIRAU06009.pdf

"A COMPARATIVE STUDY ON BRAIN TUMOR DETECTION USING IMAGE PROCESSING K-MEANS AND EM ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 3, page no. pp60-64, March-2019, Available at : http://www.jetir.org/papers/JETIRAU06009.pdf