Detection of Brain Tumour Using Image Segmentation
In the present days, some of the new methods such as Image processing are used in the field of medicine for enhancing the speedy detection of abnormalities, such as the breast cancer, lung cancer, brain cancer and so on. This paper investigates on the Image processing based segmentation methods for segmenting brain tumor in the Computed Tomography (CT) images ,MRI images and PET images. Mean and median filters are used in the image pre-processing stage and power law transformation(PLT) is employed in the image enhancing stage. For segmenting the tumour in the brain images two different image segmentation techniques such as Otsu's thresholding algorithm and k-Means clustering methods are used. The performance evaluation of the different methods used are done by calculating the performance evaluation parameters such as Signal to noise Ratio(SNR) ,Mean Square Error (MSE) and Peak Signal Noise to Ratio (PSNR) on the various images used for segmentation. Based on the investigations ,It is observed that K-Means segmentation algorithm and the PET images fetches the better results.