A Novel statistical parametric analysis of brain tumor images using contourlet transform and Fuzzy C-means clustering algorithm

Brain tumor detection is one of the most critical tasks in the field of medical image processing. Various studies reveal that the existing methods have not considered the images of poor quality like images with high noise and low brightness due to significant image processing difficulty, which can leads to error in assessment. In an image, noise may creep in at various stages such as at the time of image acquisition, during transferring the image or storing the image as data etc. As denoising filter, adaptive fuzzy filter is selected. This filter will perform in frequency domain in contourlet transform. The image is then segmented using fuzzy C means clustering technique, watershed segmentation and level set segmentation to get the best results in a noised image. These techniques are accurate and faster as they require less computational time as compared to other techniques.