Brain tumor detection is an important task in medical image analysis. Tumor in the brain leads cancer and is to be diagnosed at earlier stage itself. Image processing techniques are applied for MRI images to segment and detect tumor. Detection of tumor in brain is done manually which is complex, tedious task and experience is necessary to detect a simulation results. Hence , many Automatic image segmentation algorithms are developed to detect and segment tumor from MRI images. This image segmentation algorithms uses either texture information or edge information for segmentation of MRI images. Graph-cut optimization approach has been developed recently, which utilizes both texture and edge information. The proposed work extends the graph cut approach in three steps. First a more powerful, iterative version of the optimization is developed. In the second step in order to reduce the user interaction a powerful iterative algorithm is applied for producing the result. Finaly,, a robust algorithm has been developed for border connecting to predict both the alpha-matte around an object boundary and the colours of foreground pixels.
Selective Brain MRI Image Segmentation using Fuzzy C Mean Clustering Algorithm for Tumor Detection
