Magnetic Resonance Imaging (MRI) is one of the most important diagnostic tool. There are many methods to segment the
tumor of human brain. One of these, the conventional method that uses pure image processing techniques that are not preferred because they need human interaction for accurate segmentation. But unsupervised methods do not require any human interference and can segment the brain with high precision. In this project, the unsupervised classification methods have been used in order to detect the tumor disease from MRI images. These methods involved KÂ mean or Isodat, which were based on the digital value distribution. The results show the classification process was a powerful tool to identify the Tumor disease from MRI images. All results were evaluated by using the ENVI Version 3.2 facility.