The brain is the interior part of the central nervous system. Brain tumor is an unbalanced growth caused by cells reproducing themselves in an uncontrolled way. The seriousness of brain tumor is very high among all types of cancers because of space for me inside the skull. So, immediate detection and proper treatment can save a person’s life. In this paper a system for brain tumor extraction is designed. It uses the MRI Scanned Images at input. In the pre-processing stage, noise is removed and the texture features are extracted from it with the help of Gray level co-occurrence matrix (GLCM). Then with the help of the obtained features classification of images into normal and abnormalised one uses Probabilistic Neural Network (PNN) classifier. Then using segmentation technique and morphological operations tumorous part is located from abnormal image. The accuracy of the proposed system is 86.2% & it is evaluated in terms of confusion matrix.