Brain tumor is a major cause of death among many people. There are over one hundred and twenty types of brain and central nervous system tumors. In this paper, an automated support system has been proposed for the classification of tumor with the help of soft computing techniques. The usual detection of the brain tumor is accompanied with a lot of complexities due to the structure of the cells. Artificial Neural Network is used to classify the MRI image whether it is a tumor or benign. The constraints of manual analysis of the signal are it is time consuming, inaccurate and requires intensive trained person to avoid diagnostic errors. The Probability of proper classification has been increased by using efficient image segmentation algorithm, such as solidity and level set technique. Back Propagation Network (BPN) with image and data processing techniques is employed to implement an automated tumor classification. The performance of BPN classifier was evaluated in terms of training performance and classification accuracies.