In modern days, detection of brain tumour has turned out to be a breath taking challenge in scientific endeavours. An automatic segmentation of brain pictures has a considerable role in lessening the burden of manual labelling and increasing the accuracy of brain tumour analysis. Magnetic Resonance Imaging (MRI) has an excessive spatial reasoning view of brain and it is a productive tool used to diagnose a huge variety of diseases and verified to be an extraordinarily suitable imaging technique. This paper gives a dependable detection technique primarily based on CNN that reduces operators and errors. The Convolutional Neural Network (CNN) is used in convolving a signal or a photo with kernels to gain function maps. The image processing strategies together with image conversion, feature extraction and histogram equalization have been evolved for extraction of the tumour in the MRI images of the most cancers affected patients. An appropriate Fuzzy Classifier is developed to recognize healthier tissue from most cancers tissue. The entire gadget is divided into two stages: first off getting to know/Training Phase and secondly Recognition/Testing Phase. The purpose of the undertaking is to detect and extract the of tissue abnormalities by using the usage of the biochemical capabilities. The specificity and the sensitivity of the method are evaluated and accuracy is decided. The performance parameters display widespread outputs which are useful in extracting tumour from mind MRI image.