The presence of a brain tumor is a critical and hazardous issue that hinders the proper functioning of the human body. An unnatural and uncontrolled growth in brain cells is called a brain tumor. It is really necessary to detect brain tumors at an early stage otherwise it may become very difficult to find the complexity of brain tumors in time. Detecting brain tumors is critical for prolonging and saving patients’ lives, and therefore, advanced techniques for detecting them are necessary in the medical field (Akkus et al., Neurocomputing 392:189–195, 2019). Magnetic Resonance Imaging (MRI) is a crucial tool for examining organs and structures inside the human body, using large magnetic and radio waves. Although several existing methods, such as Random Forest, Fuzzy C-Mean, Artificial Neural Network (ANN), and Wavelet transform, have been used to detect brain tumors, their accuracy is inadequate and execution time is longer. In this project, we utilize image classification and segmentation methods to analyze images more effectively, employing algorithms such as CNN (Convolution Neural Network), U-net Architecture, Max-pooling, Up-sampling, Down-sampling, and Data augmentation, which efficiently detect brain tumors in less execution time. Initially, we obtain data from various sources and then use CNN image classification to preprocess the images, which improves accuracy. Image classification is a process that simply involves the extraction of features from the image for observing some patterns and useful information in dataset (Siar and Teshnehlab, Diagnosing and classification tumors and MS simultaneous of magnetic resonance images using convolution neural network, 7th Iranian Joint Congress on Fuzzy and Intelligent Systems (CFIS), 2019). After that extraction of essential features using the segmentation method is done, CNN U-net architecture, max-pooling, up-sampling, and data augmentation methods were applied which help accurately to detect the location and shape of brain tumors. The experimental results show that the above used methods for brain tumor detection achieve better accuracy with lower execution time than other existing methods for the detection of brain tumors.