Brain tumors can develop at any brain location with uneven boundaries and shapes. Typically, they increased rapidly due to their size doubling in twenty-five days. If they were unrecognized in earlier phases, patients suffered from various medical problems, including death. Therefore, the identification of brain tumors in the earlier stages is one of the critical aspects. In addition, an effective imaging sequence also plays a vital role in tumor diagnosis. Magnetic resonance (MR) imaging is widely used among the available scanning approaches. Therefore, in this article, we develop a distinctive novel method to classify MR-based brain images. Here, initially, we improve the brightness of brain MR images using a median filter, and then we employ image data augmentation to increase the model’s accuracy. Later, we obtain the region of interest (ROI) by Otsu’s thresholding and morphological operations. Then, we extracted relevant local textures and shaped informative features from the ROI using Enhanced gradient local binary patterns (EGLBPs) and Modified pyramid histogram of oriented gradients (MPHOG). Finally, we perform classification using various supervised learning approaches: support vector machine (SVM), K-nearest neighbors (KNN), and ensemble learning. All these experiments are implemented on Harvard Medical School (HMS) database. Based on the simulation results, our proposed imaging system outperformed state-of-the-art methods in classification and segmentation. Hence, our suggested framework can be used as a predictive tool for diagnosing patients with brain tumors.