The application of Artificial Intelligence (AI) and Machine Learning (ML) models in medical imaging represents a promising approach for the detection and classification of abnormal clinical patterns. The classification of brain tumors is crucial for diagnosis and treatment planning. In this study, the authors investigate the effectiveness of texture analysis using Haralick and Gabor descriptors, as well as LBP and HOG descriptors, in conjunction with machine learning classifiers (RF, SVM, LR, and KNN) for the classification of brain tumors based on Magnetic Resonance Imaging (MRI). This by using the Gray Level Co-occurrence Matrix (GLCM) technique, which permits the extraction of texture features through the statistical analysis of pixel neighborhood relationships. The dataset comprises MRI images representing Glioblastomas, Meningiomas, Pituitary Tumors, and non-tumor cases. Our findings suggest that texture features, notably Haralick and Gabor descriptors, hold great potential for brain tumor classification. Future research aims to employ more sophisticated deep-learning algorithms to enhance the precision of brain tumor classification.