A brain tumor is an abnormal growth of cells in the brain, which can be benign or malignant and categorized as primary or secondary. This research presents an advanced Tumor Detection System designed to improve the accuracy and speed of brain tumor diagnosis, aiding clinical decision-making. The study explores augmentation techniques, feature extraction, and various machine learning models such as CNN, SVM, KNN, Decision Tree, XGBoost, MLP, and Random Forest. We evaluate these methods, discussing their strengths, limitations, and implications for the future of automated brain tumor detection. Deep learning models, specifically the CNN model, work best for this dataset in conclusion. Alternatively, models like XGBoost, MLP Classifier, and Random Forest should be our next go-to based on the task complexity at hand and computational resources. The highest accuracy was able to be obtained using convolutional neural networks (CNN) that achieved and passed 98.6%, making them highly effective at pattern recognition tasks.