Brain hemorrhage is a critical condition that needs quick and precise response and diagnosis for timely treatment. Traditional methods like CT and MRI scans depend on expert interpretation, which can be time-consuming and prone to errors. This study introduces an automated framework with deep learning to detect and classify brain hemorrhages. By utilizing convolutional neural networks (CNNs), the system recognizes important features in medical images and classifies hemorrhages into types such as intracerebral, subarachnoid, subdural, and epidural. Trained and tested on brain scan datasets, the framework depicts the potential of deep learning to deliver quick and accurate diagnoses, avoiding delays and enhancing patient outcomes significantly.