Classification of galaxy morphology is essential to comprehending the genesis and development of galaxies. Astronomers must manually analyze data using traditional classification techniques, which is laborious and sensitive to subjectivity. It is now essential to have automatic, precise, and effective categorization systems due to the growing availability of large-scale astronomical datasets. This paper suggests automating the classification of galaxy morphologies with a deep learning-based method that uses Convolutional Neural Networks (CNNs) and transfer learning with VGG16. To overcome issues like dataset variability and class imbalance, the system uses data augmentation and picture preparation techniques. Refined to extract hierarchical features from galaxy images, the VGG16 model greatly increases classification accuracy after being pre-trained on the ImageNet dataset. Spiral, edge-on, cigar-shaped smooth, in-between smooth, and completely round smooth galaxies are among the morphological categories in which the model performs well when trained and assessed using publicly accessible astronomical datasets. Astronomers and researchers can input galaxy photos and receive real-time classification results by integrating the trained model into a web application based on Streamlit, which makes practical deployment easier. Performance indicators including F1-score, recall, accuracy, and precision attest to the suggested system’s resilience and dependability. The results show that deep learning models may greatly improve the efficiency of morphological classification while preserving high accuracy and lowering manual labor. This study explores more developments in automated astronomical image processing, which will help with scientific discovery and large-scale galaxy classification.