In tunnel construction, complex geological conditions are crucial for the selection of construction schemes and safety control. Accurate classification of surrounding rock is essential for optimizing construction processes and reducing risks. However, traditional methods for surrounding rock classification rely heavily on the experience of geological experts, which can lead to subjectivity and time-consuming processes, making it difficult to meet modern engineering demands for efficiency and precision. With the advancement of deep learning technologies, data-driven automated methods for surrounding rock classification have become feasible. This paper proposes a rapid classification method for surrounding rock at the tunnel face based on deep learning. By collecting images at the construction site and utilizing the VGGNet algorithm, a surrounding rock classifica-tion model is developed and evaluated using confusion matrices and accuracy metrics. Experi-mental results demonstrate that the proposed method achieves an accuracy of 90.42%. Integrating this model into a rapid detection device enables quick classification of surrounding rock at complex geological tunnel faces, thereby supporting decision-making for construction schemes.