An Intelligent Access Control Recognition Framework Based on Asian Face Data Augmentation and Edge Computing Optimization in Private University Campuses

To address the practical needs of campus security management in private universities, this paper proposes an intelligent access control system based on facial recognition. To enhance the accuracy of the model in Asian facial recognition tasks, we designed a data augmentation strategy tailored to Asian facial features, thereby increasing the diversity and richness of the training data. By integrating edge computing technology, we optimized the facial recognition model through methods including model pruning, quantization, and knowledge distillation, achieving efficient deployment on edge devices. Experiments conducted on a subset of Asian faces from the VGGFace2 dataset demonstrate that the proposed method outperforms traditional approaches such as LSTM and CNN in terms of accuracy, inference time, model size, and energy consumption. This success enables real-time and efficient facial recognition on edge devices. The research findings provide effective technical support for smart campus security management in private universities.