Design of Smart Grid Network Access System Based on Face Recognition

Facial recognition technology is a key technology to achieve secure access to power grid. Among them, the recognition method based on Adaboost model is fast, but the accuracy is not high. Although deep neural network has high recognition accuracy, its detection speed is slow. In this paper, an online face recognition algorithm suitable for a single regional power grid is studied based on the power grid access characteristics. This integrated system helps substation personnel reduce communication costs. In this way, equipment maintenance checks, equipment current status checks and substation area physical integrity checks are completed. Experiments show that the algorithm proposed in this paper can achieve a computation rate of 30 FPS on CPU and maintain a high accuracy rate.