This work uses a Adaboost-based face detection algorithm, which has the characteristics of high efficiency, stability and convenience, and has been widely used in the field of pattern recognition. In the face recognition module, we use a subspace-based face recognition algorithm. Specifically, we use a subspace-based recognition algorithm, which is the most classic face recognition algorithm, not only the calculation is simple and convenient, but also the recognition rate is relatively stable. This work uses a single-person face image verification method to verify the identity of the student, in other words, the system reads the face image of the student in the database through the IC card of the student's brush, and compares it with the face detected by the system. This method has a great advantage in dealing with large-scale sample sets.