The research presents a design for an automated checking system for students. The system takes a picture of the student, then it extracts his/her basic facial features. The network was trained using the reverse spreading algorithm. If a training database is generated for each student consisting of 15 training samples contained of the necessary facial expressions to identify the student for one time at the beginning of the semester, then the neural network will be trained on students database to obtain a trained neural network able to identify the students of each category depending on their physical appearance. That will result in knowing who attends and who does not attend the session. The system designed for this purpose was supplied with the trained network. The system provides the possibility of automated checking for students according to the content of the study giving the alarm in case of the existence of the picture of a student who does not belong to the same group.