Students are the main body of the classroom, and their learning status reflects the quality of classroom teaching to a certain extent. This paper aims to manage students’ learning status in industrial education classrooms by designing a student learning status assessment system. The collected video data from industrial education classrooms are processed by binarization and histogram equalization. The key technologies, such as face detection and eye movement analysis, are used to detect the learning status of students in the industrial education classroom, and the functional modules, such as data acquisition and statistical analysis, are combined to form this paper’s student learning status assessment system based on facial feature detection. It is found that the face detection algorithm of this paper’s system improves the detection accuracy by 11.35% compared with the baseline algorithm when the standard difficulty is difficult, and this paper’s algorithm is able to successfully detect and display the students’ facial features through 68 key feature points. The system in this paper is able to detect students’ learning states, such as concentration, fatigue, and doubt, by analyzing their eye movement frequency and other indicators. After applying the system discussed in this paper, the final grades of students in industrial education have significantly improved.