Extension of MDNFM For Smart Classroom Activity Monitoring

Abstract

One application of the automated smart learning process is monitoring the student's activities in the classroom. However, when indulged in taking lectures, the instructors or the faculty cannot monitor the activities of the students appropriately. Therefore, traditional methods employ several face detection algorithms to monitor the activities of the students.

 However, the results obtained through these methods are inaccurate, and hence an efficient algorithm is required to predict the active state of the student in the classroom. Hence, we use  Multi-tasking Deep Neuro-Fuzzy Model (MDNFM) for the activity monitoring (AM) of the students in the classroom. Initially, the images of the students in the classroom are captured through web cameras and other accessories placed in the smart class. Then, the acquired image is passed to the Capture, Transform, and Flow (CTF) tool for storing and transferring for further processing of images.