Evaluation of Concentration Levels in Learning Using Zoom Meeting by Applying the Histogram of Oriented Gradients (HOG) Method and Support Vector Machine (SVM) Classification

Abstract

The online learning system is one of the implementations of distance education through a learning system, an application that is widely used during distance learning is Zoom Meeting. This research was conducted to find out how effective the use of the Zoom Meeting application is for online learning. This study focuses on detecting eyes to determine a person’s focus during online learning using zoom meetings, analyzing a person’s blink of an eye, and then determining the size of the ratio in the eyes. It has been established that the level of attention of a normal person can be determined by the degree of a blink of an eye. If a person feels tired or out of focus, it can be seen in the blink of his eyes. In this study, a method will be developed that is used to detect eyes using a Histogram of Oriented Gradients (HOG) as the extraction of its characteristics and a Support Vector Machine (SVM) as its classification. Base on data analysis proved that the ability of the process to identify a person’s focus level is 99.59% accuracy, 99.49% accuracy, and 98.84% recall. The system’s skills to detect blinking eyes with PERCLOS is excellent with an average 98.60%.

Access to Document

Other files and links

• Link to publication in Scopus

Fingerprint