Perancangan Sistem Monitoring Dan Alarm Atensi Mahasiswa Untuk Mobile Learning Menggunakan Pengenalan Ekspresi Wajah Dengan Metode Jaringan Saraf Tiruan

Fatima, Azzezza Nurul (2023) Perancangan Sistem Monitoring Dan Alarm Atensi Mahasiswa Untuk Mobile Learning Menggunakan Pengenalan Ekspresi Wajah Dengan Metode Jaringan Saraf Tiruan. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02311940000100-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (6MB) | Request a copy

Abstract

Seiring dengan perkembangan teknologi, pembelajaran juga mengalami perubahan, salah satunya adalah pembelajaran secara daring atau online learning. Media belajar juga sudah berkembangan dengan memanfaatkan perangkat seluler yang dikenal sebagai Mobile Learning (m-learning). Kondisi m-learning cenderung mempersulit pengajar untuk mengevaluasi komitmen siswa dalam hybrid maupun kelas daring, seperti kurangnya perhatian dari siswa. Agar dosen menyadari akan kurangnya perhatian siswa selama pertemuan kelas, maka akan lebih bermanfaat terdapat sistem monitoring dan alarm atensi mahasiswa selama pembelajaran m-learning berlangsung yang dapat membantu dosen dalam mengetahui kondisi kelas daring. Sistem FER ini dirancang agar dapt mendeteksi ekspresi wajah Drowsy dan Neutral. FER diawali proses face detection dengan Haar-Cascade classifier, kemudian dilanjutkan facial feature extraction menggunakan 68 facial landmarks, dan terakhir adalah model klasifikasi menggunakan CNN. Hasil traning model CNN diperoleh akurasi train, validasi, dan test secara berturut-turut adalah 99.7%, 94.9%, dan 99.0%. Sistem kemudian diimplementasikan pada Raspberry Pi 4 dan Pi Camera V2 untuk mendeteksi secara real-time proses pembelajaran m-learning pada data rekaman pembelajaran kelas daring mahasiswa Teknik Fisika ITS. Berdasarkan hasil implementasi, jumlah hasil deteksi ekspresi wajah pada satu frame yang sama paling banyak adalah 11 ekspresi wajah (5 Drowsy dan 6 Neutral) dari 20 mahasiswa yang tertangkap kamera, atau sekitar 55% berhasil terdeteksi ekspresi wajahnya. Sistem juga dapat memberikan peringatan kepada dosen seberapa sering mahasiswa pada kelas tersebut terdeteksi Drowsy atau tidak atensi dalam interval waktu 10 menit. Resolusi dan tingkat pencahayaan akan berpotensi meningkatkan performa sistem monitoring dan alarm ini.
===============================================================================================================================
Along with technological advancements, education has also undergone changes, one of which is the shift towards online learning or e-learning. Learning media has also evolved by leveraging mobile devices known as Mobile Learning (m-learning). The m-learning environment tends to challenge educators in assessing students' commitment in hybrid or online classes, such as the lack of attention from students. To address this issue and help educators be aware of students' lack of attention during online learning sessions, a monitoring system with an attention alarm for students during m-learning was proposed. This system aims to assist educators in understanding the condition of online classes. The Facial Expression Recognition (FER) system is designed to detect Drowsy and Neutral facial expressions. The FER system starts with face detection using the Haar-Cascade classifier, followed by facial feature extraction using 68 facial landmarks, and lastly, classification using Convolutional Neural Network (CNN). The trained CNN model achieved training, validation, and test accuracies of 99.7%, 94.9%, and 99.0%, respectively. The system was then implemented on a Raspberry Pi 4 and Pi Camera V2 to detect real-time facial expressions during m-learning sessions, using recorded data from online physics classes at ITS (Institut Teknologi Sepuluh Nopember). Based on the implementation results, the highest number of facial expression detections in a single frame was 11 facial expressions (5 Drowsy and 6 Neutral) out of 20 students captured by the camera, approximately detecting 55% of their facial expressions. The system also provides alerts to educators about how often students are detected as Drowsy or inattentive during a 10-minute interval. Improving the resolution and lighting conditions could potentially enhance the performance of this monitoring and alarm system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Mobile learning, atensi, monitoring, CNN, ekspresi wajah , attention, facial expressions
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1573 Detectors. Sensors T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TA Engineering (General). Civil engineering (General) > TA1650 Face recognition. Optical pattern recognition. T Technology > TA Engineering (General). Civil engineering (General) > TA593.35 Instruments, cameras, etc.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Azzezza Nurul Fatima
Date Deposited:	11 Sep 2023 02:29
Last Modified:	11 Sep 2023 02:29
URI:	http://repository.its.ac.id/id/eprint/103682

Actions (login required)

View Item