Yasmin, Andri (2015) Rancang bangun sistem monitoring kelas terintegrasi berbasis rasberry PI dan webcam sebagai pendeteksi kecurangan saat ujian. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2211100194-Undergraduate_Thesis.pdf Download (3MB) | Preview |
Abstract
Kecurangan dalam dunia pendidikan bukanlah hal yang aneh. Kecurangan ini memiliki banyak bentuk, akan tetapi kecurangan yang paling umum ialah kecurangan saat ujian. Hal ini sangat bertentangan dengan salah satu fungsi sekolah, yaitu membentuk moral siswa-siswi.
Terdapat banyak cara untuk menanggulangi kecurangan ini, mulai dari penempatan guru, pemberian jarak antar murid, hingga pemberian soal yang berbeda-beda. Dengan mengintegrasikan beberapa Raspberry Pi dengan kamera dan dihubungkan dengan server via wifi, maka akan dibuat sistem monitoring kelas terintegrasi yang dapat dilihat langsung dari server sehingga dapat memonitor beberapa kelas sekaligus. Sistem ini juga memiliki beberapa fitur yang membuatnya lebih unggul jika dibandingkan dengan CCTV pada umumnya, yaitu sistem pengiriman suara ke berbagai channel dan sistem pendeteksi gerakan yang dapat memonitor apakah ada gerakan mencurigakan atau tidak.
Dari hasil pengujian yang dilakukan, pengiriman gambar terjadi secara langsung, akan tetapi frame per second maksimal berbeda antara Raspberry Pi B+ dan Raspberry Pi 2. Raspberry Pi 2 dapat mencapai 4 frame per second tanpa pengiriman, sedangkan Raspberry Pi B+ dapat mencapai 3 frame per second tanpa pengiriman. Jika terjadi kecurangan, maka akan terdapat gerakan yang banyak dan cepat yang kemudian akan terdeteksi. Jarak efektif pendeteksi gerakan ialah 2 hingga 10 meter. Presentase gerakan bergantung pada batasan yang diatur. Apabila batasan diatur 4000 dengan jarak 5 meter, maka gerakan akan terdeteksi apabila objek bergerak sebanyak 25%.
=======================================================================================================
Fraud, in education world, is very common. This fraud has many forms, but the most common one is cheating in examination. Cheating has many forms, ranging from working with a friend, using notes, to use existing technology, such as mobile phones. This is contradictory to one of the school’s function, which is shaping the moral of students.
There are many ways to handle cheating, ranging from placing teacher in each class, giving distance between students, to giving different sheet. By integrating some of Raspberry Pis with cameras and connect them into a server via wifi, it is possible to create an integrated class monitoring system that can be viewed directly from the server so that it can monitor several classes at once. This system has a few features that differ it from ordinary CCTV, which is audio transfer that can send audio into two different channels, and motion detector that can detect suspicious movement. From the results of tests performed, sending images occur directly, but frames per second maximum of the difference between the Raspberry Pi B + and Raspberry Pi 2. Raspberry Pi 2 can reach 4 frames per second without delivery, while the Raspberry Pi B + can reach 3 frames per second without delivery. Motion can only be detected if there are a lot of fast movement. Motion detection effective range is 2 to 10 meters. Percentage movement depends on the threshold. If the threshold is set 4000 at a distance of 5 meters, then the movement will be detected when a moving object as much as 25%.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSE 004.678 Yas r |
| Uncontrolled Keywords: | Raspberry Pi; Sistem Monitoring Terintegrasi; Background Subtraction |
| Subjects: | L Education > LB Theory and practice of education > LB1044.87 Internet in education (e-learning). Virtual reality in education. |
| Divisions: | Faculty of Electrical Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
| Depositing User: | - Taufiq Rahmanu |
| Date Deposited: | 26 Jun 2019 03:45 |
| Last Modified: | 26 Jun 2019 03:45 |
| URI: | http://repository.its.ac.id/id/eprint/63269 |
Actions (login required)
 |
View Item |