Adlian, Jefiza (2017) SISTEM PENDETEKSI JATUH BERBASIS SENSOR GYROSCOPE DAN SENSOR ACCELEROMETER MENGGUNAKAN BACKPROPAGATION. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
adlian jefiza tesis.pdf - Published Version Download (3MB) | Preview |
Abstract
Terjatuh merupakan aspek eksternal yang bisa mengakibatkan kematian pada lansia. Dengan banyaknya kegiatan yang bisa mereka kerjakan berakibat meningkatnya kemungkinan terjatuh. Sebuah perangkat pendeteksi terjatuh dirancang untuk meminimalisir risiko pasca terjatuh. Sebuah sensor dengan 3 sumbu accelerometer dan 3 sumbu gyroscope digunakan untuk mendeteksi kegiatan para lansia. Penelitian ini menggunakan metode BackPropagation untuk mengenali gerakan jatuh kedepan, jatuh kesamping, jatuh kebelakang, duduk, tidur, jongkok, naik tangga, turun tangga dan sholat. Total data yang di uji adalah 720 data dari 16 partisipan. Pengenalan gerakan ini berdasarkan 10 variabel inputan dari data sensor accelerometer dan sensor gyroscope Untuk menghitung akurasi, presisi, sensitifitas dan spesifikasi pada penelitian ini menggunakan persamaan Receiver Operating Characteristic (ROC). Pengenalan gerakan dilakukan sebanyak 3 kali pada iterasi 10.000, 20.000 dan 30.000. Hasil akurasi rata-rata adalah 97.12%.
==============================================================================
Falling is an external aspect that can lead to death in the elderly. With so many activities they can do that will increase the likelihood of falling. A fall detection device is designed to minimize post-fall risk. A sensor with 3 Axis accelerometer and 3 gyroscope Axis is used to detect the activities of elder. This research uses BackPropagation method to recognize falling forward movement, falling aside, falling backward, sitting, sleeping, squatting, up stairs, down stairs and praying. Total data tested is 720 data from 16 participants. The introduction of this movement is based on 10 input variables from the accelerometer sensor data and gyroscope sensor. To calculate the accuracy, precision, sensitivity and specification in this research using Receiver Operating Characteristic (ROC) formula. The introduction of movement is done 3 times at 10.000, 20.000 and 30.000 iterations. The result of average accuracy is 97.12%.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Pendeteksi jatuh, Accelerometer, Gyroscope, BackPropagation. |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | Adlian Jefiza . |
| Date Deposited: | 19 Jan 2018 07:19 |
| Last Modified: | 06 Mar 2019 07:47 |
| URI: | http://repository.its.ac.id/id/eprint/44739 |
Actions (login required)
 |
View Item |