BibTex Citation Data :
@article{ROTASI54012, author = {Firman Isma Serdana and Akif Rahmatillah and Soegianto Soelistiono}, title = {EMG Based HCI Device to Support Computer Operation}, journal = {ROTASI}, volume = {25}, number = {2}, year = {2023}, keywords = {artifical neural network; electromyography; human computer interface; human interface device}, abstract = { The human computer interface is a method of interaction between a person and a computer that utilizes a human interface device. One example of this is using hand movements to control a computer pointer, which produces a unique electromyography signal for each basic movement direction (up, down, right, and left). This project utilized an artificial neural network with a structure of seven inputs, ten hidden layer nodes, and four outputs to classify electromyography signals from the brachioradialis and flexor carpum ulnaris muscles into four basic movement categories within an Arduino Uno. The artificial neural network was trained offline using a high-capacity machine for efficiency since the Arduino Uno has low raw processing capability. The Sparkfun Pro Micro's HID function and the mouse.move() library were used to translate the classification results into pointer movement on a PC. The classification rate for the prerequisites setting resulted in an average of 93.7375%, with individual classification rates of 96.55% for up movement, 93.4% for down movement, 90.85% for right movement, and 91.95% for left movement. }, issn = {2406-9620}, pages = {30--36} doi = {10.14710/rotasi.25.2.30-36}, url = {https://ejournal.undip.ac.id/index.php/rotasi/article/view/54012} }
Refworks Citation Data :
The human computer interface is a method of interaction between a person and a computer that utilizes a human interface device. One example of this is using hand movements to control a computer pointer, which produces a unique electromyography signal for each basic movement direction (up, down, right, and left). This project utilized an artificial neural network with a structure of seven inputs, ten hidden layer nodes, and four outputs to classify electromyography signals from the brachioradialis and flexor carpum ulnaris muscles into four basic movement categories within an Arduino Uno. The artificial neural network was trained offline using a high-capacity machine for efficiency since the Arduino Uno has low raw processing capability. The Sparkfun Pro Micro's HID function and the mouse.move() library were used to translate the classification results into pointer movement on a PC. The classification rate for the prerequisites setting resulted in an average of 93.7375%, with individual classification rates of 96.55% for up movement, 93.4% for down movement, 90.85% for right movement, and 91.95% for left movement.
Article Metrics:
Last update:
Last update: 2024-11-03 23:53:43
Penerbit: Departemen Teknik Mesin, Fakultas Teknik, Universitas Diponegoro
Alamat Redaksi: Gedung Administrasi Lantai II Departemen Teknik Mesin Fakultas Teknik Undip Telp.(024)7460059, Facsimile: (024)7460059 ext.102 Email: rotasi@live.undip.ac.id