Claim Missing Document
Check
Articles

Found 1 Documents
Search

Pengenalan Gesture Tangan Untuk Otomatisasi Switching Saklar Menggunakan Metode KNN Berbasis Raspberry Pi Misran Misran; Fitri Utaminingrum; Rizal Maulana
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 5 (2021): Mei 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Research on switching automation has been done a lot, both by using smartphone control, infrared sensors to using voice commands. The switching automation process used to turn lights on or off can also be done using hand gestures. By using the K-Nearest Neighbor method the computer can understand human interaction quite well using the method of decision making from existing patterns. In this study, the K-Nearest Neighbor method was used to translate hand signals or hand gestures into a command to control the LED. The test was carried out using 5 volunteers, each of whom tested each hand gesture given. To get the results of gesture recognition, there are several steps that must be taken, namely skin detection, preprocessing process, Feature Extraction, K-NN, and finally the system output. 3. The accuracy produced by the system is very good, where by conducting several experiments, the accuracy results obtained for five volunteers is 80%. Research on switching automation has been done a lot, both by using smartphone control, infrared sensors to using voice commands. The switching automation process used to turn lights on or off can also be done using hand gestures. By using the K-Nearest Neighbor method the computer can understand human interaction quite well using the method of decision making from existing patterns. In this study, the K-Nearest Neighbor method was used to translate hand signals or hand gestures into a command to control the LED. The test was carried out using 5 volunteers, each of whom tested each hand gesture given. To get the results of gesture recognition, there are several steps that must be taken, namely skin detection, preprocessing process, Feature Extraction, K-NN, and finally the system output. 3. The accuracy produced by the system is very good, where by conducting several experiments, the accuracy results obtained for five volunteers is 80%.