[vc_empty_space][vc_empty_space]
Recognition of sign language hand gestures using leap motion sensor based on threshold and ANN models
Rusydi M.I.a, Syafiib, Hadelina R.c, Kimin E.d, Setiawan A.W.e, Rusydi A.f
a Department of Electrical Engineering, UniversitasAndalas, Padang, Indonesia
b Department of Electrical Engineering, UniversitasAndalas, Padang, Indonesia
c Department of Electrical Engineering, UniversitasAndalas, Padang, Indonesia
d Sekolah Tinggi Ilmu Kesehatan Mercubaktijaya, Padang, Indonesia
e School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung, Indonesia
f Department of Physics, National University of Singapore, Singapore
[vc_row][vc_column][vc_row_inner][vc_column_inner][vc_separator css=”.vc_custom_1624529070653{padding-top: 30px !important;padding-bottom: 30px !important;}”][/vc_column_inner][/vc_row_inner][vc_row_inner layout=”boxed”][vc_column_inner width=”3/4″ css=”.vc_custom_1624695412187{border-right-width: 1px !important;border-right-color: #dddddd !important;border-right-style: solid !important;border-radius: 1px !important;}”][vc_empty_space][megatron_heading title=”Abstract” size=”size-sm” text_align=”text-left”][vc_column_text]© 2020, Institute of Advanced Engineering and Science. All rights reserved.Hand gesture recognition is a topic that is still investigated by many scientists for numerous useful aspects. This research investigated hand gestures for sign language number zero to nine. The hand gesture recognition was based on finger direction patterns. The finger directions were detected by a Leap Motion Controller. Finger direction pattern modeling was based on two methods: Threshold and artificial neural network. Threshold model 1 contained 15 rules based on the range of finger directions on each axis. Threshold model 2 was developed from model 1 based on the behavior of finger movements when the subject performed hand gestures. The ANN model of the system was designed with four neurons at the output layer, 15 neurons at the input layer, seven neurons at the first hidden layer and 5 neurons at the second hidden layer. The artificial neural network used the logsig as the activation function. The result shows that the first threshold model has the lowest accuracy because the rule is too complicated and rigid. The threshold model 2 can improve the threshold model, but it still needs development to reach better accuracy. The ANN model gave the best result among the developed model with 98% accuracy. LMC produces useful biometric data for hand gesture recognition.[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Author keywords” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Indexed keywords” size=”size-sm” text_align=”text-left”][vc_column_text]ANN,Direction,Finger,Hand gesture,Threshold[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Funding details” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”DOI” size=”size-sm” text_align=”text-left”][vc_column_text]https://doi.org/10.11591/eei.v9i2.1194[/vc_column_text][/vc_column_inner][vc_column_inner width=”1/4″][vc_column_text]Widget Plumx[/vc_column_text][/vc_column_inner][/vc_row_inner][/vc_column][/vc_row][vc_row][vc_column][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][/vc_column][/vc_row]