Pengenalan Angka Sistem Isyarat Bahasa Indonesia Dengan Menggunakan Metode Convolutional Neural Network

Authors

  • Mochamad Bagus Setiyo Bakti Teknologi Informasi, Sekolah Tinggi Teknik Surabaya Indonesia
  • Yuliana Melita Pranoto Teknologi Informasi, Sekolah Tinggi Teknik Surabaya Indonesia

DOI:

https://doi.org/10.29407/inotek.v3i1.504

Abstract

Bahasa isyarat adalah metode komunikasi dengan menggunakan gerakan tangan yang biasanya dilakukan oleh tuna rungu. Di Indonesia sendiri mempunyai 2 jenis bahasa isyarat yaitu BISINDO dan SIBI (Sistem Isyarat Bahasa Indonesia). Perbedaannya dalam isyarat dasar BISINDO menggunakan dua tangan sedangkan SIBI hanya satu tangan. Dalam paper ini akan membahas pengenalan isyarat angka SIBI dengan menggunakan metode Convolutional  Neural Network  (CNN).  Arsitektur CNN yang  digunakan adalah arsitektur  LeNet.  Arsitektur  CNN diproses dalam 3 tahap, 25 epoch, 50 epoch dan 100 epoch. Berdasarkan percobaan yang dilakukan nilai akurasi yang didapat terus meningkat dalam tiap tahapnya, mulai 67.66%, 89.44% sampai nilai akurasi tertinggi dalam proses training sebesar 96.44%. Begitupun dalam proses prediksi data juga mengalami kenaikan dalam tiap tahapnya, mulai 79.23%, 90.45% sampai didapatkan nilai akurasi tertinggi dalam prediksi data 98.89%. Dari total 90 data testingset, hanya sekali keasalahan dalam prediksi data.

Keywords:

SIBI, CNN, LeNet, Machine Learning

##plugins.themes.default.displayStats.downloads##

##plugins.themes.default.displayStats.noStats##

References

Sulfayanti, Dewiani, and Armin Lawi, “A real time alphabets sign language recognition system using hands tracking,” 2016 Int. Conf. Comput. Intell. Cybern., pp. 69–72, 2016.

C. M. Jin, Z. Omar, and M. H. Jaward, “A mobile application of American sign language translation via gambar processing algorithms,” 2016 IEEE Reg. 10 Symp., pp. 104–109, 2016.

T. Dharamsi, R. Jawahar, K. Mahesh, and G. Srinivasa, “Stringing Subtitles in Sign Language,” 2016 IEEE Eighth Int. Conf. Technol. Educ., pp. 228–231, 2016.

I. Kautsar, R. I. Borman, and A. Sulistyawati, “Aplikasi pembelajaran bahasa isyarat bagi penyandang tuna rungu berbasis android dengan metode bisindo,” Semin. Nas. Teknol. Inf. dan Multimed. 2015, vol. 3, no. 2003, pp. 6–8, 2015.

Menteri Pendidikan dan Kebudayaan Republik Indonesia "Pembakuan Sistem Isyarat Bahasa Indonesia Bagi Kaum Tuna Rungu" No 0161/U/1994, 1994.

A. Cordiner, “Illumination invariant face detection,” MComSc thesis, Univ. Wollongong, p. 175, 2009.

S. Yang and Q. Zhu, “Video-based Chinese sign language recognition using convolutional neural network,” 2017 9th IEEE Int. Conf. Commun. Softw. Networks, ICCSN 2017, vol. 2017– January, pp. 929–934, 2017.

D. Chai and K. N. Ngan, “Locating facial region of a head-and-shoulders color gambar,” Proc. - 3rd IEEE Int. Conf. Autom. Face Gesture Recognition, FG 1998, pp. 124–129, 1998.

Y. Xie, H. Jin, and E. C. C. Tsang, “Improving the lenet with batch normalization and online hard example mining for digits recognition,” Int. Conf. Wavelet Anal. Pattern Recognit., vol. 1, pp. 149–153, 2017.

G. Antonellis, A. G. Gavras, M. Panagiotou, B. L. Kutter, G. Guerrini, A. C. Sander, and P. J. Fox, “Shake Table Test of Large-Scale Bridge Columns Supported on Rocking Shallow Foundations,” J. Geotech. Geoenvironmental Eng., vol. 12, p. 04015009, 2015.

Mattew D. Zeiller " Adadelta : An AdaptiveLearning Rate Mehod” Cornell University, 2012.

CNN Indonesia, "Masih Ada 40,5 Juta Orang Indonesia Berjuang Lawan Ketulian", 2017. avaible : https://www.cnnindonesia.com/gaya- hidup/20170323180138-255-202335/masih-ada- 405-juta-orang-indonesia-berjuang-lawan-ketulian, diakses 25 Agustus 2018.

Downloads

Published

2021-03-07

Issue

Vol. 3 No. 1 (2019): PROSIDING SEMNAS INOTEK Ke-III Tahun 2019

Section

Articles

License

Copyright (c) 2020 Mochamad Bagus Setiyo Bakti, Yuliana Melita Pranoto

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Authors who publish with this journal agree to the following terms:

  1. Copyright on any article is retained by the author(s).
  2. The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
  3. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
  4. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
  5. The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License

How to Cite

Pengenalan Angka Sistem Isyarat Bahasa Indonesia Dengan Menggunakan Metode Convolutional Neural Network. (2021). Prosiding SEMNAS INOTEK (Seminar Nasional Inovasi Teknologi), 3(1), 011-016. https://doi.org/10.29407/inotek.v3i1.504