Sistem Pendeteksian Bahasa Isyarat Alfabet Sibi Menggunakan Algoritma You Only Look Once Dan Roboflow
Abstract
Sign language is a language used by deaf people to communicate with others. The lack of access to information and communication for people with disabilities, especially the deaf, encourages the development of solutions that are able to translate sign language into text and speech form. Seeing these problems and gaps, this research was conducted to develop a SIBI alphabet sign language translation system using the YOLOv8 algorithm. The YOLOv8 algorithm is used to recognize and translate SIBI language characters into Indonesian text. The dataset used consists of images taken manually as well as those sourced from the internet. The test results show that the model trained with 80:10:10 split dataset achieves the highest accuracy, with mAP50 value of 0.99 and mAP50-95 value of 0.91. Besides that, the SGD optimizer provides the best performance compared to other optimizers such as Adam and RMSprop. The sign language translation system as a product in this research is expected to improve the accessibility of deaf people and sign language learning in Indonesia.
Keywords: machine learning, optimizer, sign language, SIBI, YOLOv8
References
J. K. Hikmalansya dan D. Cahyono. “Aplikasi pembelajaran bahasa isyarat berbasis android.” Jurnal INFORM, vol. 1, no. 2, pp. 118-124, Jul. 2016, doi: 10.25139/inform.v1i2.849. [Online]. Available: https://ejournal.unitomo.ac.id/index.php/inform/article/view/849
D. Naren, “Aksinya viral saat debat pilpres 2024, ternyata segini gaji fantastis juru bicara isyarat, tertarik?” Ayobandung.com, 9 Jan. 2024. [Online]. Available: https://www.ayobandung.com/umum/7911457619/aksinya-viral-saat-debat-pilpres-2024-ternyata-segini-gaji-fantastis-juru-bicara-isyarat-tertarik
S. Shania, M. F. Naufal, V. R. Prasetyo, dan M. S. Bin Azmi, “Translator of Indonesian sign language video using convolutional neural network with transfer learning,” Indonesian Journal of Information Systems, vol. 5, no. 1, pp. 17-27, Agu. 2022, doi: 10.24002/ijis.v5i1.5865. [Online]. Available: https://api.semanticscholar.org/CorpusID:257295702
N. Nurrahma, R. Yusuf, dan A. S. Prihatmanto. “Indonesian sign language fingerspelling recognition using vision-based machine learning,” in 2021 International Conf. on Intelligent Cybernetics Technology & Applications (ICICyTA), Des. 1-2, 2021. hlm. 28–33, 2021. [Online]. Available:
https://api.semanticscholar.org/CorpusID:246533726
Direktorat Pendidikan Masyarakat dan Pendidikan Khusus. (30 Des. 2020). Kamus SIBI Kerjasama antara Kementerian Pendidikan dan Kebudayaan dengan Lembaga Penelitian dan Pengembangan Sistem Isyarat Bahasa Indonesia. [Online]. Available: https://pmpk.kemdikbud.go.id/sibi/
M. Alaftekin, I. Pacal, dan K. Cicek. “Real-time sign language recognition based on YOLO algorithm,” Neural Comput Appl, vol. 36, no. 14, pp. 7609–7624, Feb. 2024, doi: 10.1007/s00521-024-09503-6. [Online]. Available: https://link.springer.com/article/10.1007/s00521-024-09503-6
Ultralytics, “Ultraltics YOLOv8.” Ultralytics YOLOv8 Docs, Accessed: 14 Agu. 2024. [Online].
L. Shen, B. Lang, dan Z. Song, “DS-YOLOv8-based object detection method for remote sensing images,” IEEE Access, vol. 11, pp. 125122–125137, Nov. 2023, doi: 10.1109/ACCESS.2023.3330844. [Online]. Available: https://ieeexplore.ieee.org/iel7/6287639/6514899/10310143.pdf
NVIDIA. “NVIDIA-SMI,” NVIDIA. Accessed: [Online]. Available: www.nvidia.com
N. A. Ahmed, “What is a confusion matrix in machine learning? the model evaluation tool explained.” Datacamp, Accessed: 14 Agu. 2024. [Online]. Available: https://www.datacamp.com/tutorial/what-is-a-confusion-matrix-in-machine-learning
J. Solawetz, “What is mean average precision (mAP) in object detection?” roboflow, Accessed: 14 Agu. 2024. [Online]. Available: https://blog.roboflow.com/mean-average-precision/#:~:text=Mean%20Average%20Precision%20(mAP)%20is%20used%20to%20measure%20the%20performance,versions%20of%20the%20same%20model.
Downloads
Published
Versions
- 2025-11-05 (2)
- 2025-11-05 (1)
