DETEKSI MASKER WAJAH MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN) MENINGKATKAN NILAI AKURASI MELALUI ARSITEKTUR LAYER KONVOLUSI
DOI:
https://doi.org/10.33143/jics.Vol8.Iss1.2123Abstrak
Abstrak— Di Era pandemi covid-19 seperti yang dirasakan semenjak Tahun 2019 memiliki dampak yang sangat signifikan diberbagai bidang kehidupan. Hal ini menjadi fokus pemerintah dalam memulihkan berbagai masalah khususnya kesehatan. Virus covid-19 menjadi momok menakutkan untuk dihadapi yang kian mempengaruhi psikologis. Karena penyebaran begitu cepatnya dari satu daerah bahkan sekarang dunia dan tentunya menyababkan vatalitas kematian yang terus meningkat. Oleh karena itu, pemerintah Indonesia melalui satgas covid-19 dan jajarannya terus mengkampanyekan protokol kesehatan khususnya penggunaan masker kesehatan. Pada penelitian ini berfokus pada penggunaan teknologi untuk mengatasi permasalah penegakkan aturan protokol kesehatan tersebut melalui meneliti penggunaan metode deteksi masker untuk keperluan tertentu. Penelitian ini menghasilkan hasil riset teknologi pendeteksi masker (masked detection) menggunakan metode CNN yang merupakan salah satu metode dari cabang computer vision. Hasil yang diperoleh dari dataset 7500 citra training dengan epoch 20 adalah akurasi sebesar 0.92287 dari loss sebesar 0.18338. Pengujian dari akurasi ini sangat baik sehingga pendeteksian pada citra diberbagai sudut pengambilan menghasilkan false negatif yang minim.
Kata kunci: Deteksi masker, covid-19, convolutional neural netowrk, deep learning, kesehatan
Abstract—In the era of the COVID-19 pandemic, as it has been felt since 2019, it has had a very significant impact in various areas of life. This has become the focus of the government in restoring various problems, especially health. The Covid-19 virus is a frightening specter to face which is increasingly affecting psychologically. Because the spread is so fast from one area to even now the world and of course causes the mortality rate to continue to increase. Therefore, the Indonesian government through the COVID-19 task force and its staff continues to campaign for health protocols, especially the use of health masks. This study focuses on the use of technology to overcome the problem of enforcing the health protocol rules through examining the use of mask detection methods for certain purposes. This research produces the results of research on mask detection technology (masked detection) using the CNN method which is one of the methods of the computer vision branch. The results obtained from the 7500 training image dataset with 20 epochs is an accuracy of 0.92287 from a loss of 0.18338. Testing of this accuracy is very good so that detection of images at various angles of capture produces minimal false negatives
Keywords: masked detection, covid-19, convolutional neural network, deep learning, health
Referensi
WHO, 2020. Anjurnan mengenai penggunaan masker dalam konteks covid-19. World Heath Organization 2020.
Keputusan Gubernur Aceh, 2020. Pembentukan Satuan Tugas Penanganan Corona Virus Disease 2019 Aceh.
Shereen, M. A., Khan, S., Kazmi, A., Bashir, N., & Siddique, R. (2020). COVID19 infection: Origin, transmission, and characteristics of human coronaviruses. In Journal of Advanced Research (Vol. 24)
Yesudhas, D., Srivastava, A., & Gromiha, M. M. (2021). COVID-19 outbreak: history, mechanism, transmission, structural studies and therapeutics. In Infection (Vol. 49, Issue 2)
Buku: Hermawati, FA, “Pengolahan citra digital Konsep & teoriâ€, 2013, Penerbit Andi
Dwipayana, M., Arnia, F., & Musliyana, Z. (2018). Histogram equalization smoothing for determining threshold accuracy on ancient document image binarization. Journal of Physics: Conference Series, 1019(1).
Sulistyowati, A., Hariyani, Y. S., & Novianti, A. (2018). Perancangan Aplikasi Pembaca Warna dan Bentuk Berbasis Pengolahan Citra untuk Daftar Katalog Perpustakaan. E-Proceeding of Applied Science, 4(3).
Wira, J. Putra G, “Pengenalan Konsep Pembelajaran Mesin dan Deep Learningâ€, Edisi 1.4, 2020.
Gu, J., Wang, Z., Kuen, J., Ma, L., Shahroudy, A., Shuai, B., Liu, T., Wang, X., Wang, G., Cai, J., & Chen, T. (2018). Recent advances in convolutional neural networks. Pattern Recognition, 77.
Unduhan
Terbitan
Bagian
Lisensi
COPYRIGHT TRANSFER FORM
The copyright to this article is transferred to Universitas Ubudiyah Indonesia (UUI) if and when the article is accepted for publication. The undersigned hereby transfers any and all rights in and to the paper including without limitation all copyrights to UUI. The undersigned hereby represents and warrants that the paper is original and that he/she is the author of the paper, except for material that is clearly identified as to its original source, with permission notices from the copyright owners where required. The undersigned represents that he/she has the power and authority to make and execute this assignment.
We declare that:
- This paper has not been published in the same form elsewhere.
- It will not be submitted anywhere else for publication prior to acceptance/rejection by this Journal.
- A copyright permission is obtained for materials published elsewhere and which require this permission for reproduction.
Furthermore, I/We hereby transfer the unlimited rights of publication of the above mentioned paper in whole to UUI. The copyright transfer covers the right to reproduce and distribute the article, including reprints, translations, photographic reproductions, microform, electronic form (offline, online) or any other reproductions of similar nature. The corresponding author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. After submission of this agreement signed by the corresponding author, changes of authorship or in the order of the authors listed will not be accepted.
Retained Rights/Terms and Conditions
- Authors retain all proprietary rights in any process, procedure, or article of manufacture described in the work.
- Authors may reproduce or authorize others to reproduce the work or derivative works for the author’s personal use or for company use, provided that the source and the UUI copyright notice are indicated, the copies are not used in any way that implies UUI endorsement of a product or service of any employer, and the copies themselves are not offered for sale.
- Although authors are permitted to re-use all or portions of the work in other works, this does not include granting third-party requests for reprinting, republishing, or other types of re-use.
.png)
