Implementasi ESPCN untuk Meningkatkan Kualitas Foto dan Akurasi Model Klasifikasi Menggunakan CNN
DOI:
https://doi.org/10.61769/telematika.v18i1.559Keywords:
CNN, photo quality, super resolution, accuration, OpenCVAbstract
In conducting research, there are often obstacles to supporting media to support the observations and experiments under study. Especially in the case of research involving photos, not a few quality problems that use cameras show results that are not ideal, such as dimness, disturbing-colored dots, or other disturbances. In line with the rapid development of technology today, these problems can be overcome by computer programming through the Opensource Computer Vision Library (OpenCV). OpenCV is a programming module that contains various features, one of which is improving image quality with super-resolution. In practice, photos that have low quality will be enhanced using the efficient subpixel convolutional neural network (ESPCN) model. The deep learning algorithm used is a convolutional neural network (CNN) to support the testing means. CNN works to obtain the percentage accuracy of the photos under study as a representation of the final test results. This test aims to improve the low quality of photos with the ESPCN model to compare the accuracy with the original photos. The test result is the application of ESPCN to low-quality photos. The test result is higher accuracy than the original photo with a difference of 1.2%. The original photo had an accuracy of 90.6%, while the enhanced photo had an accuracy of 91.8%. The final result shows that low-quality photos can be upscaled using ESPCN to produce better accuracy.
References
S. Anwar, S. Khan, dan N. Barnes, “A deep journey into super-resolution: a survey,” ACM Comput. Surv., vol. 53, no. 3, hlm. 1–21, 2020, doi: 10.1145/3390462.
Z. Chu, dkk., “A generalizable smple resolution augmentation method for mechanical fault diagnosis based on ESPCN,” J. Sensors, vol. 2021, 2021, doi: 10.1155/2021/7496007.
P. N. Michelini, Y. Lu, dan X. Jiang, “Edge-SR: super-resolution for the masses,” Pros. - 2022 IEEE/CVF Winter Conf. Appl. Comput. Vision, WACV 2022, hlm. 4019–4028, 2022, doi: 10.1109/WACV51458.2022.00407.
S. Zhang, G. Liang, S. Pan, dan L. Zheng, “A fast medical image super resolution method based on deep learning network,” IEEE Access, vol. 7, hlm. 12319–12327, 2019, doi: 10.1109/ACCESS.2018.2871626.
R. Ravikumar dan V. Arulmozhi, “Digital image processing-a quick review,” Int. J. Intell. Comput. Technol., vol. 2, no. 2, hlm. 16–24, 2019.
F. F. Maulana dan N. Rochmawati, “Klasifikasi citra buah menggunakan convolutional neural network,” J. Informatics Comput. Sci., vol. 1, no. 02, hlm. 104–108, 2020, doi: 10.26740/jinacs.v1n02.p104-108.
C. Chazar dan B. Erawan, “Machine learning diagnosis kanker payudara menggunakan algoritme support vector machine,” Inf. (Jurnal Inform. dan Sist. Informasi), vol. 12, no. 1, hlm. 67–80, 2020, doi: 10.37424/informasi.v12i1.48.
N. Giarsyani, A. F. Hidayatullah, dan R. Rahmadi, “Komparasi algoritme machine learning dan deep learning untuk named entity recognition (studi kasus: data kebencanaan),” J. Inform. Rekayasa Elektron., vol. 3, no. 1, hlm. 48–57, 2020, doi: 10.20961/ijai.v4i2.41317.
B. W. Kurniadi, H. Prasetyo, G. L. Ahmad, B. Aditya Wibisono, dan D. Sandya Prasvita, “Analisis perbandingan algoritme SVM dan CNN untuk klasifikasi buah,” dalam Seminar Nasional Mahasiswa Ilmu Komputer dan Aplikasinya (Senamika), 2021, vol. 2, no. 2, hlm. 1–11.
S. Sinaga, “Pengaruh motivasi dan pengalaman kerja terhadap produktivitas kerja karyawan pada PT Trikarya Cemerlang Medan,” J. Ilm. Metadata, vol. 2, no. 2, hlm. 159–169, 2020, doi: 10.47652/metadata.v2i2.28.
Y. Fitrianto, Dasar-Dasar Digital Imaging. Semarang: Yayasan Prima Agus Teknik, 2021.
A. Stanley, "Penerapan kode Huffman dalam proses kompresi dan dekompresi format gambar JPEG", Makalah IF2120 Matematika Diskrit-Sem. I Tahun 2018, 2018.
X. Hou, dkk., “Learning based image transformation using convolutional neural networks,” IEEE Access, vol. 6, hlm. 49779–49792, 2018, doi: 10.1109/ACCESS.2018.2868733.
T. R. Ravi, K. Kuhikar, Y. Kalambe, A. Wagh, dan U. Deshkar, “Well detection using image processing,” J. Res. Eng. Appl. Sci., vol. 5, no. 4, hlm. 142–145, 2020, doi: 10.46565/jreas.2020.v05i04.004.
H. Adusumalli, D. Kalyani, dan R. K. Sri, “Face mask detection using OpenCV,” dalam 2021 Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV), 2021, hlm. 1304–1309. doi: 10.1109/ICICV50876.2021.9388375.
B. Hardiansyah, A. P. Armin, dan A. B. Yunanda, “Rekonstruksi citra pada super resolusi menggunakan interpolasi bicubic,” Integer J. Inf. Technol., vol. 4, no. 2, hlm. 1–12, 2019, doi: 10.31284/j.integer.2019.v4i2.684.
B. Garber, A. Grossman, dan S. Johnson-Yu, “Image Super-Resolution via a Convolutional Neural Network,” 2020.
P. A. Nugroho, I. Fenriana, dan R. Arijanto, “Implementasi deep learning menggunakan convolutional neural network (CNN) pada ekspresi manusia,” J. Algor Data Syst., vol. 2, no. 1, hlm. 12–21, 2020, doi: 10.31253/algor.v2i1.
A. A. Sulaeman dan A. Susilo, “Implementasi convolutional neural network untuk klasifikasi Covid-19,” Sigma- J. Teknol. Pelita Bangsa, vol. 13, no. 2, hlm. 113–118, 2022.
D. Efendi, J. Jasril, S. Sanjaya, F. Syafria, dan E. Budianita, “Penerapan algoritme convolutional neural network arsitektur ResNet-50 untuk klasifikasi citra daging sapi dan babi,” Jurikom (Jurnal Ris. Komputer), vol. 9, no. 3, hlm. 607–614, 2022, doi: 10.30865/jurikom.v9i3.4176.
P. Zhou, J. Feng, C. Ma, C. Xiong, S. Hoi, dan E. Weinan, “Towards theoretically understanding why SGD generalizes better than ADAM in deep learning,” dalam 34th Conference on Neural Information Processing Systems (NeurIPS 2020), 2020, vol. 2020-Decem, no. 1, hlm. 19–21.
Downloads
Published
Issue
Section
License
Copyright (c) 2023 Andre Daegal, Rianto Rianto
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation.
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
