Peningkatan Resolusi Citra dengan Menggunakan Metode GAN untuk Aplikasi Peningkatan Gambar

Marniati Triningsi Tamo Ama(1*), Agustinus Rudatyo Himamunanto(2), Gogor Christmass Setyawan(3),

(1) Universitas Kristen Immanuel, Yogyakarta, Indonesia
(2) Universitas Kristen Immanuel, Yogyakarta, Indonesia
(3) Universitas Kristen Immanuel, Yogyakarta, Indonesia
(*) Corresponding Author

Abstract


This research proposes the use of a Generative Adversarial Network (GAN), a deep learning approach consisting of two neural networks: a generator that generates high-resolution images from low-resolution images, and a discriminator that distinguishes between original high-resolution images. and the image the generator produces. Through joint training, the generator learns to produce increasingly realistic and detailed images. This research uses training data of 400 image data, 100 images consisting of training data and test data. The GAN model trial showed a success rate of 80% training data, 20% test data. This process continued through repeated testing and 10,000 epoch training periods using Pytorch to train the GAN, with sharper and more detailed results than conventional methods. The application of GANs in various applications such as medical image processing, video restoration, and security surveillance shows great potential in improving image quality. Challenges such as training stability and computational time are overcome through more efficient regularization and optimization techniques, so that GANs prove to be a powerful tool for image resolution enhancement with a significant contribution to the development of more advanced image processing technologies.

Full Text:

PDF

References


Liu, Ming Yu, et al. “Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications.” Proceedings of the IEEE, vol. 109, no. 5, 2021, pp. 839–62, https://doi.org/10.1109/JPROC.2021.3049196.

Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2020). Generative adversarial networks. Communications of the ACM, 63(11), 139–144. https://doi.org/10.1145/3422622.

C. Ledig et al., “済無No Title No Title No Title,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 2017-Janua, no. 12, pp. 1–23, 2020, doi: 10.1007/s10994-023-06367-0.

P. Welander, S. Karlsson, and A. Eklund, “Generative Adversarial Networks for Image-to-Image Translation on Multi-Contrast MR Images - A Comparison of CycleGAN and UNIT,” 2018, [Online]. Available: http://arxiv.org/abs/1806.07777.

Ricky, M., & Al Rivan, M. E. (2022). Implementasi Deep Convolutional Generative Adversarial Network untuk Pewarnaan Citra Grayscale. Jurnal Teknik Informatika Dan Sistem Informasi, 8(3), 556–566. https://doi.org/10.28932/jutisi.v8i3.5218

Ledig, Christian, Lucas Theis, Ferenc Huszár, Jose Caballero, Andrew Cunningham, Alejandro Acosta, Andrew Aitken, et al. “済無No Title No Title No Title.” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 2017-Janua, no. 12, 2020, pp. 1–23, https://doi.org/10.1007/s10994-023-06367-0.

Qin, Chongli, et al. “Training Generative Adversarial Networks by Solving Ordinary Differential Equations.” Advances in Neural Information Processing Systems, vol. 2020-December, no. NeurIPS, 2020, pp. 1–19.

Kong, Jungil, et al. “HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis.” Advances in Neural Information Processing Systems, vol. 2020-December, no. NeurIPS, 2020.

C. Ledig et al., “href{https://ieeexplore.ieee.org/abstract/document/8099502}{Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network},” Cvpr, vol. 2, no. 3, p. 4, 2017, [Online]. Available: http://openaccess.thecvf.com/content_cvpr_2017/papers/Ledig_Photo-Realistic_Single_Image_CVPR_2017_paper.pdf

V. V. Pramansah, D. I. Mulyana, T. Silfia, and R. T. Jaya, “Penciptaan Karakter Anime Otomatis Dengan Menggunakan Generative Adversarial Networks,” J. Tek. Elektro dan Komputasi, vol. 4, no. 1,pp.2129,2022,[Online].Available:http://jurnal.unmuhjember.ac.id/index.php/ELKOM/article/view/7105.




DOI: https://doi.org/10.30645/kesatria.v5i3.447

DOI (PDF): https://doi.org/10.30645/kesatria.v5i3.447.g442

Refbacks

  • There are currently no refbacks.


Published Papers Indexed/Abstracted By: