PEMULIHAN CITRA AKIBAT DEGRADASI DIGITAL MENGGUNAKAN PIX2PIX GENERATIVE ADVERSARIAL NETWORKSS (GAN)
DOI:
https://doi.org/10.70248/jrsit.v3i1.2959Abstract
Kerusakan digital pada citra dapat menghilangkan detail penting dan menurunkan kualitas visual. Penelitian ini bertujuan mengembangkan sistem restorasi citra berbasis Pix2Pix Generative Adversarial Networkss (GAN) dengan U-Net sebagai Generator dan PatchGAN sebagai Discriminator. Dataset yang digunakan adalah CelebA-HQ beresolusi 256×256 pixel yang dimodifikasi menjadi tiga tingkat kerusakan yaitu ringan, sedang, dan berat. Proses pelatihan dilakukan selama 100 epoch menggunakan optimizer Adam dengan penyesuaian rasio pembaruan Generator–Discriminator (1:1, 2:1, dan 3:1) untuk menjaga stabilitas model. Evaluasi dilakukan menggunakan Structural Similarity Index Measure (SSIM) dan Peak Signal-to-Noise Ratio (PSNR). Hasil menunjukkan model mampu mencapai SSIM 0,9226 dan PSNR 32,68 dB, yang menandakan keberhasilan dalam mempertahankan struktur spasial serta meningkatkan kualitas visual citra hasil restorasi. Pengujian pada tiga kategori kerusakan menunjukkan tingkat keberhasilan rata-rata di atas 90% pada kerusakan ringan dan sedang, serta 85–90% pada kerusakan berat, meskipun detail halus seperti hidung dan rambut belum sepenuhnya sempurna. Dibandingkan metode konvensional, pendekatan GAN terbukti lebih efektif dalam menghasilkan citra yang realistis dan natural. Selain itu, sistem diimplementasikan dalam bentuk aplikasi web yang memudahkan pengguna melakukan unggah, restorasi, dan unduh citra secara praktis. Dengan demikian, penelitian ini menyimpulkan bahwa pendekatan Pix2Pix berbasis GAN efektif untuk pemulihan citra digital dan memiliki potensi untuk dikembangkan lebih lanjut pada berbagai domain citra termasuk medis maupun arsip sejarah digital.
References
L. Galteri, L. Seidenari, M. Bertini, and A. Del Bimbo, “Deep Universal Generative Adversarial Compression Artifact Removal,” IEEE Trans. Multimed., vol. 21, no. 8, pp. 2131–2145, 2019, doi: 10.1109/TMM.2019.2895280.
K. Zhang, W. Zuo, L. Zhang, and D. Computing, “Deep plug-and-play super-resolution for arbitrary blur kernels,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2019-June, pp. 1671–1681, 2019, doi: 10.1109/CVPR.2019.00177.
S. Karavarsamis, I. Gkika, V. Gkitsas, K. Konstantoudakis, and D. Zarpalas, “A Survey of Deep Learning-Based Image Restoration Methods for Enhancing Situational Awareness at Disaster Sites: The Cases of Rain, Snow and Haze,” Sensors, vol. 22, no. 13, 2022, doi: 10.3390/s22134707.
M. Triningsi, T. Ama, A. R. Himamunanto, C. Setyawan, P. S. Informatika, and U. K. Immanuel, “Peningkatan Resolusi Citra dengan Menggunakan Metode GAN untuk Aplikasi Peningkatan Gambar,” vol. 5, no. 3, pp. 1287–1296, 2024.
S. Gotamaa and I. W. S. Santiyasaa, “Image Restoration Menggunakan Metode Lucy-Richardson Pada Citra Underwater,” Resolusi Rekayasa Tek. Inform. dan Inf., vol. 1, no. 3, pp. 179–187, 2021, [Online]. Available: https://djournals.com/resolusi/article/view/128
M. Cao, H. Feng, H. Xiao, S. Computer, and S. Information Technology, “An Improved GAN-Based Image Restoration Method for Imaging Logging Images,” Appl. Sci., vol. 13, no. 16, 2023, doi: 10.3390/app13169249.
I. Goodfellow et al., “Generative Adversarial Networkss,” Commun. ACM, vol. 63, no. 11, pp. 139–144, 2020, doi: 10.1145/3422622.
M. Maharani and T. A. Salim, “Strategi Pemerintah menggunakan AI : Studi Kasus Restorasi Warisan Budaya ( Koleksi Foto Prajurit ) untuk Peringatan 70 Tahun Gencatan Senjata Korea Selatan Government Strategies Using AI : A Case Study of Cultural Heritage Restoration ( Soldier Photo Col,” vol. 15, no. 2, pp. 117–130, 2024.
D. Caruso et al., “Low-dose liver CT: image quality and diagnostic accuracy of deep learning image reconstruction algorithm,” Eur. Radiol., vol. 34, no. 4, pp. 2384–2393, 2024, doi: 10.1007/s00330-023-10171-8.
P. Salehi and A. Chalechale, “Pix2Pix-based Stain-to-Stain Translation: A Solution for Robust Stain Normalization in Histopathology Images Analysis,” Iran. Conf. Mach. Vis. Image Process. MVIP, vol. 2020-Febru, 2020, doi: 10.1109/MVIP49855.2020.9116895.
W. Weng and X. Zhu, “INet: Convolutional Networks for Biomedical Image Segmentation,” IEEE Access, vol. 9, pp. 16591–16603, 2021, doi: 10.1109/ACCESS.2021.3053408.
V. Kushwaha, P. Shukla, and G. C. Nandi, “Generating quality grasp rectangle using Pix2Pix GAN for intelligent robot grasping,” Mach. Vis. Appl., vol. 34, no. 1, 2023, doi: 10.1007/s00138-022-01362-2.
M. Liu, X. Su, X. Yao, W. Hao, and W. Zhu, “Lensless Image Restoration Based on Multi-Stage Deep Neural Networks and Pix2pix Architecture,” Photonics, vol. 10, no. 11, 2023, doi: 10.3390/photonics10111274.
Y. He, D. Zhao, Z. Ren, H. Qin, and P. Jiang, “基于改进 pix2pix 框架的人脸图像修复模型,” 2024.
A. Sriram, A. Vatsa, A. Kumar, S. Vats, and A. Kumar, “Challenges and Opportunities in Malignant Image Reconstruction Using GAN: A Review,” 2024 IEEE Integr. STEM Educ. Conf. ISEC 2024, no. March, 2024, doi: 10.1109/ISEC61299.2024.10664911.
H. Hoyez, C. Schockaert, J. Rambach, B. Mirbach, and D. Stricker, “Unsupervised Image-to-Image Translation : A Review,” pp. 1–27, 2022.
M. Trigka and E. Dritsas, “A Comprehensive Survey of Deep Learning Approaches in Image Processing,” Sensors, vol. 25, no. 2, 2025, doi: 10.3390/s25020531.
A. Maulana, C. Fatichah, N. Suciati, D. Teknik Informatika, and N. Surabaya, “Facial Inpainting Pada Citra Wajah Unaligned Menggunakan Generative Adversarial Networks Dengan Feature Reconstruction Loss,” JUTI J. Ilm. Teknol. Inf., vol. 18, no. 2, pp. 171–178, 2020.
X. Wang, L. Sun, A. Chehri, and Y. Song, “A Review of GAN-Based Super-Resolution Reconstruction for Optical Remote Sensing Images,” Remote Sens., vol. 15, no. 20, pp. 1–34, 2023, doi: 10.3390/rs15205062.
K. Liu, J. Li, and S. S. Hussain Bukhari, “Overview of Image Inpainting and Forensic Technology,” Secur. Commun. Networks, vol. 2022, 2022, doi: 10.1155/2022/9291971.
L. Galteri, L. Seidenari, M. Bertini, and A. Del Bimbo, “Deep Universal Generative Adversarial Compression Artifact Removal,” IEEE Trans. Multimed., vol. 21, no. 8, pp. 2131–2145, 2019, doi: 10.1109/TMM.2019.2895280.
K. Zhang, W. Zuo, L. Zhang, and D. Computing, “Deep plug-and-play super-resolution for arbitrary blur kernels,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2019-June, pp. 1671–1681, 2019, doi: 10.1109/CVPR.2019.00177.
S. Karavarsamis, I. Gkika, V. Gkitsas, K. Konstantoudakis, and D. Zarpalas, “A Survey of Deep Learning-Based Image Restoration Methods for Enhancing Situational Awareness at Disaster Sites: The Cases of Rain, Snow and Haze,” Sensors, vol. 22, no. 13, 2022, doi: 10.3390/s22134707.
M. Triningsi, T. Ama, A. R. Himamunanto, C. Setyawan, P. S. Informatika, and U. K. Immanuel, “Peningkatan Resolusi Citra dengan Menggunakan Metode GAN untuk Aplikasi Peningkatan Gambar,” vol. 5, no. 3, pp. 1287–1296, 2024.
S. Gotamaa and I. W. S. Santiyasaa, “Image Restoration Menggunakan Metode Lucy-Richardson Pada Citra Underwater,” Resolusi Rekayasa Tek. Inform. dan Inf., vol. 1, no. 3, pp. 179–187, 2021, [Online]. Available: https://djournals.com/resolusi/article/view/128
M. Cao, H. Feng, H. Xiao, S. Computer, and S. Information Technology, “An Improved GAN-Based Image Restoration Method for Imaging Logging Images,” Appl. Sci., vol. 13, no. 16, 2023, doi: 10.3390/app13169249.
I. Goodfellow et al., “Generative Adversarial Networkss,” Commun. ACM, vol. 63, no. 11, pp. 139–144, 2020, doi: 10.1145/3422622.
M. Maharani and T. A. Salim, “Strategi Pemerintah menggunakan AI : Studi Kasus Restorasi Warisan Budaya ( Koleksi Foto Prajurit ) untuk Peringatan 70 Tahun Gencatan Senjata Korea Selatan Government Strategies Using AI : A Case Study of Cultural Heritage Restoration ( Soldier Photo Col,” vol. 15, no. 2, pp. 117–130, 2024.
D. Caruso et al., “Low-dose liver CT: image quality and diagnostic accuracy of deep learning image reconstruction algorithm,” Eur. Radiol., vol. 34, no. 4, pp. 2384–2393, 2024, doi: 10.1007/s00330-023-10171-8.
P. Salehi and A. Chalechale, “Pix2Pix-based Stain-to-Stain Translation: A Solution for Robust Stain Normalization in Histopathology Images Analysis,” Iran. Conf. Mach. Vis. Image Process. MVIP, vol. 2020-Febru, 2020, doi: 10.1109/MVIP49855.2020.9116895.
W. Weng and X. Zhu, “INet: Convolutional Networks for Biomedical Image Segmentation,” IEEE Access, vol. 9, pp. 16591–16603, 2021, doi: 10.1109/ACCESS.2021.3053408.
V. Kushwaha, P. Shukla, and G. C. Nandi, “Generating quality grasp rectangle using Pix2Pix GAN for intelligent robot grasping,” Mach. Vis. Appl., vol. 34, no. 1, 2023, doi: 10.1007/s00138-022-01362-2.
M. Liu, X. Su, X. Yao, W. Hao, and W. Zhu, “Lensless Image Restoration Based on Multi-Stage Deep Neural Networks and Pix2pix Architecture,” Photonics, vol. 10, no. 11, 2023, doi: 10.3390/photonics10111274.
Y. He, D. Zhao, Z. Ren, H. Qin, and P. Jiang, “基于改进 pix2pix 框架的人脸图像修复模型,” 2024.
A. Sriram, A. Vatsa, A. Kumar, S. Vats, and A. Kumar, “Challenges and Opportunities in Malignant Image Reconstruction Using GAN: A Review,” 2024 IEEE Integr. STEM Educ. Conf. ISEC 2024, no. March, 2024, doi: 10.1109/ISEC61299.2024.10664911.
H. Hoyez, C. Schockaert, J. Rambach, B. Mirbach, and D. Stricker, “Unsupervised Image-to-Image Translation : A Review,” pp. 1–27, 2022.
M. Trigka and E. Dritsas, “A Comprehensive Survey of Deep Learning Approaches in Image Processing,” Sensors, vol. 25, no. 2, 2025, doi: 10.3390/s25020531.
A. Maulana, C. Fatichah, N. Suciati, D. Teknik Informatika, and N. Surabaya, “Facial Inpainting Pada Citra Wajah Unaligned Menggunakan Generative Adversarial Networks Dengan Feature Reconstruction Loss,” JUTI J. Ilm. Teknol. Inf., vol. 18, no. 2, pp. 171–178, 2020.
X. Wang, L. Sun, A. Chehri, and Y. Song, “A Review of GAN-Based Super-Resolution Reconstruction for Optical Remote Sensing Images,” Remote Sens., vol. 15, no. 20, pp. 1–34, 2023, doi: 10.3390/rs15205062.
K. Liu, J. Li, and S. S. Hussain Bukhari, “Overview of Image Inpainting and Forensic Technology,” Secur. Commun. Networks, vol. 2022, 2022, doi: 10.1155/2022/9291971.
