Enhancing Blurred Facial Images Using Generative Adversarial Networks

Year 2024, Volume: 19 Issue: 2, 305 - 313, 30.09.2024

Kenan Bakır , Yaman Akbulut

https://doi.org/10.55525/tjst.1381587

Abstract

This research examines the improvement of facial images using generative adversarial networks (GANs). The significance of this topic lies in its potential for enhancing image processing and facial recognition systems. The primary objective of this study is to evaluate the effectiveness of GANs in enhancing the quality of facial images. The hypotheses put forth in this thesis suggest that GAN-based methods can succeed in increasing the resolution and realism of facial images. The sample consists of 70.000 different facial images, representing the primary data source for this study. The method primarily involves the creation and training of a GAN model. A GAN consists of a generator that attempts to mimic real images during the learning process and a discriminator network that evaluates the realism of these images. The findings of the study demonstrate the effectiveness of GANs in making facial images higher in resolution and more realistic. This has the potential to improve the performance of facial recognition systems and enable more precise diagnoses in medical imaging applications. This information underscores the importance of GAN-based methods in enhancing facial images.

Keywords

GAN, facial images, image processing, resolution enhancement, face recognition

Üretici Çekişmeli Ağlar ile Bulanık Yüz Görüntülerinin Geliştirilmesi

Abstract

Bu araştırma, üretici çekişmeli ağlar (ÜÇA) kullanarak yüz görüntülerinin geliştirilmesini incelemektedir. Bu konunun önemi, görüntü işleme ve yüz tanıma sistemlerinin gelişmesine yönelik potansiyeli içermektedir. Bu çalışmanın ana hedefi, ÜÇA’ların yüz görüntülerinin kalitesini artırma yeteneğini değerlendirmektir. Bu tezin önerdiği hipotezler, ÜÇA temelli tekniklerin yüz görüntülerinin çözünürlüğünü artırma ve daha gerçekçi hale getirme konularında başarılı olabileceğini iddia etmektedir. Örnekleme, 70,000 farklı yüz görüntüsünden oluşmaktadır ve bu örneklem boyutu, bu çalışmanın temel veri kaynağını temsil etmektedir. Yöntem öncelikle ÜÇA modelinin oluşturulmasını ve eğitilmesini içermektedir. ÜÇA, öğrenme süreci sırasında gerçek görüntülerin taklit edilmesine çalışan bir üretici ve bu görüntülerin gerçekçilik derecesini değerlendiren bir ayrımcı ağdan oluşur. Araştırma sonuçları, ÜÇA’ların yüz görüntülerini daha yüksek çözünürlükte ve daha gerçekçi hale getirme yeteneğini göstermektedir. Bu, yüz tanıma sistemlerinin performansını artırabilir ve tıbbi görüntüleme uygulamalarında daha hassas teşhisler koyma potansiyelini sunar. Bu bilgi, ÜÇA temelli yöntemlerin yüz görüntülerinin geliştirilmesindeki önemini vurgulamaktadır.

Keywords

ÜÇA, yüz görüntüleri, görüntü işleme, çözünürlük arttırma, yüz tanıma

References

• Allebach J, Wong PW. Edge-directed interpolation. Proceedings of International Conference on Image Processing. 1996; 707–710.
• Nasrollahi K, Moeslund TB. Super-resolution: A comprehensive survey. Machine Vision and Applications 2014; 25: 1423–1468.
• Yang C-Y, Ma C, Yang M-H. Single-image super-resolution: A benchmark. European Conference on Computer Vision (ECCV), Springer, 2014; 372–386.
• Borman S, Stevenson RL. Super-Resolution from Image Sequences - A Review. Midwest Symposium on Circuits and Systems, 1998; 374–378.
• Farsiu S, Robinson MD, Elad M, Milanfar P. Fast and robust multiframe super resolution. IEEE Transactions on Image Processing 2004; 13(10): 1327–1344.
• Duchon CE. Lanczos Filtering in One and Two Dimensions. Journal of Applied Meteorology 1979; 18: 1016–1022.
• Li X, Orchard MT. New edge-directed interpolation. IEEE Transactions on Image Processing 2001; 10(10): 1521–1527.
• Freeman WT, Jones TR, Pasztor EC. Example-based superresolution. IEEE Computer Graphics and Applications 2002; 22(2): 56–65.
• Freeman WT, Pasztor EC, Carmichael OT. Learning low-level vision. International Journal of Computer Vision 2000; 40(1): 25–47.
• Yang J, Wright J, Huang T, Ma Y. Image super-resolution as sparse representation of raw image patches. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2008; 1–8.
• Dong W, Zhang L, Shi G, Wu X. Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization. IEEE Transactions on Image Processing 2011; 20(7): 1838–1857.
• Do N-T, Na I-S, Kim S-H. Forensics face detection from GANs using convolutional neural network. ISITC, 2018.
• Glasner D, Bagon S, Irani M. Super-resolution from a single image. IEEE International Conference on Computer Vision (ICCV) 2009; 349–356.
• Huang JB, Singh A, Ahuja N. Single image super-resolution from transformed self-exemplars. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2015; 5197–5206.
• Gu S, Zuo W, Xie Q, Meng D, Feng X, Zhang L. Convolutional sparse coding for image super-resolution. IEEE International Conference on Computer Vision (ICCV) 2015; 1823–1831.
• Tai Y-W, Liu S, Brown MS, Lin S. Super Resolution using Edge Prior and Single Image Detail Synthesis. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2010; 2400–2407.
• Zhang K, Gao X, Tao D, Li X. Multi-scale dictionary for single image super-resolution. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2012; 1114–1121.
• Yue H, Sun X, Yang J, Wu F. Landmark image superresolution by retrieving web images. IEEE Transactions on Image Processing 2013; 22(12) 4865–4878.
• Timofte R, De Smet V, Van Gool L. Anchored neighborhood regression for fast example-based super-resolution. IEEE International Conference on Computer Vision (ICCV) 2013; 1920–1927.
• Timofte R, De Smet V, Van Gool L. A+: Adjusted anchored neighborhood regression for fast super-resolution. Asian Conference on Computer Vision (ACCV) 2014; 111–126.
• Kim KI, Kwon Y. Single-image super-resolution using sparse regression and natural image prior. IEEE Transactions on Pattern Analysis and Machine Intelligence 2010; 32(6): 1127–1133.
• He H, Siu WC. Single image super-resolution using gaussian process regression. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2011; 449–456.
• Salvador J, Perez-Pellitero E. Naive bayes super-resolution forest. IEEE International Conference on Computer Vision (ICCV) 2015; 325–333.
• Dai D, Timofte R, Van Gool L. Jointly optimized regressors for image super-resolution. Computer Graphics Forum 2015; 34: 95–104.
• Johnson J, Alahi A, Fei-Fei L. Perceptual losses for real-time style transfer and super-resolution. European Conference on Computer Vision (ECCV) 2016.
• Mechrez R, Talmi I, Shama F, Zelnik-Manor L. Maintaining natural image statistics with the contextual loss 2018.
• Ledig C, Theis L, Huszár F, Caballero J, Cunningham A, Acosta A, Aitken A, Tejani A, Totz J, Wang Z, et al. Photo-realistic single image super-resolution using a generative adversarial network. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017.
• Sajjadi MS, Schölkopf B, Hirsch M. Enhancenet: Single image super-resolution through automated texture synthesis. IEEE International Conference on Computer Vision (ICCV) 2017.
• Wang X, Yu K, Dong C, Loy CC. Recovering realistic texture in image superresolution by deep spatial feature transform. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018.
• Bevilacqua M, Roumy A, Guillemot C, Morel MLA. Low-complexity single-image super-resolution based on nonnegative neighbor embedding. Proceedings of the British Machine Vision Conference (BMVC) 2012; 1–10.
• Chang H, Yeung DY, Xiong Y. Super-resolution through neighbor embedding. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Washington, DC, USA, 2004.
• Freeman WT, Pasztor EC, Carmichael OT. Learning low-level vision. International Journal of Computer Vision 2000; 40(11): 25–47.
• Mo H, Chen B, Luo W. Fake faces identification via convolutional neural network. ACM IH&MMSEC, 2018.
• Yang J, Wang Z, Lin Z, Cohen S, Huang T. "Coupled dictionary training for image super-resolution." IEEE Transactions on Image Processing 2012; 21(11): 3467–3478.
• Lago F, Pasquini C, Böhme R, et al. More real than real: A study on human visual perception of synthetic faces. 2021.
• Yang J, Wright J, Huang TS, Ma Y. Image super-resolution via sparse representation. IEEE Transactions on Image Processing 2010; 19(11): 2861–2873.
• Zeyde R, Elad M, Protter M. On single image scale-up using sparse-representations. Proceedings of the 7th International Conference on Curves and Surfaces 2012; 711–730.