Improvement of blurred face images with generative adversarial networks

This study examines the enhancement of facial images using GAN (Generative Adversarial Network). The significance of the topic lies in its potential for improving image processing and facial recognition systems. The aim of this study is to evaluate the effectiveness of GANs in enhancing the quality of facial images. The hypotheses of this thesis propose that GAN-based methods are successful in increasing the resolution and realism of facial images. The sample consists of 70000 different facial images, representing the primary data source for this study. The method primarily involves the creation and training of the GAN model, which consists of a generator striving to mimic real images and a discriminator network assessing the realism of these images. The findings of the study demonstrate the effectiveness of GANs in producing higher-resolution and more realistic facial images, potentially improving the performance of facial recognition systems and enabling sharper diagnostics in medical imaging applications. The generated knowledge underscores the significance of GAN-based methods in enhancing facial images.