This research presents a novel approach to image enhancement and photo restoration by employing Generative Adversarial Networks (GANs) to tackle the intricate issue of moire pattern removal. Moire patterns, unwanted interference patterns that arise when capturing images of fine repetitive structures, such as textiles or screens, can severely degrade image quality. Conventional techniques for moire pattern removal involve filtering, which can inadvertently result in the loss of fine details or introduce blurriness. In contrast, GANs offer a promising solution by leveraging their generative capabilities to produce enhanced images while preserving crucial details. Through adversarial training, the generator learns to produce images that effectively eliminate moire artifacts while maintaining or even enhancing the overall image quality. This dataset serves as the foundation for training the GAN, enabling it to learn the intricate patterns and characteristics of moire artifacts. Quantitative metrics encompass peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and other image quality assessment measures, while qualitative analysis involves visual comparison between original moire-laden images and the GAN produced enhanced images. The results demonstrate the remarkable potential of GAN in tackling moire patterns and enhancing image quality. The proposed method outperforms conventional techniques in terms of preserving fine details and eliminating moire artifacts.