//U-Soar Generative Adversarial Networks

IUP Undergraduate Summer Opportunities for Applying Research (U-SOAR)

~IUP U-Soar Website~
https://www.iup.edu/research/events/u-soar/index.html

GitHub  GAN-Image-Denoising

Summary

The subject of GAN Image Denoising was a research subject, done by Samuel Rocco, for the Indiana University of Pennsylvania Undergraduate Summer Opportunities for Applied Research (U-SOAR) Program for summer 2023.

This GitHub repository, contains a project focused on using Generative Adversarial Networks (GANs) for image denoising. The project involves applying GANs to remove noise from images, enhancing their quality and clarity. The repository includes the complete codebase, along with detailed documentation on the implementation process, dataset preparation, model architecture, and training methodology. Additionally, sample results are provided to demonstrate the effectiveness of the GAN in restoring noisy images to their original, noise-free state.

My Contributions

Technologies Used: Python, TensorFlow, Matplotlib, Pandas, NumPy

  • Conducted research on utilizing Generative Adversarial Networks (GANs) to mitigate noise from images.
  • Designed and implemented a GAN model trained on the current image while attempting to de-noise in real time, all done while achieving a significant improvement in image clarity.
  • Presented findings at the 2023 U-SOAR Research Symposium.
  • Collaborated with faculty advisors to refine the model and explore potential applications in computer vision, and utilizing GAN for real-time network security.