Administrative Technical Institute Staff Member Defends Master’s Thesis at the University of Basra
Sahar Jabbar Mohammed, a staff member of the Department of Business Administration Techniques at the Administrative Technical Institute in Basra, successfully defended her master’s thesis at the College of Computer Science and Information Technology, University of Basra. The thesis was entitled:
“A Deep Learning Based Framework for Image-to-Image Translation on Aerial Maps.”
The research aimed to develop a deep learning framework for transforming aerial images into digital maps by designing a hybrid model that combines Variational Autoencoders (VAE) and Generative Adversarial Networks (GAN) within a cVAE-cGAN architecture. The proposed framework addressed the challenge of balancing image quality with output diversity in image-to-image translation tasks.
The study’s findings demonstrated a significant improvement in the quality of the generated images compared with the conventional Pix2Pix model. The performance was evaluated using several specialized metrics, including PSNR, SSIM, LPIPS, and FID, all of which indicated the effectiveness of the proposed approach in enhancing image generation accuracy and quality.
