|
|
The Modeling and Simulation in Electrical and Electronics Engineering، جلد ۴، شماره ۳، صفحات ۱۱-۲۴
|
|
|
| عنوان فارسی |
|
|
| چکیده فارسی مقاله |
|
|
| کلیدواژه‌های فارسی مقاله |
|
|
| عنوان انگلیسی |
SRGAN Enhancement through Autoencoder-Pretrained U-Net with Residual Blocks for Improved Image Super-Resolution |
|
| چکیده انگلیسی مقاله |
Super-resolution is a crucial task in image processing, enhancing the resolution of low-quality images for applications such as surveillance, remote sensing, and autonomous systems. Traditional methods often struggle to preserve fine details, leading to artifacts and reduced visual fidelity. This study introduces the Pretrained RU-SRGAN, an enhanced Super-Resolution Generative Adversarial Network (SRGAN) that incorporates U-Net architecture, residual learning, and autoencoder pretraining to improve both image quality and computational efficiency, particularly in resource-constrained environments like UAVs. The goal of this research is to evaluate how these architectural modifications can enhance super-resolution performance with limited data. Autoencoder pretraining enables the generator to leverage learned features from low-resolution images, accelerating convergence and improving high-resolution reconstructions. Experimental results show that Pretrained RU-SRGAN outperforms baseline models, achieving a PSNR of 25.7 dB and an SSIM of 0.83. These results highlight the model's ability to preserve fine details and structural integrity, making it particularly effective for real-time image enhancement in UAV applications. The Pretrained RU-SRGAN provides a robust solution for super-resolution tasks, balancing high-quality image reconstruction with computational efficiency, and is well-suited for practical deployment in dynamic, resource-limited environments. |
|
| کلیدواژه‌های انگلیسی مقاله |
Transfer learning,UAV,Image Reconstruction,RU-Net,SSIM,PSNR |
|
| نویسندگان مقاله |
Amirreza Rouhbakhshmeghrazi |
School of Electronics and Information, Northwestern Polytechnical University, Xi'an, China.
Shayan Nalbandian |
School of Software Engineering, Northwest Polytechnic University, Xi'an, China.
Shayan Nalbandian |
School of Software Engineering, Northwest Polytechnic University, Xi'an, China.
Chao Song |
School of Electronics & Information, Northwest Polytechnic University, Xi'an, China.
Ghazal Alizadeh |
School of Aeronautics, Northwestern Polytechnical University, Xi'an, China.
Mohammad Reza Hassannezhad |
School of
Aeronautics, Northwestern Polytechnical University, Xi’an, Shaanxi, China.
|
|
| نشانی اینترنتی |
https://mseee.semnan.ac.ir/article_9634_e06839e6edcd0d189572b45859a5f9e0.pdf |
| فایل مقاله |
فایلی برای مقاله ذخیره نشده است |
| کد مقاله (doi) |
|
| زبان مقاله منتشر شده |
en |
| موضوعات مقاله منتشر شده |
|
| نوع مقاله منتشر شده |
|
|
|
|
برگشت به:
صفحه اول پایگاه |
نسخه مرتبط |
نشریه مرتبط |
فهرست نشریات
|