摘要
围绕图像超分辨率重建技术,列举并总结了传统图像超分辨率重建方法的优点及其局限性。回顾了几种经典的基于学习的超分辨率方法,并分析对比了不同模型的特点及在不同数据集上的实验结果。最后,对图像超分辨率重建技术的未来发展趋势进行展望。
Focusing on super-resolution(SR)reconstruction methods,the advantages and limitations of traditional image SR reconstruction methods have been introduced and summarized.Several classical learning based super-resolution methods are reviewed.The characteristics of different models and the experimental results on different data sets are analyzed and compared.Finally,prospects for the future development of SR methods are presented.
作者
李莹华
刘悦
刘颖
LI Yinghua;LIU Yue;LIU Ying(Institute of Image and Information Processing,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;Key Laboratory of Electronic Information Application Technology for Crime Scene Investigation,Ministry of Public Security,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;International Cooperation Research Center for Wireless Communication and Information Processing Technology,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)
出处
《西安邮电大学学报》
2022年第2期72-87,共16页
Journal of Xi’an University of Posts and Telecommunications
基金
国家自然科学基金项目(62071380)
陕西省自然科学基础研究计划项目(2021JQ-708)。
关键词
图像超分辨率重建
深度学习
对抗网络
稀疏表示
神经网络
image super-resolution reconstruction
deep learning
countermeasure network
sparse representation
neural network
