中国科学院大学学报2025，Vol.42Issue(6)：769-780,12.DOI:10.7523/j.ucas.2024.007
基于双时相特征的SAR生成光学影像方法
A method for SAR-to-optical image synthesis based on bi-temporal features
摘要
Abstract
The robust optical image time series are of great value in many applications of remote sensing.However,due to the effects of weather conditions like clouds and rain,it is very difficult to obtain such robust time series of optical images in many regions.Using the all-weather imaging capability of synthetic aperture radar(SAR)to generate optical images from SAR images is an effective solution to the missing data of optical images.But there is still a problem that the quality of generated images in complicated scenarios is much worse than that in simple scenarios.In this paper,we build bi-temporal datasets of different scenarios based on Sentinel imagery and propose an improved generator of conditional generation adversarial network.The encoder-decoder-based generator learns to extract and fuse the bi-temporal polarized SAR features and the additional optical features from the source time phase.In addition,a strategy to balance the weights of SAR and optical features is adopted.Comparison experiments show that our method is the best on FID and PSNR among all evaluated methods.The proposed method significantly reduces the gap in the quality of generated images between simple scenario and complicated scenario.The ablation study shows that our method outperforms the baseline model by 46 in FID,6.6 dB in PSNR and 0.44 in SSIM.Our method efficiently improves the quality of generated images in different scenarios.关键词
可见光影像生成/双时相数据集/合成孔径雷达/生成对抗网络Key words
optical image synthesis/bi-temporal dataset/SAR/generative adversarial network分类
计算机与自动化引用本文复制引用
翁永椿,马勇,陈甫,尚二萍,姚武韬,仉淑艳,杨进,刘建波..基于双时相特征的SAR生成光学影像方法[J].中国科学院大学学报,2025,42(6):769-780,12.基金项目
国家自然科学基金(42201063)、海南省重点研发计划(ZDYF2021SHFZ260)和海南自然科学青年基金(520QN295)资助 （42201063）
