《计算机应用研究》|Application Research of Computers

压缩感知和相似性约束的图像超分辨率重构算法

Compressed sensing and similarity constraint image super-resolution

免费全文下载 (已被下载 次)  
获取PDF全文
作者 吴科永,陈东,辛宁,曹桂兴
机构 1.西安电子科技大学 综合业务网理论及关键技术国家重点实验室,西安 710071;2.中国空间技术研究院 通信卫星事业部,北京 100094
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)05-057-1555-05
DOI 10.19734/j.issn.1001-3695.2017.12.0833
摘要 针对通过外部学习进行超分辨率存在图像质量不佳、细节不真实的问题提出一种压缩感知和相似性约束的单帧图像超分辨率算法。算法首先利用压缩感知中测量域与频域的线性关系对训练库图像在测量域分类,对不同类别图像块训练对应类别的字典,提高字典的表示能力;然后在重构时利用图像的非局部相似性,将图像在分类字典下的稀疏性和相似块信息同时作为先验信息联合约束重构过程,最后恢复出高分辨率图像。实验结果表明算法重构出的高分辨率图像具有丰富的细节以及清晰的边缘,重构图像主观质量良好。
关键词 超分辨率; 压缩感知; 测量域字典分类; 非局部相似; 联合重构
基金项目 国家自然科学基金资助项目(61372068,61772387)
本文URL http://www.arocmag.com/article/01-2019-05-057.html
英文标题 Compressed sensing and similarity constraint image super-resolution
作者英文名 Wu Keyong, Chen Dong, Xin Ning, Cao Guixing
机构英文名 1.State Key Laboratory of Integrated Services Networks,Xidian University,Xi'an 710071,China;2.Institute of Telecommunication Satellite,China Academy of Space Technology,Beijing 100094,China
英文摘要 Aiming at the problem of poor quality and contrived details in super-resolution reconstruction by external learning, this paper proposed a single image super-resolution based on compressed sensing and similarity constraint. Firstly, this paper proposed a classified dictionary based on measurement domain by using the linear relationship between the measurement domain and the frequency domain in compressed sensing, which improved the representation of dictionaries, then, it used the non-local similarity in the reconstruction process, combined the sparsity under the feature dictionary and similar block information as priori information to regularize super-resolution reconstruction, and finally recovered the high resolution image. Experimental results show that the reconstructed image has rich details and defined edges, and the subjective quality is good.
英文关键词 super-resolution; compressed sensing; dictionary classification in measurement domain; non-local similarity; joint reconstruction
参考文献 查看稿件参考文献
 
收稿日期 2017/12/10
修回日期 2018/1/25
页码 1555-1559
中图分类号 TP391.4
文献标志码 A