辐射保真的红外遥感图像超分辨率重建

doi:10.19665/j.issn1001-2400.2019.02.018

西安电子科技大学学报 ›› 2019, Vol. 46 ›› Issue (2): 107-113.doi: 10.19665/j.issn1001-2400.2019.02.018

辐射保真的红外遥感图像超分辨率重建

时文俊¹,郭从洲²,童晓冲²,田园²,曹闻³

^1. 郑州升达经贸管理学院应用数学研究所,河南郑州 451191
^2. 信息工程大学基础部,河南郑州 450001
^3. 郑州大学智慧城市研究院,河南郑州 450001

收稿日期:2018-06-30 出版日期:2019-04-20 发布日期:2019-04-20
作者简介:时文俊(1980-),女,副教授,E-mail:shiwj_1980@126.com.
基金资助:
河南省科技攻关项目(172102210553);河南省高等学校青年骨干教师培养计划项目(2017GGJS193);国家社会科学基金(18BJL073)

Super-resolution reconstruction of infrared remote sensing images with radiation fidelity

SHI Wenjun¹,GUO Congzhou²,TONG Xiaochong²,TIAN Yuan²,CAO Wen³

^1. Institute of Applied Mathematics, Zhengzhou ShengDa Univ. of Economic, Business & Management, Zhengzhou 451191, China
^2. Basis Department, Information Engineering Univ., Zhengzhou 450001, China
^3. Intelligent City Research Institute, Zhengzhou Univ., Zhengzhou 450001, China

Received:2018-06-30 Online:2019-04-20 Published:2019-04-20

摘要/Abstract

摘要：

为了解决红外遥感图像超分辨率重建与辐射定标精度保真之间的矛盾,结合正则化超分辨率重建理论,建立了基于二阶总广义变分的超分辨率重建模型。通过分析重建模型的特点,引入交替方向乘子法进行数值求解;重建过程利用双边滤波器原理,将图像的高低频信息分离;针对分离后高频信息图像进行重建处理,将低频信息图像与重建后的高频信息图像融合达到超分辨率的目的。利用风云四号气象卫星得到的真实红外图像进行实验验证和定量化分析,表明该方法对辐射定标精度的影响要小于常规意义下的超分辨率重建的影响。

关键词: 辐射保真, 超分辨率重建, 二阶总广义变分, 交替方向乘数法, 高频信息图像

Abstract:

In order to overcome the contradiction between the super resolution reconstruction of an infrared remote sensing image and the fidelity of radiometric calibration, combined with the theory of regularized super-resolution reconstruction, a super resolution reconstruction model based on two-order total generalized variation is established. Through the analysis of the characteristics of the reconstruction model, the ADMM algorithm is introduced to solve the numerical solution. In the reconstruction process, the bilateral high-frequency filter is used to separate the high and low frequency information of the image, and after separation only the high frequency information image is dealt with. Finally, the low frequency information image and the reconstructed high frequency information image are fused to achieve super resolution. Experimental verification and quantitative analysis of the FY-4 meteorological satellite infrared image show that the influence of this method on radiometric calibration is less than that of conventional super resolution reconstruction.

Key words: radiation fidelity, super-resolution reconstruction, two order generalized variation, alternating directions method of multipliers, high frequency information image

中图分类号:

TP394.1

时文俊,郭从洲,童晓冲,田园,曹闻. 辐射保真的红外遥感图像超分辨率重建[J]. 西安电子科技大学学报, 2019, 46(2): 107-113.

SHI Wenjun,GUO Congzhou,TONG Xiaochong,TIAN Yuan,CAO Wen. Super-resolution reconstruction of infrared remote sensing images with radiation fidelity[J]. Journal of Xidian University, 2019, 46(2): 107-113.

图/表 9

图1

图2

图3

图4

图5

图6

表1

表2

表3

参考文献 14

[1]	顾行发, 田国良, 余涛 , 等. 航天光学遥感器辐射定标原理与方法[M]. 北京: 科学出版社, 2013: 1-68.
[2]	TOMASI C, MANDUCHI R . Bilateral Filtering for Gray and Color Images[C]// Proceedings of the 1998 IEEE International Conference on Computer Vision. Piscataway: IEEE, 1998: 839-846.
[3]	ZHENG Z, JIA S, ZHANG G , et al. An Approach of Point Cloud Denoising Based on Improved Bilateral Filtering[C]// Proceedings of the SPIE: 10615. Bellingham: SPIE, 2018: 1061557.
[4]	周峰, 金炜, 龚飞 , 等. 主题学习和稀疏表示的MODIS图像超分辨率重建[J]. 遥感学报, 2017,21(2):253-261.
	ZHOU Feng, JIN Wei, GONG Fei , et al, Super Resolution Reconstruction of MODIS Image Based on Topic Learning and Sparse Representation[J]. Journal of Remote Sensing, 2017,21(2):253-261.
[5]	BREDIES K, KUNISCH K, POCK T . Total Generalized Variation[J]. SIAM Journal on Imaging Sciences, 2010,3(3):492-526. doi: 10.1137/090769521
[6]	许建楼, 冯象初, 郝岩 . 二阶总广义变分图像修复模型及其算法[J]. 西安电子科技大学学报, 2012,39(5):18-23. doi: 10.3969/j.issn.1001-2400.2012.05.004
	XU Jianlou, FENG Xiangchu, HAO Yan . Second Order Total Generalized Variational Inpainting Model and Its Algorithm[J]. Journal of Xidian University, 2012,39(5):18-23. doi: 10.3969/j.issn.1001-2400.2012.05.004
[7]	KANG M, KANG M, JUNG M . Total Generalized Variation Based Denoising Models for Ultrasound Images[J]. Journal of Scientific Computing, 2017,72(1):172-197. doi: 10.1007/s10915-017-0357-3
[8]	伍政华, 孙明健, 顾宗山 , 等. 基于二阶广义方向性全变分的图像超分辨率重建方法[J]. 电子学报, 2017,45(11):2625-2632. doi: 10.3969/j.issn.0372-2112.2017.11.008
	WU ZHENGHUA, SUN MINGJIAN, GU ZONGSHAN , et al. Second-order Directional Total Generalized Variation Regularization for Image Super Resolution[J]. Acta Electronica Sinica, 2017,45(11):2625-2632. doi: 10.3969/j.issn.0372-2112.2017.11.008
[9]	BREDIES K, SUN H P . Preconditioned Douglas-Rachford Algorithms for TV- and TGV-regularized Variational Imaging Problems[J]. Journal of Mathematical Imaging and Vision, 2015,52(3):317-344. doi: 10.1007/s10851-015-0564-1
[10]	BREDIES K, VALKONEN T . Inverse Problems with Second-order Total Generalized Variation Constraints[EB/OL]. [ 2018- 04- 20]. http://tuomov.iki.fi/mathematics/SampTA2011.pdf.
[11]	刘孝艳, 冯象初 . 对称张量空间下高阶正则化的图像恢复模型[J]. 计算机科学, 2013,40(11A):359-362. doi: 10.3969/j.issn.1002-137X.2013.z2.090
	LIU Xiaoyan, FENG Xiangchu . Image Restoration Model of High Order Regularization in Symmetric Tensor Space[J]. Computer Science, 2013,40(11A):359-362. doi: 10.3969/j.issn.1002-137X.2013.z2.090
[12]	DENG W, YIN W . On the Global and Linear Convergence of the Generalized Alternating Direction Method of Multipliers[J]. Journal of Scientific Computing, 2016,66(3):889-916. doi: 10.1007/s10915-015-0048-x
[13]	WARREN R E, OSHER S J . Hyperspectral Unmixing by the Alternating Direction Method of Multipliers[J]. Inverse Problems and Imaging, 2017,9(3):917-933.
[14]	HE R, ZHENG W S, TAN T , et al. Half-quadratic Minimization for Robust Sparse Representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013,36(2) : 261-275. doi: 10.1109/TPAMI.2013.102

数据名称	波段1	波段2	截取区域坐标
原始数据	1.64	1.57	(1:200,1:200)
文中超分辨率重建	1.64	1.57	(1:400,1:400)
常规超分辨率重建	1.74	1.69	(1:400,1:400)

绝对误差	波段1		波段2
绝对误差	常规超分辨率	文中方法超分辨率	常规超分辨率	文中方法超分辨率
<1%	21.06	31.60	40.35	48.27
<2%	40.27	46.79	61.55	65.96
<5%	65.30	70.29	82.23	87.96
<10%	76.54	87.42	91.19	97.24

绝对误差	波段1		波段2
绝对误差	k	b	k	b
常规超分 (置信区间)	0.93690 (0.93480, 0.93810)	0.01702 (0.01593, 0.01920)	0.95890 (0.95880, 0.95910)	0.01401 (0.01393, 0.01410)
文中超分 (置信区间)	0.989800 (0.989700, 0.989900)	0.002072 (0.002026, 0.002128)	0.984800 (0.984700, 0.985000)	0.001785 (0.001766, 0.001803)

辐射保真的红外遥感图像超分辨率重建

Super-resolution reconstruction of infrared remote sensing images with radiation fidelity

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 14

相关文章 5

Metrics

本文评价

推荐阅读 10

[1]	张宇浩,程培涛,张书豪,王秀美. 一种自适应权重学习的轻量超分辨率重建网络[J]. 西安电子科技大学学报, 2021, 48(5): 15-22.
[2]	欧阳宁,王先傲,蔡晓东,林乐平. 结合小波与递归神经网络的低分辨率人脸识别[J]. 西安电子科技大学学报, 2019, 46(6): 95-101.
[3]	刘树东,王晓敏,张艳. 一种对称残差CNN的图像超分辨率重建方法[J]. 西安电子科技大学学报, 2019, 46(5): 15-23.
[4]	陈晓范,申海杰,边倩,王振铎,田新志. 结合注意力机制的人脸超分辨率重建[J]. 西安电子科技大学学报, 2019, 46(3): 148-153.
[5]	吴玉莲;冯象初. 利用小波和二阶总广义变分的图像放大方法[J]. J4, 2013, 40(6): 155-161.