混合稀疏表示模型的超分辨率重建

杨雪; 李峰; 鹿明; 辛蕾; 鲁啸天; 张南

doi:10.11834/jrs.20219409

遥感技术方法 | 浏览量 : 0 下载量: 283 CSCD: 1 更多指标

R-PDF
PDF
导出
分享
收藏
专辑

混合稀疏表示模型的超分辨率重建
New super-resolution reconstruction method based on Mixed Sparse Representations
2022年26卷第8期页码：1685-1697
纸质出版日期： 2022-08-07 ，
DOI： 10.11834/jrs.20219409

扫描看全文

杨雪，李峰，鹿明，辛蕾，鲁啸天，张南.2022.混合稀疏表示模型的超分辨率重建.遥感学报，26（8）： 1685-1697

Yang X，Li F，Lu M，Xin L，Lu X T and Zhang N. 2022. New super-resolution reconstruction method based on Mixed Sparse Representations. National Remote Sensing Bulletin， 26（8）：1685-1697
杨雪，李峰，鹿明，辛蕾，鲁啸天，张南.2022.混合稀疏表示模型的超分辨率重建.遥感学报，26（8）： 1685-1697 DOI： 10.11834/jrs.20219409.

Yang X，Li F，Lu M，Xin L，Lu X T and Zhang N. 2022. New super-resolution reconstruction method based on Mixed Sparse Representations. National Remote Sensing Bulletin， 26（8）：1685-1697 DOI： 10.11834/jrs.20219409.

摘要

超分辨率重建是当前卫星遥感数据空间分辨率提升的重要技术，但目前现有的超分辨率重建方法在处理具有复杂地物特征的影像时效果往往不佳。当遥感影像中包含有各种非均匀地物信息时，难以构建一种通用的模型来解决遥感影像的病态问题。基于此，本文结合图像稀疏表达与非凸高阶全变分理论，提出了一种混合稀疏表示模型的新型超分辨率重建方法（MSR-SRR）。这种方法以遥感图像在多重变换域的稀疏性表达作为先验概率模型，通过正则化方法来完成超分辨率重构，不仅保留了超分重建结果影像的边缘信息，而且对影像中产生的“阶梯效应”进行了适当的平滑处理。该方法利用迭代重加权

http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=35277473&type=

http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=35277470&type=

2.53999996

2.28600001

交替方向乘子方法进行求解，提高了算法的运行效率，改善了影像质量。为了证明所提出方法的有效性，MSR-SRR结果与非均匀插值、POCS和IBP等传统超分方法的重建结果进行了对比验证。结果表明，MSR-SRR方法的图像清晰度平均提升了31.74%，PSFs半峰宽度最大，高斯方差值达到1.8415，效果明显优于其他方法。为进一步评估MSR-SRR结果的实用性，本文以高分四号卫星（GF-4）影像作为样例，利用支持向量机（SVM）分类方法对超分重建前后的影像进行了分类试验和精度验证。结果表明，超分辨率重建后的影像结果相对于原始影像的分类结果，Kappa系数提升了9.7%，OA值提升了5.96%。这表明MSR-SRR方法可以有效提升影像清晰度，丰富影像纹理细节，增强图像质量，有效提升影像分类精度。

Abstract

When processing remote sensing images with complex features

the conventional Super-Resolution Reconstruction (SRR) methods are often not ideal

especially for remote sensing images containing various non-uniform object information. A universal method to solve this problem is difficult to construct at present. A new SR reconstruction method of mixed sparse representation model (MSR-SRR) combined with the sparse representation and non-convex high-order total variational regularizer has been proposed to solve this problem. In this method

the sparse representation of remote sensing images in multiple transform domains is regarded as a prior probability model

and the SR reconstruction is completed by regularization. The obtained image not only retains the edge information of the image result by SR reconstruction

but also smoothens the “ladder effect” of the image. The efficiency of operation and the quality of SR reconstruction results are improved by an effective re-weighted

alternating direction method. Results show that the sharpness of the image increases by 31.74% on the average

the half-peak width of PSFs is the largest

and the Gaussian variance value reaches 1.8415. The GF-4 satellite images have been selected to carry out validation experiment to verify the feasibility and validity of MSR-SRR. The reconstruction results show that the images using the MSR-SRR method have better definition

richer details

and higher quality than those with non-uniform interpolation

the POCS method

and IBP method. The support vector machine method is used to classify and evaluate the accuracy of the images before and after SR reconstruction. The results show that the overall accuracy and Kappa coefficient of the reconstructed super-resolution image are improved more significantly than the original image classification results. The OA value increases by 5.96%

and the Kappa coefficient increases by 9.7%. The findings confirmed that the MSR-SRR method is effective and feasible and has extensive practical value.

关键词

遥感高分四号超分辨率重建混合稀疏表示全变分非凸

Keywords

remote sensingGF-4Super-Resolution Reconstruction (SRR)Mixed Sparse Representation (MSR)Total Variation (TV)Non-convex

references

Adam T and Paramesran R. 2019. Image denoising using combined higher order non-convex total variation with overlapping group sparsity. Multidimensional Systems and Signal Processing, 30: 503-527 [DOI: 10.1007/s11045-018-0567-3http://dx.doi.org/10.1007/s11045-018-0567-3]

Babacan S D, Molina R and Katsaggelos A K. 2011. Variational Bayesian super resolution. IEEE Transactions on Image Processing, 20(4): 984-999 [DOI: 10.1109/TIP.2010.2080278http://dx.doi.org/10.1109/TIP.2010.2080278]

Bai M R, Zhang X J and Shao Q Q. 2016. Adaptive correction procedure for TVL1 image deblurring under impulse noise. Inverse Problems, 32(8): 085004 [DOI: 10.1088/0266-5611/32/8/085004http://dx.doi.org/10.1088/0266-5611/32/8/085004]

Bao H, Li Z L, Chai F M and Yang H S. 2015. Filter wheel mechanism for optical remote sensor in geostationary orbit. Optics and Precision Engineering, 23(12): 3357-3363

鲍赫, 李志来, 柴方茂, 杨会生. 2015. 静止轨道光学遥感器的滤光轮机构. 光学精密工程, 23(12): 3357-3363 [DOI: 10.3788/OPE.20152312.3357http://dx.doi.org/10.3788/OPE.20152312.3357]

Beck A and Teboulle M. 2009. A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM Journal on Imaging Sciences, 2(1): 183-202 [DOI: 10.1137/080716542http://dx.doi.org/10.1137/080716542]

Chang H B, Lou Y F, Duan Y P and Marchesini S. 2018. Total variation--based phase retrieval for Poisson noise removal. SIAM Journal on Imaging Sciences, 11(1): 24-55 [DOI: 10.1137/16M1103270http://dx.doi.org/10.1137/16M1103270]

Cheng W. 2018. Detection of Sea Motion Targets in Multi-spectral Imagery of Static Orbiting Staring Satellites. Wuhan: Huazhong University of Science and Technology (程伟. 2018. 静轨凝视多光谱影像海面运动目标检测. 武汉: 华中科技大学)

Condat L. 2014. A generic proximal algorithm for convex optimization-application to total variation minimization. IEEE Signal Processing Letters, 21(8): 985-989 [DOI: 10.1109/LSP.2014.2322123http://dx.doi.org/10.1109/LSP.2014.2322123]

Crete F, Dolmiere T, Ladret P and Nicolas M. 2007. The blur effect: perception and estimation with a new no-reference perceptual blur metric//Proceedings of SPIE 6492, Human Vision and Electronic Imaging XII. San Jose: SPIE [DOI: 10.1117/12.702790http://dx.doi.org/10.1117/12.702790]

He C, Hu C H, Li X L, Yang X G and Zhang W. 2016. A parallel alternating direction method with application to compound l1-regularized imaging inverse problems. Information Sciences, 348: 179-197 [DOI: 10.1016/j.ins.2016.01.087http://dx.doi.org/10.1016/j.ins.2016.01.087]

Irani M and Peleg S. 1991. Improving resolution by image registration. CVGIP: Graphical Models and Image Processing, 53(3): 231-239 [DOI: 10.1016/1049-9652(91)90045-Lhttp://dx.doi.org/10.1016/1049-9652(91)90045-L]

Jiang C, He H Y and Ma Z Q. 2019. Instrument simulation of multispectral remote sensing images in the frame of GF-4 satellite system//Proceedings of SPIE 11156, Earth Resources and Environmental Remote Sensing/GIS Applications X. Strasbourg: SPIE [DOI: 10.1117/12.2532656http://dx.doi.org/10.1117/12.2532656]

Lei J F, Zhang S Y, Luo L, Xiao J S and Wang H. 2018. Super-resolution enhancement of UAV images based on fractional calculus and POCS. Geo-spatial Information Science, 21(1) 56-66 [DOI: 10.1080/10095020.2018.1424409http://dx.doi.org/10.1080/10095020.2018.1424409]

Li F, Li C R, Tang L L and Guo Y. 2014. Elastic registration for airborne multispectral line scanners. Journal of Applied Remote Sensing, 8(1): 083614 [DOI: 10.1117/1.JRS.8.083614http://dx.doi.org/10.1117/1.JRS.8.083614]

Li F, Xin L, Guo Y, Gao D S, Kong X H and Jia X P. 2018a. Super-resolution for GaoFen-4 remote sensing images. IEEE Geoscience and Remote Sensing Letters, 15(1): 28-32 [DOI: 10.1109/LGRS.2017.2768331http://dx.doi.org/10.1109/LGRS.2017.2768331]

Li F, Xin L, Guo Y, Gao J B and Jia X P. 2017. A framework of mixed sparse representations for remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 55(2): 1210-1221 [DOI: 10.1109/TGRS.2016.2621123http://dx.doi.org/10.1109/TGRS.2016.2621123]

Li F, Xin L, Guo Y and Jia X P. 2018b. Multitemporal mid-infrared imagery based calibration and super resolution for gaofen-4//IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium. Valencia: IEEE: 7038-7041 [DOI: 10.1109/IGARSS.2018.8517414http://dx.doi.org/10.1109/IGARSS.2018.8517414]

Liu J, Huang T Z, Selesnick I W, Lv X G and Chen P Y. 2015. Image restoration using total variation with overlapping group sparsity. Information Sciences, 295: 232-246 [DOI: 10.1016/j.ins.2014.10.041http://dx.doi.org/10.1016/j.ins.2014.10.041]

Liu Y, Yao L B, Xiong W and Zhou Z M. 2019. GF-4 satellite and automatic identification system data fusion for ship tracking. IEEE Geoscience and Remote Sensing Letters, 16(2): 281-285 [DOI: 10.1109/LGRS.2018.2869561http://dx.doi.org/10.1109/LGRS.2018.2869561]

Nie J, Deng L, Hao X L, Liu M and He Y. 2018. Application of GF-4 satellite in drought remote sensing monitoring: a case study of Southeastern Inner Mongolia. Journal of Remote Sensing, 22(3): 400-407

聂娟, 邓磊, 郝向磊, 刘明, 贺英. 2018. 高分四号卫星在干旱遥感监测中的应用. 遥感学报, 22(3): 400-407 [DOI: 10.11834/jrs.20187067http://dx.doi.org/10.11834/jrs.20187067]

Nitta K, Shogenji R, Miyatake S and Tanida J. 2006. Image reconstruction for thin observation module by bound optics by using the iterative backprojection method. Applied Optics, 45(13): 2893-2900 [DOI: 10.1364/AO.45.002893http://dx.doi.org/10.1364/AO.45.002893]

Ramana M V, Reddy E S and Satayanarayana C H. 2018. Curvelet Transform for efficient static texture classification and image fusion. International Journal of Image, Graphics and Signal Processing, 10(5): 64-71 [DOI: 10.5815/ijigsp.2018.05.07http://dx.doi.org/10.5815/ijigsp.2018.05.07]

Sha F, Zandavi S M and Chung Y Y. 2019. Fast deep parallel residual network for accurate super resolution image processing. Expert Systems with Applications, 128: 157-168 [DOI: 10.1016/j.eswa.2019.03.032http://dx.doi.org/10.1016/j.eswa.2019.03.032]

Shi M Z, Han T T and Liu S Q. 2016. Total variation image restoration using hyper-Laplacian prior with overlapping group sparsity. Signal Processing, 126: 65-76 [DOI: 10.1016/j.sigpro.2015.11.022http://dx.doi.org/10.1016/j.sigpro.2015.11.022]

Sun Y, Babu P and Palomar D P. 2017. Majorization-minimization algorithms in signal processing, communications, and machine learning. IEEE Transactions on Signal Processing, 65(3): 794-816 [DOI: 10.1109/TSP.2016.2601299http://dx.doi.org/10.1109/TSP.2016.2601299]

Sun Y J, Wang Z H, Qin Q M, Han G H, Ren H Z and Huang J F. 2018. Retrieval of surface albedo based on GF-4 geostationary satellite image data. Journal of Remote Sensing, 22(2): 220-233

孙越君, 汪子豪, 秦其明, 韩谷怀, 任华忠, 黄敬峰. 2018. 高分四号静止卫星数据的地表反照率反演. 遥感学报, 22(2): 220-233 [DOI: 10.11834/jrs.20186428http://dx.doi.org/10.11834/jrs.20186428]

Tsai R Y and Huang T S. 1984. Multiframe image restoration and registration. Advances in Computer Vision and Image Processing, 1(2) 317-339

Wang Y L, Bi S S, Sun M L and Cai M Y. 2014. Image retrieval algorithm based on SIFT, K-means and LDA. Journal of Beijing University of Aeronautics and Astronautics, 40(9) 1317-1322

汪宇雷, 毕树生, 孙明磊, 蔡月日. 2014. 基于SIFT, K-Means和LDA的图像检索算法. 北京航空航天大学学报, 40(9): 1317-1322 [DOI: 10.13700/j.bh.1001-5965.2013.0601http://dx.doi.org/10.13700/j.bh.1001-5965.2013.0601]

Woods M and Katsaggelos A. 2017. A Bayesian multi-frame image super-resolution algorithm using the Gaussian information filter//Proceedings of 2017 IEEE International Conference on Acoustics, Speech and Signal Processing. New Orleans: IEEE [DOI: 10.1109/ICASSP.2017.7952380http://dx.doi.org/10.1109/ICASSP.2017.7952380]

Wu C L and Tai X C. 2010. Augmented Lagrangian method, dual methods, and split Bregman iteration for ROF, vectorial TV, and high order models. SIAM Journal on Imaging Sciences, 3(3): 300-339 [DOI: 10.1137/090767558http://dx.doi.org/10.1137/090767558]

Xu L N and He L X. 2017. GF-4 images super resolution reconstruction based on POCS. Acta Geodaetica et Cartographica Sinica, 46(8): 1026-1033

许丽娜, 何鲁晓. 2017. 基于凸集投影的高分四号卫星影像超分辨率重建. 测绘学报, 46(8): 1026-1033 [DOI: 10.11947/j.AGCS.2017.20170070http://dx.doi.org/10.11947/j.AGCS.2017.20170070]

Yang J M, Wu Y, Wei Y X, Wang B, Ru C, Ma Y Y and Zhang Y. 2019. A model for the fusion of multi-source data to generate high temporal and spatial resolution VI data. Journal of Remote Sensing, 23(5): 935-943

杨军明, 吴昱, 魏永霞, 王斌, 汝晨, 马瑛瑛, 张奕. 2019. 多源数据融合的高时空分辨率植被指数生成. 遥感学报, 23(5): 935-943 [DOI: 10.11834/jrs.20198204http://dx.doi.org/10.11834/jrs.20198204]

Yang X, Li F, Xin L, Wang C, Wang X Y and Chang X. 2018. Destriping methods for high resolution satellite multispectral remote sensing image based on GPU adaptive partitioning technology. International Society for Optics and Photonics//Proceedings of SPIE 10783, Remote Sensing for Agriculture, Ecosystems, and Hydrology XX. Berlin: SPIE [DOI: 10.1117/12.2325311http://dx.doi.org/10.1117/12.2325311]

Yang X, Li F, Xin L, Zhang N, Lu X T and Xiao H C. 2019. Finer scale mapping with super resolved GF-4 satellite images. International Society for Optics and Photonics//Proceedings of SPIE 11155, Image and Signal Processing for Remote Sensing XXV. Strasbourg: SPIE [DOI: 10.1117/12.2532674http://dx.doi.org/10.1117/12.2532674]

Yang R, Liu Z H and She W J. 2019. Simultaneous super-resolution reconstruction based on plane array staring remote sensing images. Infrared and Laser Engineering, 48(1): 0126002

杨蕊, 刘朝晖, 折文集. 2019. 遥感面阵凝视图像并行超分辨重建方法. 红外与激光工程, 48(1): 0126002 [DOI: 10.3788/IRLA201948.0126002http://dx.doi.org/10.3788/IRLA201948.0126002]

Zhang D S. 2019. Wavelet transform//Zhang D S, ed. Fundamentals of Image Data Mining. Cham: Springer: 35-44 [DOI: 10.1007/978-3-030-17989-2_3http://dx.doi.org/10.1007/978-3-030-17989-2_3]

Zhao W, Bian X F, Huang F, Wang J and Abidi M A. 2018. Fast image super-resolution algorithm based on multi-resolution dictionary learning and sparse representation. Journal of Systems Engineering and Electronics, 29(3): 471-482 [DOI: 10.21629/JSEE.2018.03.04http://dx.doi.org/10.21629/JSEE.2018.03.04]

Zhou F, Jin W, Gong F and Fu R D. 2017. Super resolution reconstruction of MODIS image based on topic learning and sparse representation. Journal of Remote Sensing, 21(2): 253-262

周峰, 金炜, 龚飞, 符冉迪. 2017. 主题学习和稀疏表示的MODIS图像超分辨率重建. 遥感学报, 21(2): 253-262 [DOI: 10.11834/jrs.20176154http://dx.doi.org/10.11834/jrs.20176154]

Zhou X M, Wang K Y and Fu J. 2017. A method of SIFT simplifying and matching algorithm improvement//Proceedings of 2016 International Conference on Industrial Informatics-Computing Technology, Intelligent Technology, Industrial Information Integration (ICIICII). Wuhan: IEEE [DOI: 10.1109/ICIICII.2016.0029http://dx.doi.org/10.1109/ICIICII.2016.0029]

Zhu X B, Tian Q J, Xu K J, Lv C G and Wang L. 2019. Radiation performance simulation and analysis of the signal-to-noise ratio for GF-4 geostationary satellite: in the case of the coastal water in Hong Kong. Journal of Remote Sensing, 23(3): 526-546

朱小波, 田庆久, 徐凯健, 吕春光, 王玲. 2019. 高分四号静止卫星辐射性能模拟与信噪比分析——以香港近海岸水体为例. 遥感学报, 23(3): 526-546 [DOI: 10.11834/jrs.20197128http://dx.doi.org/10.11834/jrs.20197128]

文章被引用时，请邮件提醒。

提交

透视地球——新一代对地观测技术

迁移深度卷积神经网络模型秋粮作物泛化识别

Grid A*：面向野外空地协同应急处置的快速路径规划

基于高光谱卫星影像的生长期互花米草指数构建